bismillah

Copyright 1989 by The Journal of Bone and Join: Surgery. Incorporated

VOL. 71-A, NO. 6. JULY 1989 811

Non-Operative Treatment of Meniscal Tears*BY CARL B. WEISS, M.D.t, ROCHESTER, MAGNUS LUNDBERG, M.D.t, PER HAMBERG, M.D., LINKOPING,

KENNETH E. DEHAVEN, M.D.t, ROCHESTER, NEW YORK, AND JAN GILLQUIST, M.D.1, LINKOPING, SWEDEN

From the Department of Orthopaedics, University of Rochester Medical Center, Rochester.and the Department of Orthopaedic Surgery, University Hospital. Linkoping

ABSTRACT: In a retrospective review of the resultsof 3,612 arthroscopic procedures that were performedfor the treatment of an acute or a chronic meniscal le-sion, with or without an associated ligamentous lesion,we identified eighty meniscal tears (in seventy-five pa-tients) that had been assumed to be stable. Seventy werevertical longitudinal tears and ten were vertical radialtears. The seventy longitudinal tears included fifty-twolateral and eighteen medial meniscal lesions. All of theradial tears were in the lateral meniscus. Of the seventy-five patients, fifty-two had been followed for two to tenyears. At the time of follow-up, only six of these fifty-two patients had needed additional intervention becauseof symptoms that were related to the meniscal tear. Fourof them had the intervention after a sports-related trau-matic extension of a stable tear, and two, because per-sistent symptoms were caused by the original meniscallesion.

A repeat arthroscopy was performed on thirty-twopatients (twenty-six of whom had a longitudinal tear andsix of whom had a radial tear), at an average of twenty-six months after the original arthroscopy. Seventeen ofthe twenty-six longitudinal tears had completely healed.Five of the six radial tears had no evidence of healingand one had extended. Neither ligamentous laxity nor ameniscal tear that was chronic at the time when it wasdiscovered appeared to preclude healing of the stablelongitudinal tears. No localized degenerative changes inthe adjacent articular cartilage were found in associationwith any of the stable vertical longitudinal or radialmeniscal lesions.

Excluding the six patients who had had additionalsurgical treatment, none of the fifty-two patients whofilled out a questionnaire reported that they had symp-toms of a meniscal lesion, and none of the forty-twopatients who were re-examined two years or more afterthe operation had signs of a meniscal lesion.

Stable vertical longitudinal tears, which tend to oc-cur in the peripheral vascular portions of the menisci,

* No benefits in any form have been received or will be received froma commercial party related directly or indirectly to the subjectofthis article.No funds were received in support of this study.

t Department of Orthopaedics, University of Rochester Medical Cen-ter, 601 Elmwood Avenue, P.O. Box 665, Rochester, New York 14642.

Department of Orthopaedic Surgery, University Hospital, S-58 186Linkoping, Sweden.

have great potential for healing. The tear should be leftalone unless it is the only abnormality that is found andit is causing symptoms that warrant treatment. Stableradial tears, which tend to occur in the avascular innerone-third of the meniscus, have little potential for heal-ing. Whether it is best to leave these lesions alone or tofashion an intact rim by contouring the meniscus wasnot established by this study.

Recent long-term follow-up investigations have shownthat degenerative changes develop more frequently in kneesthat have had a total meniscectomy than they do in thecontralateral knees in which the menisci are found to beintactu 1.16.18.19,30 Radiographic and clinical studies havealso shown that degenerative changes develop four to seventimes more often in knees that have been treated by partialmeniscectomy than they do in the contralateral, unaffectedknees7. Finally, degenerative changes develop even morefrequently in knees that have had a meniscectomy and haveassociated ligamentous instability92 #{176}.

Biomechanical studies have shown that the meniscihave important load-transmitting and energy-absorbingfunctions2192025272933. They carry 30 to 70 per cent of thetotal load that is transmitted across the knee joint. They doso even when they are torn, provided that the peripheralcircumferential fibers are intact21272833. The menisci alsoprovide passive stability, functioning as secondary soft-tis-sue restraints that prevent anterior displacement of the tibiaon the femur3.

The blood supply of the menisci has been studied ex-tensively. The potential for healing of tears in the vascularouter 15 to 25 per cent of the width of the meniscus wasdemonstrated by Scapinelli26 and by Arnoczky et al.2. Thevalidity of these authors findings was supported by theexcellent clinical results after repair of acute and chronicperipheral meniscal tears46804526.

This understanding of the function and importance ofthe menisci has prompted a selective approach in the treat-ment of meniscal tears. It has been the general practice topreserve as much meniscal tissue as possible, in the hopethat the long-term results after such treatment will be betterthan those after total meniscectomy924.

The goal of this study was to identify a subset of din-ically stable meniscal tears that can be left alone. We ana-lyzed stable tears that had been identified arthroscopically,

FIG. 1 FIG. 2Fig. 1 : A stable partial-thickness vertical longitudinal tear of the lateral meniscus, demonstrated using a probe.Fig. 2: A stable radial tear of the inner one-third of the lateral meniscus. demonstrated using a probe.

812 C. B. WEISS ET AL.

THE JOURNAL OF BONE AND JOINT SURGERY

on the basis of arbitrary criteria that were developed throughthe clinical experience of two of us (K. E. DeH. and J. G.).We categorized stable tears on the basis of length, depth,type, and location, and we attempted to establish the im-portance, if any, of each of these parameters in the iden-tification of lesions for which no surgical intervention isneeded.

Materials and Methods

Between 1972 and 1985, 3,612 arthroscopic proce-dures were performed on patients who had signs and symp-toms of an acute or chronic meniscal tear, with or withoutan associated ligamentous tear; 1 , 177 of the procedures weredone at the University of Rochester Medical Center and2,435, at the University Hospital, Linkoping, Sweden. Oneof us (K. E. DeH.) performed all of the procedures inRochester, and one of us (J. G.), in Linkoping. At bothmedical centers, the patients were healthy and athletic. Theaverage age of all of the patients was 25.4 years (range,sixteen to forty-seven years). All of the patients were seenbecause they had acute or chronic symptoms in the knee.

