However, given the limited number of prospective randomized studies performed to date, we must also consider retrospective data in our assessment of the arthroplasty outcomes literature.

High-Flexion versus Standard Knee Designs

Kim et al.

conducted a prospective randomized study involving a bilateral total knee arthroplasty model in which a high-flexion design was compared with a standard cruciate-retaining design. Ten men and 240 women had a bilateral total knee arthroplasty in which one knee was treated with a high-flexion implant and the contralateral knee was treated with a standard-flexion design. Patients were assessed at various intervals with use of questionnaires, rating systems, and clinical and radiographic examinations.

After a minimum duration of follow-up of two years, the authors found no significant clinical differences between the groups, revealing that the high-flexion design did not have any advantage over the standard-flexion design.

Kim et al.

also performed a prospective randomized study in which fifty-four patients received a high-flexion prosthesis in one knee and a standard cruciate-retaining prosthesis in the contralateral knee.

Again, the authors found no differences between the groups in terms of knee scores, function, or range of motion at three years.

Minoda et al.

prospectively randomized 171 patients undergoing 181 cruciate-retaining total knee arthroplasties to receive standard implants or implants with a high-flexion femoral component and a polyethylene insert. The patients were followed for one year and were assessed on the basis of range of motion, knee scores, and clinical and radiographic outcomes.

There were no significant differences between groups, although the authors identified a higher average range of motion in the high-flexion group.

Seon et al.

conducted a study in which fifty knees were randomized to total knee arthroplasty with a high-flexion design and fifty were randomized to total knee arthroplasty with a standard design.

The high-flexion design was a cruciate-retaining implant with a high-flexion femoral component and high-flexion polyethylene insert similar to that used in the study by Minoda et al. The authors prospectively followed the patients for a minimum of two years in terms of range of motion, function, and knee ratings.

They found no significant differences between the groups, suggesting that the high-flexion implants alone did not provide improvement in terms of clinical outcome after total knee arthroplasty.

McCalden et al.

also compared a high-flexion design with a standard-flexion design in a prospective randomized study of 100 patients. Fifty patients received a high-flexion posterior-stabilized polyethylene insert, and fifty received a standard posterior stabilized polyethylene insert.

After an average duration of follow-up of 2.7 years, the authors found no differences in range of motion between the implant designs, again suggesting that the implant alone is not associated with clinically important improvement in postoperative flexion.

It appears that high-flexion implant designs do not provide a significant improvement of range of motion after total knee arthroplasty.

We await longer-term clinical results to determine if there are differences in implant survival due to differences in contact stress between high-flexion and standard total knee arthroplasty designs.

The literature supporting decision-making for revision total knee arthroplasty is less developed and less rigorous than that for primary total knee arthroplasty.

Furthermore, the choices and strategies shown to be optimal for total knee arthroplasty are not always optimal for revision surgery, which carries a higher risk of complication and failure.

The following studies specifically address outcomes after revision total knee arthroplasty and are compared to our own outcomes in Guadeloupe.

The use of constrained condylar implants

is probably too common for revision total knee arthroplasty as there are often challenges to attaining optimal stability through the range of motion.

Kim and Kim

retrospectively studied the results of revision total knee arthroplasty in ninety-seven patients (114 knees) who had an average age of sixty-five years at the time of the index procedure.

Aseptic loosening, polyethylene wear, and infection were the reasons for revision in the majority of cases, and constrained condylar implants were used when the posterior cruciate ligament and either the medial or lateral collateral ligament was deficient.

Most implants were cemented in the metaphyseal region only, leaving the canal free of cement for a press fit : we use the same technique.

Patients were followed for an average of 7.2 years and were assessed with use of several knee scores and osteoarthritis indices.

The authors found that the ten-year survival rate, with revision or radiographic failure as the end point, was 96%. Five knees required repeat revision, and three knees sustained an extensor mechanism rupture.

