Archives of Orthopaedic and Traumatic Surgery

Arch Orthop Trauma Surg (1985) 104:1-6

© Springer-Verlag 1985

OriginalArticles Clinical and Radiological Long-term Results After Primary Knee Ligament Surgery M Jarvinen and P Kannus Department of Clinical Sciences, University of Tampere, and Department of Surgery, University Central Hospital, Tampere, Finland

Summary A total of 77 patients were reviewed 2 to 7 years (average 40 months) after primary operative treatment of an acute knee ligament injury Fifty percent of the injuries occurred during sports events, the other causes were traffic in 29% of cases and accidents at work, home etc in 21 % of the cases Crosscountry skiing was the cause in half of the sports accidents There were two annual peaks in the distribution of the injuries: January-March (40 % of the cases) and August-October ( 29%) The total number of operations increased steeply during our observation period; 57 % of the operations were performed from 1980 to 1981 The most common types of injury were a combination rupture of the anterior cruciate and the medial collateral ligaments ( 36 % of the cases) and a medial collateral ligament rupture alone (31 %) According to the follow-up study, excellent or good results were obtained in 59% of the cases and fair or poor in 41% The pathological clinical and radiological changes after knee ligament injury accumulated markedly under some specific findings. Characteristic of the patients with excellent healing results was that they were younger, received isolated ligament injury and played a great amount of sport, and of the patients with fair or poor results that they were older, received combined ligament injury, suffered quadriceps muscle atrophy and had diminished sporting activity. Zusammenfassung Bei 77 Patienten mit primer operierten frischen Kniebandverletzungen wurde 2-7 Jahre (Mittel: 40 Monate) nach der Verletzung eine Nachuntersuchung durchgefiihrt Atiologisch dominierten Sportverletzungen (50 %, die Halfte davon Offprint requests to: M Jarvinen, M D , Department of Clinical Sciences, University of Tampere, Teiskontie 35, SF-33520 Tampere 52, Finland

Skilanglaufverletzungen), gefolgt von Verkehrsunfallen (29 %) und anderen (21 %, Arbeits-, Haushalts u a Unfalle) In der jahrlichen Verteilung der Verletzungen gab es 2 Spitzen: Januar-Mirz (40 %) und August-Oktober (29 %) Wahrend der Untersuchungszeitspanne nahm die Zahl der Operationen erheblich zu: 57 % aller Operationen wurden 1980-81 durchgefiihrt Am hufigsten waren kombinierte Rupturen des vorderen Kreuzbandes und medialen Seitenbandes ( 36%) sowie isolierte Rupturen des medianen Seitenbandes ( 31%) Sehr gute oder gute Resultate fanden sich bei 59%, befriedigende oder ma Bige bei 41 % der Falle Pathologische klinische und radiologische Veranderungen traten bei bestimmten Diagnosen gehauft auf Patienten mit sehr guten oder guten Ergebnissen waren jiinger, hatten isolierte Bandverletzungen und trieben aktiv Sport; befriedigende oder m 13Bige Ergebnisse fanden sich bei alteren Patienten, bei kombinierten Bandverletzungen, bei Atrophie des M quadriceps femoris sowie bei geringer Sportaktivitat.

In the past decades, the knowledge of knee ligament injuries, especially in the area of diagnosis and treatment, has increased significantly Today, in the diagnosis of acute knee ligament injuries new clinical tests combined with arthroscopy are regarded as the method of choice l9, 17 l Early surgical intervention after a careful clinical examination under anesthesia has already for some decades been generally accepted as the prevalent method in the treatment of these injuries l 15, 18, 20, 23 l However, the treatment of knee ligament injuries is still a serious problem owing to the functionally very complex role of the knee ligaments which makes them difficult to restore and