Although different arthroscopic portals may have beenused, for each patient a thorough arthroscopic examinationwas performed by two orthopaedists who had comparableexpertise in arthroscopic surgery. The two populations ofpatients were assumed to be a roughly homogeneous groupand were treated by two surgeons who both believed thatnot all meniscal tears are symptomatic or cause secondaryarticular changes and that some meniscal tears can be leftalone.

As a result of the arthroscopic examinations, 1,287patients who had 1,316 meniscal tears were identified. Inthe remainder of the patients, either isolated ligamentouschondral lesions or symptomatic plicas were identified. Ofthese 1 ,316 tears, eighty in seventy-five patients were class-ified as stable in accordance with the following criteria.

A vertical longitudinal tear involving the body of themeniscus was classified as stable when the portion of themeniscus that was central to the tear could not be displacedmore than three millimeters from the intact peripheral rim(Fig. 1). The amount ofdisplacement was determined whileeach tear was being thoroughly probed, on both the tibialand the femoral surfaces, using a five-millimeter nerve-hookthat had a three-millimeter-long tip below the curve of thehook. Each tear was probed carefully to determine whetherit was full thickness or partial thickness. The length of eachtear was determined using the tip of the nerve-hook as aruler, and the surface or surfaces on which the tear hadoccurred were also documented. All partial-thickness tearswere classified as stable.

Radial tears that involved less than the inner one-thirdof the width of the meniscus were considered to be stablein Linkoping but not in Rochester (Fig. 2). In LinkOping,ten such tears were left alone. In Rochester, an undeterminednumber of such tears were treated by arthroscopic d#{233}bride-ment, with removal of the torn portion of the inner rim andcontouring of the meniscus so that its inner margin wassmooth and not indented. Although only a few radial tearswere left alone in this study, and the findings are thereforeof uncertain significance, the follow-up data on these tearsare interesting and are included in this report.

Slightly more than 6 per cent (eighty) of the 1,316meniscal tears were stable, as defined by the previouslydescribed criteria. If the ten radial tears that were followedin Linkoping are excluded, the frequency of stable verticallongitudinal meniscal tears is slightly more than S per cent(seventy of 1 ,306). The seventy vertical longitudinal me-niscal tears in sixty-five patients were in the lateral meniscusin forty-six, in the medial meniscus in fourteen, in both themedial and the lateral meniscus in four, and in both lateralmenisci in one.

To determine the locations of the tears within the me-

Tibial Surface 10 (56%) Tibial Surlace 3 (7%)

Femoral Surface 2 (11%) Femoral Surface 31 (70%)

Tibial & Femoral Surface 0

Full Thickness 6 (33%)

Tibial & Femoral Surface 2 (5%)

Full Thickness 8 (18%)FIG. 3

The locations of the stable longitudinal meniscal tears.

NON-OPERATIVE TREATMENT OF MENISCAL TEARS 813

VOL. 71-A, NO. 6. JULY 1989

MEDIAL MENISCUS(1 8 Vertical longitudinal tears)

LATERAL MENISCUS(44 Vertical longitudinal tears)

isci, the torn surface or surfaces of the menisci were iden-tified (Fig. 3) and the body of each meniscus was dividedinto three equal areas and five equal zones, which were usedto locate the tears more exactly (Fig. 4).

The seventy vertical longitudinal tears all were leftalone. No attempt was made to debride or freshen the tornmeniscal surfaces. The tears were manipulated only duringexamination with the nerve-hook. The postoperative man-agement varied for the patients who had a vertical longi-tudinal tear. For the nine patients who had an isolatedmeniscal lesion (no associated ligamentous injuries), theknee was not immobilized. The patients walked usingcrutches, progressively increasing weight-bearing until theydiscontinued the crutches, usually within a week.

The remaining fifty-six patients had an associated acute

or chronic ligamentous lesion, and they were treated withvarying amounts of immobilization. Of the thirty patientswho had an acute injury, in twenty-four the knee was im-mobilized after repair or reconstruction of an acute ruptureof the anterior cruciate ligament. When either an anteriorcruciate ligament was reconstructed, with or without repairof other ligaments, or a torn meniscus was repaired, theknee was immobilized for six weeks. After this, a protocolfor rehabilitation, which included range-of-motion exercisesand preferential strengthening of the hamstrings, was insti-tuted. Active quadriceps exercises, in which the knee wasflexed 45 degrees and then fully extended, were not doneuntil the twelfth postoperative week.

For the remaining six patients who had an acute hg-amentous injury, the knee was immobilized for two to three

Posterior



ZONES AREAS

-Outer 1/3

sOQter 1/3

1/3

Fi. 4

Central 1/3

1/3

Each zone represents one-fifth of the length of the meniscus and each area represents one-third of the width of the meniscus.

weeks. This was followed by a rehabilitation program sim-ilar to the one for the other twenty-four patients, except forthe inclusion of quadriceps exercises against resistance, in-volving moving the knee through a full range of motion.

Of the twenty-six patients who had a chronic meniscaland ligamentous lesion, reconstruction of the anterior cru-ciate ligament was done in seventeen. Those seventeen weretherefore managed with a protocol for rehabilitation thatwas similar to that for the patients who had a repair of anacute anterior-cruciate lesion. For the nine patients who hadchronic anterior-cruciate insufficiency but did not have re-construction, immobilization was not used, as if the me-niscal lesion were isolated. No stable meniscal tear wasassociated with a rupture of the posterior cruciate ligamentor with an isolated sprain of the medial or lateral collateralligament. Of the six patients who had an acute radial tear,four were immobilized, while two had no immobilizationpostoperatively.