As well as our own experience woud confirm it, this report suggested that constrained condylar implants can yield very successful outcomes for ten years after revision total knee arthroplasty, although surgeons must be vigilant after surgery in consideration of a higher complication rate.

Peters et al.

retrospectively studied the four-year results of 184 consecutive revision total knee arthroplasties that were performed with a modular, cementless, press-fit, fluted stem.

The most common reasons for revision were aseptic loosening and infection : same indications in Guadeloupe.

At the time of the most recent follow-up, there had been only fifteen failures (8%), including nine recurrent infections and four new cases of deep infection.

There were no repeat revisions for aseptic loosening and no cases of radiographic loosening.

Similar to the previous study, revision total knee arthroplasty with a metaphyseal cementing technique and a diaphyseal press-fit yielded excellent early to intermediate-term results.

Long and Scuderi

reviewed the outcomes of sixteen revision total knee arthroplasties involving the use of porous tantalum cones for defects in the tibia. At a minimum of two years of follow-up, there were no cases of mechanical failure or loosening of any of the cones. Two patients required cone removal because of recurrent infection.

It appears that early-term results are quite positive for tantalum augments as they exhibit dependable osseointegration characteristics. Long-term studies are necessary to determine their survival rate.

The efficacy of antibiotic-loaded bone cement

may differ when revision total knee arthroplasty is compared with primary total knee arthroplasty.

Chiu and Lin

studied the prophylactic effect of vancomycin-bearing cement (1 g of vancomycin per 40 g of cement) used during revision total knee arthroplasty. The authors conducted a prospective cohort study of 183 first-time revision total knee arthroplasties in patients without an infection.

Ninety-three procedures were performed with low-concentration antibiotic cement, and ninety procedures were performed with standard cement.

All revisions were performed in a room without ultraviolet light, laminar flow, or body-exhaust suits, and all patients received intravenous antibiotics preoperatively and for thirty-six hours after surgery.

At an average of eighty-nine months after revision total knee arthroplasty, there had been six deep infections (7%) in the control group and no infections in the vancomycin-loaded cement group.

The difference was significant (p = 0.0130).

The authors recommended the use of low-dose antibiotic-laden cement for revision total knee arthroplasty but pointed out the need for larger studies.

• preemptive pain management,
• the intraoperative administration of medications,
• periarticular injections,
• and postoperative avoidance of parenteral narcotics.

Periarticular injections

Krenzel et al.

conducted a prospective randomized study to evaluate the efficacy of posterior capsular injections of a local anesthetic during sixty-six total knee arthroplasties.

During surgery, 20 mL of either ropivacaine or saline solution was injected into the posterior capsule in all four quadrants.

Additionally, all patients received a femoral nerve catheter with ropivacaine, a spinal block with ropivacaine, and patient-controlled intravenous fentanyl after surgery.

Outcomes were evaluated with use of functional tests, pain scales, and the assessment of narcotic usage at frequent intervals after surgery.

The authors found that patients receiving the posterior capsular injection were more likely to perform a straight leg raise at eight and ten hours after surgery (!!!?) and were less likely to rate the pain as severe at twelve hours after surgery.

After twelve hours, there was no apparent difference between the study group and the control group.

Christensen et al.

also conducted a prospective randomized study focusing on periarticular injection during total knee arthroplasty.

However, those authors compared a periarticular injection with and without the presence of methylprednisolone.

Seventy-six patients were randomized to either the study or control group.

All patients received preemptive pain medication, general anesthesia, and a femoral nerve block, and all were permitted to resume walking in the early postoperative period.

The authors found no differences in terms of pain or function after surgery; however, the patients who received the methylprednisolone injection were discharged from the hospital an average of one day sooner.

The only complications in the study occurred in patients who received the periarticular steroid injection; specifically, one patient had a deep infection, and two patients required manipulation.

The rate of complications was not significantly different between the groups; however, if periarticular corticosteroid injection does increase the risk of deep infection, the study likely did not have enough power to reveal significance.

One concern regarding the use of periarticular injections is the potential for local anesthetic toxicity in association with the simultaneous use of a reinfusion drain after total knee arthroplasty.