M Jirvinen and P Kannus: Knee Ligament Surgery

2

reconstruct This fact, together with the general tendency toward an increasing number of acute knee ligament injuries l2, 3, 4, 8 l, has played an important part in focusing our attention also on this interesting topic. Little material dealing at the same time with subjective, clinical, and radiological long-term results of primary knee ligament surgery has been published. There has been hardly any material attempting to combine these aspects in the analysis of the overall results and grading the patients for each healing group However, the subjective evaluation and clinical and radiological objective findings are equally important when presenting the overall healing results after knee ligament surgery Another interesting question related to this topic is what factors are associated with each healing group Are there any distinctive characteristics in the patients with excellent, or poor, healing compared with the other patients? The present study is a retrospective evaluation of patients operated on at Tampere University Central Hospital from 1975 to 1981 because of an acute knee ligament injury Our purpose was to clarify the etiology and the extent of these injuries and the operative findings in the patients as well as to establish the subjective, clinical, and radiological long-term healing results We were also interested in finding out the characteristics of the patients with, on the one hand, excellent and, on the other, fair or poor healing. Patients and Methods During the period 1975-1981 96 patients were operated on at Tampere University Central Hospital because of an acute knee ligament injury The patients with other simultaneous injuries (knee fractures etc ) were not counted, but the patients with ligamentous insertion fragments alone were included The group consisted of 58 men and 38 women with the left knee being operated in 46 and the right in 50 cases The mean age of the patients at the time of the operation was 36 5 years ranging from 8 to 63 years, 33 of them being under 30 years The mean age of the men was 33 and of the women 40 years The age and sex distribution of the patients is presented in Fig 1. Clinical and radiological reexamination was performed in 61 patients 2 to 7 years after the operation (mean 40 months, median 34 months) and 16 patients answered a questionnaire. Six patients had died and 13 could not be traced The follow-up study thus came to include 80 % of the patients Because of the dropout of 19/96 patients, we made a comparison analysis of these and the reviewed patients based on the hospital records There were no significant differences between these groups in any aspect studied (age, sex, pre and postoperative sporting activity, pre and postoperative status, the etiology of the injury, the time between injury and operation, the year and month of the operation, left and right knee, operative findings, postoperative complications, and sick-leave period) The reviewed 80% of the patients thus seemed to represent the whole group quite well and we were able to continue the study.

N 20.

o MALE e D FEMALE 18 -

15. 13

19 AL

11

12

10.

5,

a M

0-9

.

4 I

10-19 20-29

-" 30-39

40-49

50-59

60-69 AGE IN YEARS

Fig 1 Age and sex distribution of patients operated on because of acute knee ligament injury

From the patients' hospital records all personal data, sports information, and pre and postoperative findings were registered Through the questionnaire information on the subjective state of the injured knee and the postoperative physical activity was sought and obtained. The clinical examination was carried out by the same examiner and the stability of both knees was tested using the method described earlier by Balkfors l 2 l Instability was recorded according to the gradation described by Hughston et al. l 12l, used later also by Oretorp et al l 19l A difference between the injured and the uninjured knee less than 5 mm in a separation test of the joint surfaces was regarded as mild instability, 5-10 mm as moderate, and more than 10mm as severe The rotatory instability was estimated subjectively and classified as slight or moderate. In connection with the clinical follow-up study standard anteroposterior and lateral radiographs were taken of both knees Both these X-ray films and those taken preoperatively were analyzed without knowing which knee had been operated on The radiological changes were estimated by the methods described by Appel l1l and Balkfors l2 l Before calculating the incidences of these radiological changes the preoperative changes in the injured knee and the similar postoperative changes in the uninjured knee were excluded In this way it was possible to include in the results only the changes caused by this particular knee ligament injury. According to the hospital records 49 patients were operated on in the first 2 days In 84 (88 %) of all the cases the operation was performed within 1 week of the injury and in 94 (98 %) within 2 weeks Middle tears of the cruciate ligament were mainly sutured by the end-to-end technique In the insertion ruptures or avulsion fractures, the sutures were pulled through holes drilled in the femoral or tibial condyles and the insertion or avulsion was fixed to the site of the rupture in order to get the ruptured insertion or avulsion as close as possible to its normal position. Medial and lateral collateral ligament ruptures were identified by dissection with immediate primary repair and suturing of the torn ligaments Accompanying meniscus tears were treated by meniscectomy All operations were done in a bloodless field Vacuum wound drainage was applied generally for 24 h The patients were immobilized in plaster for 5 to 6 weeks after the operation Standard statistical methods, the 2 test and the Student's t test, were used to determine the significance of the differences.

3

M Jdrvinen and P Kannus: Knee Ligament Surgery N 30

TOTAL SPORT

r E

31

Table 1 Distribution of patients according to cause and month of the injury

.7

January February March April May June July August September October November December

25

N

20.

