Histologic Evaluation of Posterior Cruciate Ligaments From Osteoarthritic Knees

Abstract

It is controversial whether posterior cruciate ligament-retaining or posterior cruciate ligament-sacrificing (-substituting) implants should be used in total knee arthroplasty. The use of posterior cruciate ligament-retaining implants implies that the residual ligament is functional, and presumably normal, but few studies have been conducted to elucidate the histologic appearance of the posterior cruciate ligament taken from osteoarthritic knees. The purposes of the current study were (1) to evaluate the histologic appearance of posterior cruciate ligaments excised from osteoarthritic knees during primary total knee arthroplasty and to compare their appearance with posterior cruciate ligaments from knees of cadavers that were not operated on; and (2) to determine whether a correlation exists between the histologic appearance of the posterior cruciate ligament and the clinical status of the patients studied. Twenty-six posterior cruciate ligament specimens from patients with osteoarthritis and four specimens from cadavers were evaluated with the use of light and electron microscopy. Posterior cruciate ligaments from osteoarthritic knees showed greater degeneration than those from cadavers by light microscopy. Age greater than 60 years was associated with decreased collagen diameter in posterior cruciate ligaments from osteoarthritic knees as determined by electron microscopy. With the number of specimens available, the authors could not find a significant correlation between tibiofemoral alignment and mean collagen diameter or percentage of collagen occupancy. The extent of tissue degeneration of the posterior cruciate ligament could not be predicted by clinical findings. Additional studies identifying the mechanical competency of the posterior cruciate ligament in osteoarthritis would be valuable.

It is controversial whether posterior cruci-ate ligament-retaining or posterior cruciate ligament-sacrificing implants should be used in total knee arthroplasty. ^{7,13,14,16,20,24,26,29,32} Proponents for posterior cruciate ligament-retaining implants have argued that these designs provide implanted knees with posterior stability, ^12,22 an increased range of motion, ^2,18 femoral roll back to increase the moment arm of the quadriceps, ² and more physiologic proprioception ^4,16 as compared with posterior cruciate ligament-sacrificing implants. Advocates of posterior cruciate ligament sacrifice in total knee arthroplasty have shown that posterior cruciate ligament-sacrificing designs have excellent clinical long-term results, ^{5,26,28,31,32} which are comparable with the results of posterior cruciate ligament-retaining designs. ^3,7,29 Previous studies have shown that posterior cruciate ligament-sacrificing implants, especially those with posterior-stabilizing mechanisms, have no difference in outcome with respect to postoperative range of motion, ^14,32,33 gait pattern, ^5,36 or proprioception ⁸ from that of posterior cruciate ligament-retaining implants. Several biomechanical studies also have failed to show that posterior cruciate ligament-retaining knee implants have biomechanical advantages compared with posterior cruciate ligament-substituting implants. ^{10,11,21,25,34,36}

Of concern with retention of the posterior cruciate ligament is documentation that strain of the posterior cruciate ligament is variable after total knee arthroplasty, as shown by intraoperative measurement and in normal cadaver knees. ¹⁵ It seems difficult to reproduce a physiologic strain behavior, and therefore it remains questionable whether the posterior cruciate ligament functions as a posterior restraint in posterior cruciate ligament-retaining total knee arthroplasty. In addition, Alexiades et al ¹ and Kleinbart et al ¹⁹ showed that posterior cruciate ligament specimens from the patients with osteoarthritis show degenerative changes. These reports of the histologic appearance were by means of light microscopy and may be subjective. It is uncertain whether a specimen termed normal by light microscopy has normal biomechanical properties. Previous investigations, in which electron microscopy was used to evaluate collagen fibrils of ligaments, have documented that the biomechanical properties of ligaments may depend on the distribution and the diameter of collagen fibrils. ^6,23,35 An evaluation of the posterior cruciate ligament of osteoarthritic knees using electron microscopy, with quantitative evaluation of collagen diameter, is desirable and may be more relevant to the biomechanical status of the ligament. Moreover, it also is unknown whether the changes of the morphologic features of the posterior cruciate ligament identified by microscopy correlate with clinical evaluation of the corresponding osteoarthritic knees (the radiographic appearance of degenerative changes, the alignment of the knee, or its functional ability).

The purposes of the current study were to evaluate the histologic appearance of posterior cruciate ligaments harvested from osteoarthritic knees during primary total knee arthroplasty by use of light and electron microscopy, and to identify whether a correlation exists between the histologic appearance of the posterior cruciate ligaments and the clinical assessment of the knee preoperatively.

MATERIALS AND METHODS

Subjects

Twenty-six posterior cruciate ligaments were harvested from 24 consecutive patients at primary total knee arthroplasty (Osteoarthritis Group) by two of the authors (BNS and AHW). This group consisted of 15 men and nine women, with an average age of 64.2 years (range, 40–84 years). The study was approved by the institutional review board at the hospital where the total knee arthroplasties were done and informed consent was obtained from all patients. The Control Group consisted of four posterior cruciate ligaments from four cadavers (two males and two females), with an average age of 87 years (range, 78–98 years). The medical history for each patient was reviewed to ensure compliance with the inclusion and exclusion criteria for the Osteoarthritic Group (Table 1). All patients of the Osteoarthritic Group were suitable for evaluation based on the criteria listed in Table 1. Insufficient medical history for each cadaver precluded inclusion and exclusion compliance. Cadavers with osteoarthritis in the knees were excluded from the Control Group if typical degeneration of the joint cartilage was seen at the harvest of the posterior cruciate ligament.