The average age of the seventy-five patients who hadeighty stable meniscal tears (seventy longitudinal and tenradial) was 25.4 years (range, sixteen to forty-seven years).Fifty-six patients were male and nineteen, female. Sixty-five patients had a sports-related injury, and in ten the injurywas caused by another type of trauma. Of the eighty men-iscal lesions, forty-five were diagnosed during an arthros-copic examination for an acute injury and thirty-five, duringarthroscopic evaluation for chronic symptoms. Eighty-fourper cent of the forty-five acute meniscal tears were asso-ciated with an acute tear of the anterior cruciate. Seventy-seven per cent of the thirty-five chronic tears were associatedwith a chronic rupture of the anterior cruciate.

For all seventy-five patients, the arthroscopic findingswere reviewed by re-examination ofthe videotapes that weremade during the operation. Written descriptions of the op-erative findings and postoperative drawings ofeach meniscaltear were also reviewed. From these sources, data werecollected on the type, location, length, and depth of eachtear.

All data were reviewed by two of us (C. B. W. at theUniversity of Rochester and P. H . at the University Hospital,Linkoping) who were not directly involved in the surgicaltreatment or postoperative management but who did performthe follow-up examinations. Of the seventy-five patientswho were originally identified, four were lost to follow-up.Eight patients who had a vertical longitudinal tear of thelateral meniscus were excluded because the information con-cerning the location, length, and depth of the tear was in-complete. Five other patients were excluded because theyhad not been followed for at least two years. At the timeof the most recent follow-up, none of those seventeen pa-tients had had any persistent or recurrent symptoms relatedto the meniscal tear. However, by this time, six additionalpatients had had further surgical treatment of a meniscaltear that originally had been thought to be stable (as willbe described).

Of the fifty-two patients who had been followed for aminimum of two years (range, two to ten years; average,4.3 years) and had had no further surgical treatment, twenty-four were re-examined at the University of Rochester andtwenty-eight, at the University Hospital, Linkoping. A ques-tionnaire was used to determine the function of the kneeand the patients level of activity. The answers to the ques-


Points

530

5.,

0

1510

6

0

25201510

50

25

20

10

5

0

10630

10630

54.,

0

* Each of the five zones represents one-fifth of the length of the

meniscus. See text and Fig. 4.t Each of the three areas represents one-third of the width of the

meniscus. See text and Fig. 4.1: All radial tears.

VOL. 71-A, NO. 6. JULY 1989

TABLE I

THE LYSH0LM-II SCORE2332

LimpNoneSlight or periodicSevere and constant

SupportNoneStick or crutch neededWeight-bearing impossible

LockingNoneCatching sensation, but no lockingLocking occasionallyLocking frequentlyLocked joint at examination

InstabilityNeverRarely during athletic activitiesFrequently during athletic activitiesOccasionally during daily activitiesOften during daily activitiesEvery step

PainNoneInconstant and slight during

strenuous activitiesMarked during or after

walking >2 kmMarked during or after

walking

TABLE III

RESULTS AT THE TIME OF REPEAT ARTHROSCOPY

Thick- StabilityType of Length ness of the Immobili- Length of

Case Age Meniscus Tear Zone* Areat of Tear of Tear Knees zation Follow-up Result(Yrs.) (itt,::) (Mos.)

Acute veIl. Post. Outer 5-10 Full Stable No 52 Healedlongit.

Acute vert. Post. Central 5-10 Full Unstable No 16 Healedlongit.

Acute veIl. Post. Outer 5-10 Full Unstable Yes 7 Healedlongit. junci.

Acute vert. Post. Outer 5-10 Full Unstable Yes 9 Healedlongit.

Chronic vert. Post. Outer 5-10 Full Stable No 10 Healedlongit.

Chronic yen. Post. Outer 10 Full Unstable Yes 25 Healedlongit. junct.

Chronic vert. Post. Outer 5-10 Full Unstable Yes 25 Healedlongit.

Chronic vert. Post. Outer 5-10 Full Unstable Yes 12 Healedlongit.

Acute vert. Post. Outer 5-10 Partial Stable Yes 50 Healedlongit. junct.

Acute vert. Post. Central 5-10 Partial Stable Yes 6 Healedlongit.

Acute vert. Post. Outer 5-10 Partial Unstable Yes 26 Healedlongit.

Chronic veIl. Post. Outer 10 Partial Unstable Yes 35 Healedlongit.

Acute vert. Post. Outer 5-10 Partial Stable Yes 27 Healedlongit.

Acute vert. Post. Outer 5-10 Full Unstable Yes 30 Unchangedlongit. junct.

Acute veil. Ant. Central 5-10 Full Unstable Yes 100 Unchangedlongit. junct.

Acute yen. Ant. Central > 10 Partial Stable No 37 Unchangedlongit. junct.

Acute vert. Post. Outer 5-10 Partial Unstable Yes 79 Unchangedlongit.

Chronic yen. Post. Outer 5-10 Partial Stable No 21 Unchangedlongit. junct.

Acute radial Mid. Inner


TABLE IV-A

CORRELATION OF THE LOCATION OF THE TEARS AND

HEALING AT THE TIME OF REPEAT ARTHROSCOPY

Zone*

Lateral Meniscus (N = 24) Medial Meniscus ( N = 8)Outer1 Centralt Innert Outert Centralt Innert

PosteriorHealed 8 2 0 2 0 0Unhealed 1 0 0 2t 0 0Retear or progression 1 2 0 0 0 0

Total 10 4 0 4 0 0

Posterior junctionalHealed I 0 0 2 0 0Unhealed 0 0 0 2 0 0Retear or progression 0 0 0 0 0 0

Total I 0 0 4 0 0

MiddleHealed li 0 0 0 0 0Unhealed 0 0 5 0 0 0Retear or progression 0 0 1 0 0 0

Total 1 0 6 0 0 0

Anterior junctionalHealed 0 0 0 0 0 0Unhealed 0 2 0 0 0 0Retear or progression 0 0 0 0 0 0

Total 0 2 0 0 0 0

AnteriorHealed 0 0 0 0 0 0Unhealed 0 0 0 0 0 0Retear or progression 0 0 0 0 0 0

Total 0 0 0 0 0 0

* Each of the five zones represents one-fifth of the length of the meniscus. See

text and Fig. 4.t Each of the three areas represents one-third of the width of the meniscus. See

text and Fig. 4.t These two tears were repaired three months after the diagnosis. at which time

one of them was partially healed. A vertical longitudinal tear.qT All radial tears.

area of the middle zone had healed. However, the two tearsin the central area of the anterior junctional zone showedno signs of healing at the time of the repeat arthroscopy(Tables III and IV-A).