15

10

SSSSSSSTOO SSSSSSSSSSTOO SSSSSSSSSSSSTOO SST STTT SSSTTO STTT SSTTTTTTO SSSTTOOOO SSTTTOOOOO SSTTTOO SSSTTO Total 96

5

S = sport; T = traffic; O = other (work, spare time, etc ) U

1915

1976

1977 1978 1979

1980

1981 YEAR

Fig 2 Distribution of patients according to year of operation and the role of sports as the cause of the injury

Results During our observation period there was an enormous increase in the number of operations: in the years 1980-1981 55 ( 57%) patients were operated on (Fig 2) In half of the operated patients the trauma was sustained during sports events (Fig 2 and Table 1) The other causes were traffic in 28 and accidents at work, home etc in 20 cases Cross-country skiing was the cause in nearly half of the sports accidents ( 23 persons, 48%) Other forms of sports were football in eight, volleyball in four, ice hockey in three, motorsports and track and field in two cases each, and one case each in boxing, cycling, gymnastics, slalom, weight-lifting, and wrestling. There were two annual peaks in the distribution of the injuries: January-March (38 cases, 29 sports injuries) and August-October ( 28 cases, seven sports injuries) (Table 1) Rupture of the anterior cruciate ligament combined with the medial compartment was found in 35 (36 %) patients and an isolated rupture of the medial collateral ligament was operated on in 30 ( 31%) patients (Table 2). According to the subjective evaluation 14 ( 18 %) patients reported excellent healing of the operated knee, in 24 ( 31 %) patients the healing was good, in 31 (40 %) fair, and in eight ( 11%) poor Sporting activities had diminished or were given up by 45 patients and of these 31 reported that the knee ligament injury was the cause of this reduction The sickleave period was on average 3 6 months and the physically active patients were in general unable to

Table 2 Classification of injuries and incidence of different combinations of ligament tears Group

Associated injury

N

ACL

Isolated Medial compartment Lateral compartment

5 35 (36 %) 3

PCL

Isolated Anterior cruciate, Medial compartment Medial compartment

10 7

MED

Isolated

30 (31 %)

LAT

Isolated Both cruciates

Total

3

2 1 96

The classification of the injuries is made according to Lysholm et al l 17 l because of the fact that the majority of the injuries are combinations of several ligament tears: ACL: Rupture of the anterior cruciate ligament with or without associated medial or lateral compartment injury Patients with a rupture of both cruciate ligaments are classified as group PCL PCL: Rupture of the posterior cruciate ligament with or without associated anterior cruciate ligament or medial compartment injury MED: The medial ligament complex included, besides the medial collateral ligament, also the medial meniscus LA T: Single lateral compartment tears

participate in sports for about 6 0 months In six cases the knee injury was the reason for a change of occupation. The clinical and radiological changes accumulated markedly with some specific findings The most frequent clinical pathological findings were a mild loss of flexion (17 patients), crepitation of the patella (31), slight abduction-adduction instability in full

M Jarvinen and P Kannus: Knee Ligament Surgery

4

LIGAMENT CALCIFICATION FFMORAI MFDIA CONDYL OSTEOPHYTES OF FEMOR MEDIALCONDYLE (14) NARROWING OFMEDIAL JOINT SPACE120)

OSTEOPHYTES OF >PATELLA 113) _OSTEOPHYTES OF EMINENTIA (25)

OSTEOPHYTES OF TIBIAL MEDIAL CONDYLE (151

Fig 3 Frequency of the six most common roentgenological changes caused by knee ligament injuries

extension (18) and in mild flexion ( 37), and a positive anterior ( 19) and posterior (8) drawer sign These findings together covered 65 % of all the clinical pathological findings and other recorded findings, altogether 24 different factors, were quite uncommon Mild anteromedial instability was registered in 16 patients and mild anterolateral in four A decrease in the circumference of the thigh indicating quadriceps muscle atrophy, mean 1 3 cm, was found in 31 patients When a comparison was made between the patients with anterolateral instability or instability in abduction-adduction with the knee in extension and the patients with anteromedial rotatory instability or abduction-adduction instability with the knee in mild flexion, no difference was noted in the final results with respect to poor healing In contrast a highly significant difference was found between the patients with poor healing results and the others with respect to quadriceps muscle atrophy (P
In the postoperative roentgen analysis of these knees, all pathological changes caused by the injury to the knee ligaments could be accommodated under 26 specific roentgenological diagnoses The six most characteristic roentgen changes were osteophytes in the patella, in the eminentias of the tibia, and in the medial condyles of the femur and the tibia, narrowing of the medial joint space as well as ligament calcification in the medial femoral condyle (Fig 3) These changes together covered 70 % of all the pathological changes caused by the knee ligament injury and other findings occurred only in single cases. In order to find out the characteristics of the different healing groups we established subjective, clinical, and radiological criteria for assigning the patients, on the one hand, to an excellent and, on the other, to a fair or poor overall healing group According to these criteria 16 patients with excellent healing were found The patients were put in this group if in the questionnaire they reported excellent or good repair of the operated knee and if in the clinical follow-up study and roentgen analysis at most only one pathological change was found In the comparison analysis the four characteristics typical of this patient group were: young in age, isolated ligament injury, injury sustained during sports, and sporting activity continued after trauma at at least the level of before the trauma (Table 3). In the follow-up study, the group of patients with fair or poor results consisted of 25 patients In the patients belonging to this group four or more pathological findings in the clinical examination with at least two pathological changes in the roentgen analysis were recorded in addition to the fact that the