Parker et al.

actually measured plasma ropivacaine concentrations in twenty patients who had reinfusion from a drain after having received periarticular ropivacaine during total knee arthroplasty.

Although the periarticular injection had 150 mg of ropivacaine, patients who were reinfused with an average of only 1.9 mg of ropivacaine and plasma concentrations reached peaks of only 1.5 µg/mL.

Interestingly, the reinfused quantity of ropivacaine was quite trivial compared with the ropivacaine plasma levels resulting from the periarticular injection itself. It is important to note that the periarticular injection is administered into the tissue, not into the joint space.

I use commonly periarticular injections, as part of a pain-management protocol.

They are easy to administer, are generally safe, and appear to have a benefit in the early postoperative period.

Nerve blocks

A few groups of investigators have also conducted excellent studies on nerve blocks during the perioperative period.

Hunt et al.

randomized patients to receive either a sham injection (n = 24), a femoral nerve block (n = 33), or femoral and sciatic nerve blocks (n = 31) before total knee arthroplasty.

Postoperative pain management included patient-controlled intravenous morphine in addition to oral medications, and patients were assessed with pain scales at four-hour intervals.

The groups receiving nerve blocks had significantly lower pain scores through the morning of the second postoperative day when compared with the sham injection group.

On the day of surgery, patients receiving the femoral and sciatic nerve blocks had less pain than those receiving a femoral block only; however, there was no significant difference after the day of surgery.

Opioid use was also significantly less for the groups receiving the nerve blocks, with the combined sciatic and femoral nerve block group having the least opioid usage through the afternoon of the first postoperative day.

Shum et al.

studied the use of continuous femoral nerve blocks for analgesia after total knee arthroplasty.

Sixty patients undergoing a unilateral total knee arthroplasty were randomized to receive a high-dose continuous femoral nerve block (0.15% ropivacaine), a low-dose femoral nerve block (0.2% ropivacaine), or no femoral nerve block.

All patients also received patient-controlled intravenous morphine after surgery. Both nerve block groups received continuous ropivacaine administered at a rate of 10 mL/hr for the first twenty-four hours and 5 mL/hr for the next twenty-four hours. The authors assessed pain scores and opioid use for seventy-two hours after total knee arthroplasty and also evaluated clinical scores and range of motion two years after surgery.

Patients who had a continuous nerve block had significantly less postoperative pain for six hours after surgery; however, there was no effect beyond this time period.

Additionally, patients with a continuous block had lower cumulative morphine usage for seventy-two hours after surgery. There were no differences between the high and low-dose ropivacaine groups. Furthermore, there were no differences between any of the groups in terms of clinical scores at two years.

On the basis of these two studies addressing the use of nerve blocks after total knee arthroplasty, it appears that there is an early postoperative benefit that decreases pain and opioid use for up to twenty-four hours after surgery. However, there do not appear to be clinically relevant differences beyond this time period.

Blood-salvage techniques and the timing of tourniquet release…

Another perioperative decision involves the use of blood-salvage techniques and the timing of tourniquet release during total knee arthroplasty.

Steffin et al. (2009)

conducted a prospective randomized study comparing tourniquet release before or after wound closure and studied the effects on hematocrit drop, drainage output, and operative time.

Thirty-seven patients were randomized by means of a coin flip and were followed through the perioperative period.

The authors found no difference between the groups in terms of any of the parameters studied.

Outcomes and benefits remain mixed.

Karpman and Smith (2009)
performed a randomized prospective study of total knee arthroplasties performed through a medial parapatellar approach (n = 19), a mini-midvastus approach (n = 20), or a quadriceps-sparing approach (n = 20).

The same surgeon performed all procedures, and identical perioperative protocols were used for all patients.

Although the quadriceps-sparing approach was associated with improvements in terms of pain, function, and satisfaction, these improvements were not consistent or persistent.