Three longitudinal tears in the posterior zone of thelateral meniscus (two in the central and one in the outerarea) that had been previously designated as stable becamesymptomatic after a subsequent acute sports-related in-jury. Repeat arthroscopy, eleven, sixteen, and twenty-fourmonths after the initial arthroscopy, showed that each ofthe three tears had extended and become unstable. In eachof these knees, the torn portion of the meniscus was de-brided. In two of them, the anterior cruciate ligament wasreconstructed initially, and at the time of follow-up theLachman-test rating was 1 + or more. The third knee hadchronic anterior-cruciate insufficiency, and reconstructionwas performed at the time of the repeat arthroscopy.

Of the six radial tears of the lateral meniscus that werere-examined (Table III), five had no evidence of healingand one (Case 3 1) had extended to the peripheral rim aftera sports-related injury. Any healing of this radial tear thatmight have occurred could not be evaluated because of thesecond injury. This tear was treated by d#{233}bridement.

At the time of the second-look arthroscopy, five of thenine partial-thickness and seven of the nine full-thicknessvertical longitudinal tears had healed (Table IV-B). Also atthis time, the lengths of the vertical longitudinal tears werecorrelated with healing (Table IV-C), but there were toofew tears in each category to permit definite conclusions.

TABLE IV-B

CORRELATION OF THE THICKNESS OF THE TEARS AND HEALING AT THE TIME OF REPEAT ARTHROSCOPY

Lateral Meniscus (N = 24) Medial Meniscus (N = 8)Outert Centralt Innert Outert Centralt Innert

Zone* Partial Full Partial Full Partial Full Partial Full Partial Full Partial Full

PosteriorHealed 3 5 I 1 0 0 2 0 0 0 0 0Unhealed I 0 0 0 0 0 2 0 0 0 0 0Retear or progression 1 0 1 1 0 0 0 0 0 0 0 0

Total 5 5 2 2 0 0 4 0 0 0 0 0

Posterior junctionalHealed 0 1 0 0 0 0 1 1 0 0 0 0Unhealed 0 0 0 0 0 0 1 1 0 0 0 0Retear or progression 0 0 0 0 0 0 0 0 0 0 0 0

Total 0 1 0 0 0 0 2 2 0 0 0 0

MiddleHealed l 0 0 0 0 0 0 0 0 0 0 0Unhealed 0 0 0 0 0 51 0 0 0 0 0 0Retear or progression 0 0 0 0 0 1 0 0 0 0 0 0

Total 1 0 0 0 0 6 0 0 0 0 0 0Anterior junctional

Healed 0 0 0 0 0 0 0 0 0 0 0 0Unhealed 0 0 l 11 0 0 0 0 0 0 0 0Retear or progression 0 0 0 0 0 0 0 0 0 0 0 0

Total 0 0 1 1 0 0 0 0 0 0 0 0

AnteriorHealed 0 0 0 0 0 0 0 0 0 0 0 0Unhealed 0 0 0 0 0 0 0 0 0 0 0 0Retear or progression 0 0 0 0 0 0 0 0 0 0 0 0

Total 0 0 0 0 0 0 0 0 0 0 0 0

* Each of the five zones represents one-fifth of the length of the meniscus. See text and Fig. 4.t Each of the three areas represents one-third of the width of the meniscus. See text and Fig. 4.1:A vertical longitudinal tear. All radial tears.

818 c. B. WEISS ET AL.

TABLE IV-C

CORRELATION OF THE LENGT H OF THE TEARS (IN MILL IMETE RS) AND HEAL ING AT THE T IME OF REPEAT ARTHROSCOPY

Lateral Meniscus (N = 24) Medial Meniscus (N = 8)Outert Centralt Innert Outert Centralt Innert

Zone* 10 10 10 10 10 10

PosteriorHealed 0 8 0 0 2 0 0 0 0 1 1 0 0 0 0 0 0 0Unhealed 0 1 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0Retear or progression 0 1 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0

Total 0 10 0 0 4 0 0 0 0 2 2 0 0 0 0 0 0 0

Posterior junctionalHealed 0 1 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0Unhealed 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0Retear or progression 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Total 0 1 0 0 0 0 0 0 0 3 1 0 0 0 0 0 0 0

MiddleHealed 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Unhealed 0 0 0 0 0 0 5 0 0 0 0 0 0 0 0 0 0 0Retear or progression 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0

Total 0 0 1 0 0 0 6 0 0 0 0 0 0 0 0 0 0 0

Anterior junctionalHealed 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Unhealed 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0Retear or progression 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Total 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0

AnteriorHealed 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Unhealed 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Retear or progression 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Total 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

* Each of the five zones represents one-fifth of the length of the meniscus. See text and Fig. 4.

t Each of the three areas represents one-third of the width of the meniscus. See text and Fig. 4.1:All radial tears.