Table 3 Some characteristics of patients with excellent healing results (compared with those of other patients)

Age Sports injury Isolated medial collateral ligament rupture Sports activity not diminished after operation

Excellent results (16 patients)

Others (45 patients)

Statistical significance

23 5 (mean) 12/16 (75%) 11/16 (69 %) 15/16 (94 %)

41 2 (mean) 18/45 (40 %) 11/45 (24%) 11/45 (24 %)

P
Table 4 Some characteristics of patients with fair or poor healing results (compared with those of other patients)

Age Combined ligamentous rupture Sports activity diminished after operation Quadriceps muscle atrophy (>lcm in the circumference of the thigh compared with the uninjured knee)

Fair of poor results (25 patients)

Others (36 patients)

Statistical significance

42 0 (mean) 17/25 (68 %) 22/25 (88 %) 18/25 (72 %)

32 8 (mean) 16/36 (44%) 17/36 (47%) 8/36 (22%)

P< O 01 P
(ttest) (X2 test) (X2 test) (X2 test)

M Jdrvinen and P Kannus: Knee Ligament Surgery

patients reported fair or poor healing of the operated knee in the questionnaire The characteristics typical of this patient group were: older in age, postoperative quadriceps muscle atrophy, combined injury in the knee ligaments, and reduced or discontinued sporting activity (Table 4) The remaining 36 patients, who did not meet the criteria for these two healing groups, did not differ from the whole group of patients in any aspect studied.

Discussion The background of the increase in the number of knee ligament operations in our hospital during the last years studied is not clear Apparently the trend toward operative treatment has increased in general, but a worldwide rise has also been observed in the number of these injuries in the past decade l2, 3, 4, 8 l The role of sports in the etiology of knee ligament injuries seems to be dominant almost everywhere, but the form of the sport is closely dependent on the sporting habits of the area, where the investigation is made In most studies soccer plays a major role as the cause of knee ligament injuries sustained in sports activities l6, 8, 11, 16l In our patients cross-country skiing was the cause of the injury in half of the cases where trauma was sustained in sports In the winter season, cross-country skiing is one of the most popular leisure time sport in Finland. The marked accumulation of the clinical and radiological changes with some specific findings was interesting The most common clinical findings were not surprising since, with the exception of crepitation of the patella, they are all typical knee ligament tests for the stability of the knee joint The most common radiological changes were all on the medial side of the injured knee The reason for this might be that two types of rupture (the combination rupture of the anterior cruciate and medial collateral ligaments and the isolated rupture of the medial collateral ligament) together covered 67 % of all the types of rupture and, thus, the ruptures were also mainly on the medial side of the knee joint The anterior cruciate and medial collateral ligaments are generally known to be the most commonly affected ligaments in knee injuries l8, 17, 22, 23 l. In all studies published, there are in the follow-up at least some patients with poor healing results Important reasons for poor healing are given as neglected primary diagnosis, meniscectomy in connection with the repair operation, the more mature age of the patients, and instability after the operation l15, 16, 22, 23l In addition, we found that combined ligament rupture, diminished sporting activity, and

5

quadriceps muscle atrophy are associated with poorer healing, while a younger age, isolated ligament injury, and a greater sporting activity are associated with better healing results. In the distribution of the patients into each overall healing group our results are in agreement with similar earlier studies l15, 22, 23 l The better healing results presented l5, 7, 17 l are apparently based on a better diagnosis and operative technique, but also partly on the qualified patient material consisting mainly of active participants in competitive sports with a higher muscular performance and a higher spontaneous capacity for healing. The loss of muscle tissue in the whole extremity through atrophy during postoperative immobilization is a common, harmful problem in the treatment of knee injuries l10, 13, 20, 21l We found significant quadriceps muscle atrophy more often in the patients with fair or poor healing results than in the others. This is due partly to combined ligament ruptures being a more serious injury, but also to the muscles being weaker in the injured extremity from the lack of active sporting habits before the trauma. Haggmark and Eriksson l14l have reported, when using a movable cast brace after reconstruction of the anterior cruciate ligament, no atrophy in the quadriceps muscle postoperatively and only limited decrease in the activity of the oxidative enzymes of the vastus lateralis muscle Because of the close connection between quadriceps muscle atrophy and many complaints in the operated knee, the postoperative treatment is important for the final results of these injuries and more attention should be focused on the postoperative treatment and rehabilitation. To sum up: · The number of knee ligament injuries has increased in the past few years with a simultaneously growing trend toward operative treatment of these injuries. · Sport is an etiological factor in the knee ligament injuries operated on in a great many of the cases; cross-country skiing was the cause in half the sports injuries in this group of patients. · The pathological clinical and radiological changes after knee ligament injuries seem to accumulate markedly with some specific findings. · A younger patient, isolated ligament injury, and greater sporting activity are connected with better overall healing results whereas a more mature age, combined ligament injury, and a limited sports are connected with fair and poor results. · Attention should be paid also to the postoperative treatment because of the close connection between quadriceps muscle atrophy and poor final results after the repair of knee ligaments.