The quadriceps-sparing approach was associated with an additional fifteen minutes of operative time on average and was the approach used for the only two postoperative infections in the study.

Lin et al. (2009)

also conducted a prospective randomized study comparing the quadriceps-sparing approach with the mini-incision medial parapatellar approach.

Sixty patients (eighty total knee arthroplasties) were included in the study, and the authors could find no significant differences in terms of postoperative pain, peak muscle torque, or functional outcomes.

The quadriceps-sparing approach was associated with the only alignment outliers (four implants) and had a surgical duration that was about twenty minutes longer than that for the medial parapatellar approach.

Given the increased surgical time, the potential for malalignment and wound complications, and the limited early benefit, the quadriceps-sparing approach is not a recommended technique for many surgeons.

Other authors have also studied the early outcome after specific approaches to total knee arthroplasty.

Dalury et al. (2009)

specifically addressed the effect of anterior tibial translation and patellar eversion on early recovery after total knee arthroplasty.

In a prospective randomized study involving forty patients, the authors tested the hypothesis that avoidance of tibial dislocation and patellar eversion leads to improved recovery.

Patients were followed with regard to Knee Society scores and testing for pain, strength, and motion.

The authors found no clinical differences between the study groups at six weeks and concluded that patellar eversion and tibial translation have no adverse effect on patients having a total knee arthroplasty.

Niki et al. (2009)

analyzed the serum levels of several biomarkers for tissue damage to compare various surgical approaches for total knee arthroplasty.

A consecutive series of 147 patients was studied; the procedures included several minimally invasive approaches and standard approaches to total knee arthroplasty.

Serum levels of creatinine phosphokinase, myoglobin, aldolase, lactate dehydrogenase, glutamic-oxaloacetic transaminase, and creatinine were tested in all patients preoperatively and at five intervals up to two weeks postoperatively.

The minimally invasive approaches were not associated with lower biomarker levels.

In addition to concerns about the limited benefit of minimally invasive surgery, there is concern that use of the minimally invasive approaches could lead to increased early failure after total knee arthroplasty, especially when new adopters are not trained.

Barrack et al. (2009)

conducted a very interesting study in which they analyzed their experience with 237 first-time revision total knee arthroplasties over a three-year period.

Although only 18% of these revisions were performed following a minimally invasive index arthroplasty, the average time to revision in this group of patients was only fifteen months.

This represented a very rapid time to failure in comparison with the authors’ overall experience with revision after standard total knee arthroplasty (average time to revision, eighty months).

Although there have been many reports from experienced centers demonstrating the safety of minimally invasive techniques, it is not clear that the results are equal among all surgeons.

The authors of two recent studies warned about the complications that may arise in association with the use of a minimally invasive approach to unicompartmental knee replacement.

Hamilton et al.

focused on complications and two-year survival in a retrospective study of 445 consecutive minimally invasive fixed-bearing unicondylar knee replacements. The authors identified an overall revision rate of 5.8% at an average of 3.25 years after surgery. The most common reasons for revision and reoperation were retained cement (4.5%), aseptic loosening (2.2%), and tibial plateau fracture or collapse (0.9%). Interestingly, the authors found that the complications could not be totally attributed to a learning curve as the revision and reoperation rates did not decrease significantly through the study period. The authors emphasized the importance of appropriate visualization and attention to excess cement removal before closure. However, it was not clear from the study whether the minimally invasive technique was responsible for the complications.

Song et al.

also reported on their early experience with unicompartmental knee arthroplasty with use of a mobile-bearing implant. They reviewed their first 100 cases of medial unicompartmental knee arthroplasty after a minimum of two years of follow-up. Complications necessitating reoperation included two medial plateau fractures, four meniscal bearing dislocations, and one case of femoral loosening.

Similar to the paper by Hamilton et al., the authors emphasized the careful attention that is required to perform a successful unicompartmental knee replacement.

Again, it is not clear whether the complications were related to minimally invasive surgery.

Unicompartmental knee replacement is considered less traumatic and likely results in better function than total knee replacement.