Still, the location of a tear within the meniscus seemed to All of the tears of the medial meniscus that were eval-be the most important determinant of healing of vertical uated by second-look arthroscopy were ten millimeters longlongitudinal tears of the lateral meniscus. or less. The length ofthe tear did not appear to affect healing

(Table IV-C). Of the four tears that were in the outer areaMedial Meniscal Tears of the posterior junctional zone, two healed (Cases 6 and

Information on type, location, depth, and length was 9) and two showed no evidence of healing (Cases 17 andavailable for all eighteen medial meniscal tears. All eighteen 21) at the time of the repeat arthroscopy. There were nowere vertical longitudinal tears. Ten tears (56 per cent) were obvious differences between the two tears that healed andpartial thickness on the tibial surface; two (1 1 per cent), the two that did not. Similarly, of the four partial-thicknesspartial thickness on the femoral surface; and six (33 per tears in the outer area of the posterior zone, two healedcent), full thickness (Fig. 4). Eight of the tears were in the (Cases 12 and 16) and two did not (Cases 27 and 28).outer area of the posterior junctional zone; six, in the outer The two partial-thickness tears of the undersurface ofarea of the posterior zone; three, in the central area of the the outer area of the posterior zone that did not heal wereposterior zone; and one, in the central area of the anterior isolated lesions. Initially, they were left alone and the kneesjunctional zone (Table II). Fifteen of the eighteen tears were were not immobilized since there was no associated liga-five to ten millimeters long; two, ten millimeters long or mentous or meniscal lesion. However, three months afterlonger; and one was less than five millimeters long. There the diagnostic arthroscopy, both patients still had pain overwere no radial tears of the medial meniscus in this series. the posteromedialjoint line. Because these isolated meniscal

During the second-look arthroscopy, it was possible to tears were thought to be the cause of the symptoms, openassess the location, depth, and length ofeight ofthe eighteen meniscal repair was performed. When the repeat arthros-tears of the medial meniscus (Tables III through IV-C). copy was done, just before the open repair, one of theHowever, because there were so few tears in each category, patients (Case 27) had some evidence of healing; about 20no conclusions could be drawn relative to each of the var- per cent of the previous partial-thickness tear had filled in.iables. Four of the eight tears in the outer areas of the The other patient (Case 28) had no evidence of healing.posterior and posterior junctional zones of the medial me-niscus healed (Table IV-A). Three of these four healed tears LysholmIl Scoreshad been partial thickness and one, full thickness (Table The responses of all fifty-two patients who answeredIV-B). the questionnaire were rated using the Lysholm-II scoring


TABLE V

CORRELATION OF HEALING OF MEDIAL AND LATERAL MENISCI AND

STABILITY OF THE KNEE AT THE TIME OF REPEAT ARTHROSCOPY

Stable Knees (N = 12) Unstable Knees (N = 20)Immobilized (N = 4) Not Immobilized (N = 8) Immobilized (N = 16) Not Immobilized (N = 4)

Longit. Radial Longit. Radial Longit. Radial Longit. Radial

Lat. Med. Lat. Med. Lat. Med. Lat. Med. Lat. Med. Lat. Med. Lat. Med. Lat. Med.

Healed 2 2 0 0 2 1 0 0 7 1 0 0 1 0 0 0Unchanged 0 0 0 0 0 3 2 0 2 1 2 0 1 0 1 0

Retear orprogression 0 0 0 0 0 0 0 0 2 0 1 0 1 0 0 0Total 4 0 6 2 13 3 3 1


VOL. 71-A, NO. 6. JULY 1989

system22. Forty-two patients (80 per cent) had a score of 80points or more (mean, 93 points), and the other ten (20 percent) had a mean score of7l points (range, 53 to 79 points).All six patients who had had subsequent surgical treatmentof the meniscal lesion eventually had a satisfactory score(average, 86 points). In the ten patients who had an unsat-isfactory score, there was no evidence that the poor resultwas due to the meniscal lesion. Six of the ten had to use afunctional brace and limit activity because of anterior cru-ciate instability. Three had difficulty going up and downstairs and performing other activities because of patello-femoral symptoms, and the remaining patient had trouble-some limitation of motion of the knee secondary to arthro-fibrosis that had developed after reconstruction of the an-tenor cruciate ligament. This patient had a repeat arthros-copy and arthroscopic lysis of adhesions twenty-sevenmonths after reconstruction ofthe anterior cruciate ligament.At that time, the tear of the lateral meniscus, which hadbeen previously left alone, had healed.

At the time of follow-up, the forty-two patients whohad a satisfactory Lysholm-II score had no effusion or ten-derness at the joint line. The McMurray test was negative,and none of the mechanical symptoms (clicking, locking,or giving-way) that are characteristic of a meniscal lesionwere noted. We searched diligently for tenderness at thejoint line in the region of the previous meniscal tear in allknees, but none was noted.

Other Factors that Could Have influenced HealingAs already noted, the repeat arthroscopic procedures

were performed three to 100 months (average, twenty-sixmonths) after the initial diagnostic arthroscopy in thirty-twoof the original seventy-five patients (Table III). Of thesethirty-two procedures, two were performed because of per-sistent symptoms and four, because another sports-relatedinjury had occurred. The other twenty-six procedures wereelective, and they were done to determine the status of thepreviously torn meniscus. Of these twenty-six asymptomatictears, twenty-one were vertical longitudinal and five, radial.Sixteen of the vertical longitudinal tears had healed and fivehad not. As already noted, there were too few vertical lon-gitudinal tears in each subset to permit statistically signif-icant correlations between location (Table IV-A), depth(Table IV-B), or length (Table IV-C) and the healing of

these tears.None of the five radial tears that were evaluated (ex-

cluding the one that had extended) had any evidence ofhealing.

Of the sixteen healed vertical longitudinal tears, at thetime of diagnosis eight had been chronic and eight, acute(Table III). At the most recent follow-up, stability of theknee was determined by clinical assessment of laxity, usingthe Lachman and pivot-shift tests. The Lachman test, asdescribed by Torg et al.32, is an anterior drawer test that isperformed with the knee flexed 20 to 30 degrees. Torg etal. classified the result as positive if the anterior tibial trans-lation on the side of the injured knee was increased com-pared with that on the side of the uninjured knee. In thepresent study, we graded the Lachman test as 1 + if therewas a one to five-millimeter increase of anterior tibial trans-lation on the side of the injured knee compared with thaton the side of the normal knee, as 2 + if there was a six toten-millimeter increase, and as 3 + if there was an elevento fifteen-millimeter increase. However, instrumented testsof laxity were not performed. Instead, the grading was basedon the examiners clinical feel of the knee.