6

References 1 Appel H (1970) Late results after meniscectomy in the knee joint Acta Orthop Scand lSuppll 133 2 Balkfors B (1982) The course of knee-ligament injuries. Acta Orthop Scand lSuppll 198 3 Burri C, Helbing G, Henkenmeyer H (1981) Surgical management of acute injury of the ligaments of the knee joint Ann Chir Gynaecol 70:169-175 4 Clancy WG (1983) Knee ligamentous injury in sports: the past, present and future Med Sci Sports 15:9-14 5 Eriksson E (1981) Rehabilitation of muscle function after sport injury major problem in sports medicine Int J Sports Med 2:1-6 6 Franke K (1976) Clinical experience in 130 cruciate ligament reconstructions Orthop Clin North Am 7:191-193 7 Franke K (1982) Kapsel-Bander-Lasionen des Kniegelenks Zentralbl Chir 107:553-561 8 Gillquist J (1971) Ledbandskador i knileden r de vanliga? Lakartidningen 68:6003-6008 9 Gillquist J, Hagberg G, Oretorp N (1977) Arthroscopy in acute injuries of knee joint Acta Chir Scand 48: 190-196 10 Gould N, Donnermeyer D, Gammon GG, Pope M, Ashikaga T (1983) Transcutaneous muscle stimulation to retard disuse atrophy after open meniscectomy Clin Orthop 178:190-197 11 Hagberg G (1978) On arthroscopy of the knee joint A clinical study with special reference to traumatic injuries. Link 6ping University Medical Dissertation No 57 12 Hughston JC, Andrews JR, Cross MJ, Moschi A (1976) Classification of knee ligament instabilities, part 1 and 2. J Bone Joint Surg lAml 58:159-179 13 Haggmark T (1978) A study of morphologic and enzymatic properties of the skeletal muscles after injuries and immobilization in man Medical Dissertation, Karolinska Institutet, Stockholm

M Jdrvinen and P Kannus: Knee Ligament Surgery 14 Haggmark T, Eriksson E (1979) Cylinder or mobile cast brace after knee ligament surgery: a clinical analysis and morphologic and enzymatic studies of changes in the quadriceps muscle Am J Sports Med 7:48-56 15 Liljedahl S-O, Nordstrand A (1969) Injuries to the ligaments of the knee Diagnosis and results of operation. Injury 1:17-24 16 Lysholm J (1981) Arthroscopy in surgery of the knee A clinical study with special emphasis on the diagnosis and treatment of ligament and meniscus injuries Link 6ping University Medical Dissertation No 106 17 Lysholm J, Gillquist J, Liljedahl S-O (1982) Long-term results after early treatment of knee injuries Acta Orthop Scand 53:109-118 18 O'Donoghue DH (1955) An analysis of end results of surgical treatment of major injuries to the ligaments of the knee J Bone Joint Surg lAml 37:1-13 19 Orctorp N, Gillquist J, Liljedahl S-O (1979) Long term results of surgery for non-acute anteromedial rotatory instability of the knee Acta Orthop Scand 50:329-336 20 Palmer I (1938) On the injuries to the ligaments of the knee joint Acta Chir Scand lSuppll 53 21 Smith MJ, Hutchins RC, Hehenberger D (1983) Transcutaneous neural stimulation use in postoperative knee rehabilitation Am J Sports Med 11: 75-82 22 Santavirta S, Kiviluoto O, Svartling N, Klossner O (1981) Operative treatment of acute ligamentous injuries of the knee joint Ital J Sports Traumatol 3:339-346 23 Solonen K, Rokkanen P (1967) Operative treatment of torn ligaments in injuries of the knee joint Acta Orthop Scand 38:67-80

Received August 6, 1984