However, the surgery requires meticulous technique in order to avoid complications, regardless of whether an extensile or a minimally invasive approach is used.

The two main technologies currently available to “guide” the surgical decision-making during a total knee arthroplasty are

1. computer navigation
2. and case-specific cutting jigs.

Both technologies aim to eradicate the natural variation in surgeon technique.

One of the main recent areas of focus has been demonstration of the importance of the mechanical axis as a predictor of function and longevity after total knee arthroplasty.

Two studies demonstrated advantages in terms of functional outcome and implant survival following total knee arthroplasties with a neutral mechanical axis.

I reach in Guadeloupe,  the same conclusions as Fang’s :  Fang et al., in a report on 6070 knees, showed that outliers in terms of overall mechanical alignment were associated with a significantly higher failure rate.

The authors observed an average failure rate of 0.84% at a mean of 6.6 years after total knee arthroplasty.

Knees that were implanted in 3° to 7° of coronal anatomic alignment had the lowest failure rate (0.5%), whereas knees outside of this range had a risk of failure that was more than three times greater.

This study was larger than previous studies with conflicting conclusions.

Our own  - same –  conclusions are drawn from a 700 hundreds cases serie.

However, I am not so lucky as Choong et al., who performed a randomized prospective study comparing

• computer-assisted surgery (sixty knees)
• with conventional surgery (fifty-five knees).

As we did, they also demonstrated a functional advantage when implants that were implanted within 3° of a neutral mechanical axis were compared with outliers.

Furthermore, computer-assisted total knee arthroplasty was associated with a higher number of knees within 3° of neutral mechanical alignment (88% versus 61%).

As we did, Longstaff et al. also demonstrated that the functional outcome after total knee arthroplasty was significantly better for patients who had neutral mechanical alignment.

1/ Kim et al.

prospectively compared computer navigation with conventional total knee arthroplasty in a bilateral total knee arthroplasty study model involving 160 patients (320 knees).

After a mean of 3.4 years follow-up, the authors found no differences in terms of knee scores or function.

They also found no differences in terms of implant alignment between techniques with use of radiographs and computed tomography.

However, the total knee arthroplasties that were performed with computer navigation required an average of almost twenty minutes of additional surgical time.

2/ Weng et al.

also prospectively compared computer navigation with conventional total knee arthroplasty with use of a bilateral knee study model involving sixty patients (120 knees).

The authors found that computer navigation was associated with a much higher rate of knees within 3° of a neutral mechanical axis (87% versus 50%) but also increased the surgical duration by an average of about twenty minutes.

3/ In a third such study, Seon et al.

prospectively compared function and implant alignment in a study involving thirty-one knees that were treated with navigation-assisted unicondylar knee arthroplasty and thirty-three knees that were treated with conventional unicondylar arthroplasty.

After a minimum duration of follow-up of two years, there were no significant differences between the groups in terms of knee scores or function; however, there were significantly more outliers in alignment (>3° from neutral) in the conventional unicondylar arthroplasty group.

The authors also reported two stress fractures due to the creation of pin tracks in the navigation group.

While most studies have demonstrated that navigation successfully reduces the number of outliers of implant alignment,

navigation requires more operative time,

and additional study will be necessary to determine whether the improvement in implant alignment impacts long-term function or survival to offset the increased surgical time and the potential for fracture.

Considering the variable efficacy and risks of bleeding, infection, and wound complications associated with anticoagulation after total knee arthroplasty, the American Academy of Orthopaedic Surgeons (AAOS) has offered me clinical practice guidelines on the prevention of symptomatic pulmonary embolism : this took place for me, as I attend the AAOS Annual Meeting San Francisco in Feb 2008.

The practice guidelines emphasize the following points:

(1) arthroplasty surgery is different now than it was fifteen years ago and older studies should not be used to create guidelines,

(2) the available literature since 1996 shows no significant differences in the rate of symptomatic pulmonary embolism with the use of low-molecular-weight heparin, warfarin, or mechanical devices and aspirin alone,

(3) the body of literature is underpowered and fails to provide considerable evidence because of the rarity of symptomatic or fatal pulmonary embolism after arthroplasty, and

(4) the risk of major bleeding should be a consideration when selecting the method of thromboprophylaxis after arthroplasty.