The findings of the pivot-shift test, as described byGalway and MacIntosh3, correlate better with functionalinstability in a knee in which the anterior cruciate ligamentis insufficient. This test demonstrates the shift or reductionof the anterolaterally displaced tibia that takes place as val-gus force is applied and the knee moves from full extensionto a flexed position. We rated this shift as mild, moderate,or severe, based on the clinical feel of the knee.

We designated a knee as stable if the patient had es-sentially normal laxity - that is, if the Lachman test wasless than 1 + and the pivot shift was negative. Knees forwhich the Lachman test was more than 1 + , or for whichthe pivot shift was positive, were considered unstable. Atthe same time, we assessed the functional instability on thebasis of the history and the Lysholm-II score, but thesesubjective data were not considered in the determination ofstability. Of the sixteen healed vertical longitudinal tears,seven were in knees that were considered stable and ninewere in unstable knees.

We also tried to determine whether immobilization ofthe knee after the tear was first diagnosed had any influenceon healing (Table V). Ofthe sixteen knees that had a vertical



longitudinal tear that healed, twelve (75 per cent) had beenimmobilized for six weeks in a plaster cast or in a locked-hinge brace in 30 to 40 degrees of flexion after the initialarthroscopy. This was done because other associated me-niscal tears had been repaired or ligaments had been recon-structed at the time of the initial arthroscopy. The other four(of the sixteen) knees had not been immobilized, but theyhad healed.

Of the seven knees that had a vertical longitudinal tearthat did not heal, three (43 per cent) had been immobilizedand four had not. Two of the three knees that had beenimmobilized had a subsequent sports-related injury that wasthought to be severe enough to cause a tear in a normalmeniscus. These two knees had a repeat tear, or an extensionof a tear that had been previously classified as stable , sixteenand twenty-four months after the initial arthroscopy. Thetwo patients had no meniscal symptoms before the reinjuryand, therefore, one can only speculate as to whether thesetears had healed before the reinjury. In the third knee inwhich a vertical longitudinal tear caused recurrent symptomsafter a sports-related injury, the injury occurred elevenmonths after the initial arthroscopy. This knee had chronicanterior-cruciate insufficiency and had not been immobilizedafter the initial arthroscopic procedure. The anterior cruciateligament was reconstructed and a partial lateral meniscec-tomy was performed after the reinjury. Eventually the symp-toms resolved.

Excluding the two isolated longitudinal tears that weretreated by open meniscal repair three months after the di-agnosis, five longitudinal tears did not heal. When thesefive tears were first diagnosed arthroscopically, four wereassociated with acute and one, with chronic symptoms.Three were partial thickness and the other two, full thick-ness. Four were five to ten millimeters long and one, tenmillimeters long or longer. Two of these five tears were inthe outer area of the posterior junctional zone of the medialmeniscus; two, in the central area of the anterior junctionalzone of the lateral meniscus; and one was in the outer areaof the posterior zone of the lateral meniscus. The reason orreasons why these tears did not heal are not clear, because,excluding the two tears in the central area of the anteriorjunctional zone of the lateral meniscus, the location, length,and depth of these tears did not differ notably from thoseof the vertical longitudinal tears that did heal.

Of the six knees that had a radial tear of the lateralmeniscus and repeat arthroscopy (including the radial tearthat extended), four had been immobilized and two had not.At the most recent follow-up, five of the six tears wereasymptomatic and one had become symptomatic because ithad extended. This traumatic extension occurred six monthsafter the initial arthroscopy, and it was treated by contouringthe flaps on each side of the tear back to form an intactmargin.

Of the six patients who had additional surgical treat-ment because of symptoms related to the meniscal tear, four(three vertical longitudinal tears and the one radial tear) hadan unstable knee . In these four knees , the stable tear had

extended as a result of a sports-related injury; the threevertical longitudinal tears became bucket-handle lesions andthe one radial tear extended to involve nearly the full widthof the meniscus. Three of these four knees were treated witharthroscopic excision of the bucket-handle fragment andone, with sufficient tapering of the two sides of the radialtear to produce a smooth, stable rim.

The two patients who had a partial-thickness verticallongitudinal tear in the outer area of the posterior zone onthe tibial surface of the medial meniscus continued to havepersistent symptoms along the posteromedial joint line afterthe original diagnostic arthroscopy. No other cause for thesymptoms was found. At the time of repair, no articulardamage was seen that could account for the posteromedialsymptoms. In fact, no localized degenerative changes wereobserved in the articular cartilage adjacent to any of thethirty-two vertical longitudinal or radial meniscal tears thatwere examined with repeat arthroscopy.

At the most recent follow-up, nineteen of the thirty-two patients who had a repeat arthroscopy had a 1 + Lach-man test (five millimeters or more of anterior tibial excursionon the side of the involved knee compared with that on theside of the contralateral knee) or a positive pivot shift.However, only six patients had sufficient instability for theLysholm-1122 score to be unsatisfactory. There may be sev-eral explanations for this finding. First, the Lysholm-II scoremay not accurately reflect laxity or give sufficient emphasisto instability of the knee (only 25 points are assigned toinstability). Second, these patients may have altered theirlevel of activity enough to compensate for the laxity of theknee. Third, some patients may be able to compensate forlaxity of the knee by setting the quadriceps and hamstringsmuscles at just the right time to avoid episodes of instabilityof the knee, while non-athletic patients who have similarlaxity of the knee may be unable to do so.

Discussion

The shortcomings of this study are quite evident. Thiswas a retrospective review that was based on arbitrarilyselected criteria to define a stable meniscal lesion. Two ofus (K. E. DeH. and J. G.) initiated the study because ofthe belief that an undefined subset of meniscal tears causeno additional symptoms in the knee if the tear is left alone.The criteria that were used to define stability were basedon clinical judgment, as described in the Materials andMethods section. Since few stable meniscal lesions fulfilledthe criteria, patients from both Sweden and New York wereselected, even though we realized that the two populationscould not be combined into a strictly homogeneous group.Finally, when we attempted to determine if several differentparameters had any value for the prediction of which lesionswould heal, the resulting subsets of knees were so smallthat only limited conclusions could be drawn.

Despite these drawbacks, this study is valuable becauserepeat arthroscopy documented healing of some of thesestable meniscal lesions. Since we found no symptoms orclinical signs in the knees that had a stable meniscal tear


VOL. 71-A, NO. 6. JULY 1959

that did not heal, one wonders whether clinical and historicalfollow-up provides reliable information as to the outcomeofsuch tears. Since the body ofthe meniscus has no sensorynerve supply, the clinical findings that are associated withmeniscal tears are often difficult to interpret. It is impossibleto estimate the number of stable meniscal lesions that arenot recognized during clinical evaluation or even duringarthroscopic examination. It should thus be emphasized that,during arthroscopic evaluation, it is essential to probe thor-oughly and visualize both surfaces of each meniscus toidentify any partial-thickness tears.

The effect of immobilization of the knee on the healingof stable vertical longitudinal meniscal tears was not estab-hished by this study. However, three-quarters (twelve ofsixteen) of the healed vertical longitudinal tears that wereexamined by repeat arthroscopy were in knees that had beenimmobilized for four to six weeks after the initial arthro-scopic examination (Table III). Of the seventeen knees thathad a vertical longitudinal tear and had been immobilized(Table V), four were stable and thirteen were unstable. Allfour vertical longitudinal tears in the stable knees healed,as did eight of the thirteen in the unstable knees. Of nineknees that had a vertical longitudinal tear and had not beenimmobilized, six were stable and three, unstable. Three ofthe six vertical longitudinal tears in stable knees and one ofthe three in unstable knees healed.

The belief that a stable knee is a prerequisite for thehealing of a torn but stable meniscus was not supported byour study, but the numbers of knees in the different groupswere small (Table V). Ofthe sixteen healed tears, nine werein knees that were classified as unstable (a Lachman test of1 + or greater or a positive pivot shift, or both) and sevenwere in knees that were classified as stable at the most recentfollow-up. Of the nine unstable knees in which the tearhealed, eight had been immobilized after arthroscopy andone had not. The immobilization of these unstable kneesmay have provided protection that encouraged healing. Intwenty-one of the thirty-two knees that had repeat arthros-copy, a reconstruction or repair of the anterior cruciate hg-ament had been attempted, but stability was restored (thatis, the Lachman test was less than 1 + and the pivot shiftwas negative) in only seventeen of the twenty-one at themost recent follow-up. Whether the repair or reconstructionof the ligament had initially been tight enough to protectthe healing meniscus from the abnormal forces that areassociated with instability of the knee is unknown. Amongthe seven stable knees in which the tear did heal, there was

no suggestion that immobilization had any effect. Of thoseseven knees, four had been immobilized and three had not(Table V).

Of the sixteen vertical longitudinal tears that healed,eight were chronic and eight were acute at the time of theinitial arthroscopy. No attempt was made to debride orfreshen the margins of the chronic tears, but the probingwith the nerve-hook conceivably could have stimulated ahealing response. Since the peripheral part of the meniscusis vascularized, tears in this area heal by the formation offibrocartilage, with the result that the gross and histologicalappearance of the meniscus is normal2626.

We believe that vertical longitudinal tears in the vas-cular outer area in the posterior zone of both the lateral andthe medial meniscus probably are avulsion injuries. Thisseems especially true of tears of the lateral meniscus thatare associated with a torn anterior cruciate ligament. Thosetears have a great potential for healing, as shown by thefact that, in this series, thirteen such tears healed (TableIV-A). As already noted, Arnoczky et al. showed that tearsinvolving the avascular inner two-thirds of the meniscus donot heal in dogs unless a vascular channel to the peripheryis created2. The findings in our study supported this obser-vation since none of the five asymptomatic radial tears thatwere re-examined during second-look arthroscopy showedany signs of healing.

The best way to manage a stable radial tear that isconfined to the avascuhar inner one-third of the meniscusremains to be established. In the few knees that had thatlesion in this study, leaving the tear alone appeared to haveno detrimental effect. However, a larger series and a longerfollow-up might show that these tears serve as areas of stressconcentration that lead to extension of the tears. Controlledprospective studies are needed to determine if this is true.

In conclusion, stable meniscal lesions are often mci-dental findings in patients who have had an injury to anotherstructure in the knee. Stable vertical longitudinal tears, es-pecially in the vascular outer area of the meniscus, shouldbe left alone unless no other lesions are present and thereis sufficient disability to warrant treatment. A chronic tearor a tear in an unstable knee is not necessarily a contrain-dication to leaving a stable meniscal tear alone.

No conclusions as to the management of stable radialtears are warranted on the basis of this study. However, thesecond-look arthroscopic examinations showed no evidenceof healing in any radial tear. Prospective longitudinal studiesare needed to establish the optimum treatment for such tears.

References1 . APPEL, HELGE: Late Results after Meniscectomy in the Knee Joint. A Clinical and Roentgenologic Follow-up Investigation. Acta Orthop.

Scandinavica, Supplementum 133. 1970.2. ARNOCZKY, S. P.; MARSHALL, J. L.; JOSEPH. A.; JAHRE, C.; and YOSHIOKA, M.: Meniscal Nutrition - An Experimental Study in the Dog.