I conducted a pooled analysis of randomized controlled trials focusing on venous thromboembolism after major orthopaedic surgery.

I included all randomized controlled trials that were cited by the American College of Chest Physicians (ACCP) guidelines and also added two trials that were excluded by the ACCP because they did not provide venographic data.

Pooled data analysis was utilized in the review, and the rates of

• symptomatic deep-vein thrombosis,
• symptomatic pulmonary embolism,
• fatal pulmonary embolism,
• major operative-site bleeding,
• and major non-operative-site bleeding

were specifically calculated.

I found, on the basis of the pooled data, that aspirin is effective for decreasing the number of venous thromboembolism events after major orthopaedic surgery.

This conclusion resulted from the Pulmonary Embolism Prevention trial that was excluded from the ACCP analysis because of the lack of venographic data.

I also found that while

• warfarin,
• low-molecular-weight heparins,
• and pentasaccharides

increase the risk of operative-site bleeding events across pooled data, there was no reduction in

• symptomatic deep-vein thrombosis,
• pulmonary embolism,
• or fatal pulmonary embolism.

The author also identified the methodological flaws that undermine the ACCP guidelines:

(1) the exclusion of any randomized studies that did not utilize venographic outcome assessments,

(2) the failure to require relevant clinical outcomes such as symptomatic deep-vein thrombosis, pulmonary embolism, fatal pulmonary embolism, and major bleeding complications for inclusion in their analysis,

(3) the failure to quantitatively analyze the incidence of these relevant clinical outcomes from randomized trials,

and (4) potential conflicts of interest with pharmaceutical companies for six of seven members of the drafting committee.

Furthermore, the ACCP guidelines adhere to strict inclusion criteria requiring outcomes assessment with venography or duplex ultrasound.

This criterion is based on their claim that asymptomatic deep-vein thromboses lead to post-thrombotic syndrome and cause morbidity after major orthopaedic surgery.

However, without any Level-1 prognostic outcome studies demonstrating post-thrombotic syndrome after asymptomatic deep-vein thrombosis following arthroplasty or hip fracture surgery, the ACCP guidelines are reduced to expert opinion.

Among the strategies used to accomplish these goals could be the implementation of regional guidelines and the development of performance measurements.

Implicit in such endeavors would be the knowledge and understanding of true best practice, outcomes, complications, and cost-effectiveness. Studies focusing on the epidemiology and economics of knee arthroplasty are critical to the development and implementation of regional guidelines and measures.

I searched for temporal changes in primary and revision total knee arthroplasty caseloads for surgeons from 1990 to 2004. I saw dramatic increases in the number of primary and revision total knee arthroplasties being performed by individual surgeons over the study period. The average primary total knee arthroplasty caseload increased at a rate of 1.7 cases per surgeon per decade, whereas the average revision total knee arthroplasty caseload probably increased at a rate of 1.4 cases per surgeon per decade. Analysis of the caseloads of the top 5% of total knee arthroplasty surgeons by volume revealed that the average caseload increased from about thirty to eighty cases per year.

Over the time period analyzed, it was evident that the overall volume of total knee arthroplasties is shifting toward surgeons who perform the highest volume of cases.

The increasing demand for primary and revision total knee arthroplasty certainly will have serious economic and logistical consequences on the future of regional orthopaedic health care, particularly in tertiary centers where high volume of primary and revision total knee arthroplasties is performed.

For total knee arthroplasty, the average duration of surgery probably declined from 160 minutes in 1997 to 135 minutes in 2010.

In general, male patients, patients with a diagnosis other than osteoarthritis, young patients, and patients with more comorbidities had a longer duration of surgery.

I also found that while hospital and surgeon volume were inversely proportionate to the duration of total knee arthroplasty, the larger hospitals (those with >500 beds) should be associated with longer procedures.

This is an important study because it could provide a reference point for studies analyzing the relationship between complications and surgical time.

Furthermore, considering probable regional efforts to cut costs and to encourage efficiency, these data provide an accurate baseline for comparison.

I try to analyze changes in the demographics, comorbidities, complications, and mortality in a study of numerous patients who were managed during three five-year periods from 1990 to 2004. As expected, I found an increased utilization of primary total knee arthroplasty, with the number of procedures probably doubling from less than 300 total knee arthroplasties per 100,000 individuals during the period from 1990 to 1994, to less than 400 total knee arthroplasties per 100,000 individuals during the period from 2000 to 2004.

Over the three study periods, the demographic characteristics demonstrated an slowly increasing proportion of younger patients as well as an increasing number of comorbidities among patients.

Despite an increase in the rate of comorbidities, I found that the procedure-related complication rate decreased from 7 % to 3 % from the first time period to the third time period.

Although the mortality rate could not declined from 0.00% during the period from 1990 to 1994 to 0.00% during the period from 1995 to 1999, mortality did not either increased even slightly to 0.00% during the period from 2000 to 2004, and from 1004 to 2010.

Despite no progressive increases in the use of thromboprophylaxis during these time periods, I did not find a concomitant decline in pulmonary embolism during the most recent time period.

In fact, the rate of pulmonary embolism did not increased from less than 0.20% in the period from 1995 to 1999 to less than 0.20% in the periods from 2000 to 2010.

Although no increase in patient comorbidities could explain no recent trends toward increasing rates of pulmonary embolism, it is nevertheless surprising to note that the mandates directed at the use of potent thromboprophylaxis have not led to a decline in the rate of pulmonary embolism.

Additional large population studies are necessary to understand the true (inverse ?)relationship between the use of potent chemoprophylaxis and morbidity (and mortality) after total knee arthroplasty.

In one case, a 32 year-old female presented to her orthopedic surgeon with knee pain. The surgeon noted that as the patient extended her knee into full extension, there was a “pop” in the posterolateral knee. She was referred for an MRI, to evaluate for internal derangement.

A sagittal proton-density weighted image reveals abnormal signal in the anterior horn of the lateral meniscus (red arrow), but is otherwise unremarkable:

Asked was the patient if she could do anything that would elicit the “pop”. She obliged, and squatted downwards, and the “pop” occurred. Placed back in the magnet, was repeated the sagittal:

After  provocative maneuver,  seen is a striking peripheral tear of the posterior horn of the lateral meniscus displaced anteriorly, leaving only fluid where the meniscus should be (green arrow).

On axial image is better revealed the avulsed lateral meniscus (blue arrows),  displaced anteriorly:

Compare this to the position of the meniscus (yellow arrow) on the initial (before the provocative maneuver) axial scan:

Intermittent meniscal dislocation has been discussed in the literature (Lyle et al.Br J Radiol. 2009, 82:374-9).

They described three patients with a strong history of intermittent knee locking, who had negative initial MR scans. The patients were able to reproduce locking of their knee voluntarily, as in our case. Further MR imaging of the knee in the “locked” position demonstrated meniscal dislocation in all three patients. All three were confirmed arthroscopically to have deficiency of the corresponding menisco-capsular ligaments (as was our patient).

When there is a strong clinical history of knee locking, all the structure of the knee must be carefully inspected on MRI, particularly the menisci, anterior cruciate ligament, and the hyaline cartilage. When no abnormality can be detected, it is a good idea to scrutinize the peripheral attachments of the meniscus. The meniscocapsular junction is a difficult area to analyze, with abnormalities easily missed (and overcalled as well). With higher resolution imaging now becoming increasingly common, it has become easier to detect abnormalities in this area with greater confidence.

In the exceptional case, provocative imaging can be performed, and may help demonstrate an intermittently dislocating mensicus.

Reference

Reference