Trans. Orthop. Res. Soc. , 26: 127, 1980.3. BALKFORS, BENGT: The Course of Knee-Ligament Injuries. Acta Orthop. Scandinavica, Supplementum 198, 1982.4. CABAUD, H. E. ; RODKEY, W. G.; and FITZWATER, J. E.: Medial Meniscus Repairs. An Experimental and Morphologic Study. Am. J. Sports

Med., 9: 129-134, 1981.5. CARGILL, A. OR. . and JACKSON, J. P.: Bucket-Handle Tear of the Medial Meniscus. A Case for Conservative Surgery. J. Bone and Joint Surg.,

58-A: 248-251, March 1976.6. CASSIDY. R. E., and SCHAFFER. A. J.: Repair of Peripheral Meniscus Tears. Am. J. Sports Med., 9: 209-214. 1981.7. DANDY, D. J.. and JACKSON. R. W.: Meniscectomy and Chondromalacia of the Femoral Condyle. J. Bone and Joint Surg., 57-A: I I 16-1119,

Dec. 1975.


8. DEHAvEN, K. E.: Commentary: Repair of Peripheral Meniscus Tears. A Preliminary Report. Am. J. Sports Med., 9: 213-214, 1981.9. DEHAVEN, K. E. : Rationale for Meniscus Repair or Excision. Clin. Sports Med. , 4: 267-273, 1985.

10. DEHAVEN, K. E. , and BLACK, K. : Personal communication, 1987.1 1. FAIRBANK, T. J.: Knee Joint Changes after Meniscectomy. J. Bone and Joint Surg. , 30-B(4): 664-670, 1948.12. FUKUBAYASHI, TORU, and KUROSAWA, HISASHI: The Contact Area and Pressure Distribution Pattern of the Knee. Acta Orthop. Scandinavica, 51:

871-879, 1980.13. GALWAY, H. R., and MACINTOSH, D. L.: The Lateral Pivot Shift: A Symptom and Sign ofAnterior Cruciate Ligamentlnsufficiency. Clin. Orthop.,

147: 45-50, 1980.14. HAMBERG, PER; GILLQUIST, JAN; and LYSHOLM, JACK: Suture of New and Old Peripheral Meniscus Tears. J. Bone and Joint Surg. . 65-A: 193-

197, Feb. 1983.15. HEATLEY, F. W.: The Meniscus - Can It Be Repaired? An Experimental Investigation in Rabbits. J. Bone and Joint Surg. , 62-B(3): 397-402,

1980.16. HENNING, C. E. ; JOLLY, B. L. ; and SCOTT, G. A. : Arthroscopic Intra-Articular Meniscus Repair - Healing Parameters. Read at the Annual

Meeting of the The American Academy of Orthopaedic Surgeons, Las Vegas. Nevada, Jan. 27, 1985.17. INSALL, J.: Surgery of the Knee. New York, Churchill Livingstone, 1984.18. JACKSON, J. P.: Degenerative Changes in the Knee after Meniscectomy. British Med. J. , 2: 525-527, 1968.19. JOHNSON, R. J.; KETTELKAMP, D. B. ; CLARK, WILLIAM; and LEAVERTON, PAUL: Factors Affecting Late Results after Meniscectomy. J. Bone

and Joint Surg. , 56-A: 719-729, June 1974.20. KETTELKAMP, D. B. , and JACOBS, A. W. : Tibiofemoral Contact Area - Determination and Implications. J. Bone and Joint Surg. , 54-A: 349-

356, March 1972.21 . KRAUSE, W. R. ; POPE, M. H. ; JOHNSON, R. J. ; and WILDER, D. G. : Mechanical Changes in the Knee after Meniscectomy. J. Bone and Joint

Surg. , 58-A: 599-604, July 1976.22. LYSHOLM, J. , and GILLQUIST, J. : Evaluation of Knee Ligament Surgery Results with Special Emphasis on Use of a Scoring Scale. Am. J. Sports

Med., 10: 150-154, 1982.23. LEVY, I. M.; T0RzILLI, P. A. ; and WARREN, R. F. : The Effect of Medial Meniscectomy on Anterior-Posterior Motion of the Knee. J. Bone and

Joint Surg. , 64-A: 883-888, July 1982.24. OCoNNoR, RICHARD: Textbook of Arthroscopic Surgery. Philadelphia, J. B. Lippincott, 1984.25. ORETORP, NILS; ALM, ANDERS; EKSTROM, HANS; and GILLQUIST, JAN: Immediate Effects of Meniscectomy on the Knee Joint. The Effects of

Tensile Load on Knee Joint Ligaments in Dogs. Ada Orthop. Scandinavica, 49: 407-414, 1978.26. SCAPINELLI, R.: Studies on the Vasculature of the Human Knee Joint. Acta Anat. , 70: 305-331 , 1968.27. SEEDHOM, B. B. ; DOWSON, D. ; and WRIGHT, V. : Functions of the Menisci - A Preliminary Study. In Proceedings of The British Orthopaedic

Research Society. J. Bone and Joint Surg. , 56-B(2): 381-382, 1974.28. SHRIVE, N.: The Weight-Bearing Role of the Menisci of the Knee. in Proceedings of The British Orthopaedic Research Society. J. Bone and Joint

Surg., 56-B(2): 381, 1974.29. SMILLIE, I. S.: Injuries of the Knee Joint. Ed. 5. New York, Churchill Livingstone, 1973.30. TAPPER, E. M., and HOOVER, N. W.: Late Results after Meniscectomy. J. Bone and Joint Surg.. 51-A: 517-526, April 1969.31. TEGNER, Y., and LYSHOLM, J.: Rating Systems in Evaluation of Knee Ligament Injuries. Clin. Orthop., 198: 43-49, 1985.32. TORG, J. S.; CONRAD, WAYNE; and KALEN, VICK1E: Clinical Diagnosis of Anterior Cruciate Ligament Instability in the Athlete. Am. J. Sports

Med., 4: 84-91, 1976.33. WALKER, P. 5., and ERKMAN, M. J.: The Role of the Menisci in Force Transmission across the Knee. Clin. Orthop.. 109: 184-192, 1975.


Documents

bismillah