TABLE 1: Inclusion and Exclusion Criteria

Sample Preparation

At the time of the total knee arthroplasty, the knee was approached surgically through a medial parapatellar incision. The posterior cruciate ligaments were harvested by sharp incision from bone-ligament junctions of the femur and tibia. Consistent orientation of the specimen was ensured by placement of a suture in the anterior femoral side of the harvested ligament. On removal, the specimen was placed in cold saline. The posterior cruciate ligament specimen was trimmed to eliminate synovial and fat tissue. At evaluation, the anterolateral bundle of the posterior cruciate ligament was identified and cut longitudinally with a razor blade. Small pieces of ligament tissue were cut away from three portions (femoral side, midportion, and tibia side) in the anterolateral bundle of the posterior cruciate ligament. Each small piece of ligament then was cut in half. On half was fixed in 10% neutral formalin for light microscopy, and the other ½ was fixed in 3.75% glutaraldehyde for electron microscopy.

Light Microscopy

The posterior cruciate ligament specimens fixed with formalin then were dehydrated and embedded in paraffin. The specimens were cut longitudinally with a microtome and stained with hematoxylin and eosin. At least three slides of each specimen were examined under a light microscope. The histologic appearance of the specimen was evaluated for degenerative changes as described previously. ^1,19 The degenerative changes were present when a slide showed abrupt zones of loose fibrous connective tissue with or without cystic, myxoid, and/or mucoid degeneration. ¹⁹ The extent of degenerative changes was categorized as normal, slight, mild, moderate, or marked. ¹⁹

Transmission Electron Microscopy

The specimens for electron microscopy were dehydrated and embedded in epoxy resin and cut into ultrathin cross sections at a 90° angle to the collagen fibril axis. These sections were stained with uranyl acetate and lead citrate. Each specimen was examined on at least three sections. Areas for examination were chosen for photography when the cross sections of collagen fibrils appeared to be round (indicating cross section at a 90° angle). Photographs were taken at magnifications of ×60,000 or 65,000. These photographs then were analyzed to quantify the collagen fibril diameter, area, and distribution using image analysis software, the BIOQUANT 95 system (R & M Biometrics, Inc, Nashville, TN). A percentage of collagen occupancy then was calculated. More specifically, if collagen fibers occupied 40% of a posterior cruciate ligament specimen under light microscopy and a percentage of collagen fibril area under electron microscopy was 60%, the percentage of collagen occupancy was calculated as 24% (0.4 × 0.6 = 0.24).

Clinical Evaluation

Clinical evaluation consisted of a comprehensive medical history, physical examination, and standardized radiographic examination. Clinical performance was assessed using the Knee Society clinical scoring system. ¹⁷ Physical examination included passive range of motion and evaluation of knee stability with the patient under anesthesia before surgery. Radiographic evaluation consisted of an anteroposterior view with the patient weightbearing taken within 3 months of surgery measured to determine alignment (femorotibial angle, femoral shaft-transcondylar angle, and tibial plateau-tibial shaft angle) ³⁰ and aggressiveness of the degenerative appearance in the knee. The degenerative change was graded using scales previously described. ²⁷

Statistical Analysis

The data were analyzed statistically using the Student’s t test or analysis of variance for continuous variables, and the chi square test for categorical variables. Spearman’s rank-order correlation test was used to test the association between two continuous variables. A value of p < 0.05 was regarded as statistically significant.

RESULTS

Twelve (46%) of 26 posterior cruciate ligaments from patients in the Osteoarthritis Group had moderate or marked degenerative change in at least one of three portions (femoral side, midportion, or tibial side) by light microscopy, whereas only one specimen in the Control Group had mild degenerative change (Fig 1, Table 2). In the Osteoarthritis Group, 19 of 26 osteoarthritic knees had varus deformity (tibiofemoral angle > 175°), five had valgus deformity (tibiofemoral angle < 170°), and two knees did not have marked deformity. Posterior cruciate ligaments from osteoarthritic knees showed greater degeneration than those from cadavers by light microscopy (p = 0.035). Knee alignment was not associated with histologic grade determined by light microscopy (p = 0.64). A significant correlation was not detected between histologic grade and age (p = 0.40), gender (p = 0.51), knee functional score (p = 0.67), patient functional score (p = 0.30), or radiographic grade for degenerative change (p = 0.23) in the Osteoarthritis Group (Table 3).

TABLE 2: Light Microscopic Appearance of Posterior Cruciate Ligament^*

TABLE 3: Correlation Between Clinical Factors and Degenerative Changes of the Posterior Cruciate Ligament Observed by Light Microscopy

Fig 1 A–D.: Photomicrographs of posterior cruciate ligaments from osteoarthritic knee show (A) unidirectional alignment of normal collagen fiber and (B) loose fibrous tissue (Stain, hematoxylin and eosin; original magnification, ×100). (C) A transmission electron micrograph shows that collagen fibers interpreted as normal under light microscopy consist of bulk collagen fibrils, whereas (D) collagen fibers with marked degeneration under light microscopy consist of a small amount of thin collagen fibrils under electron microscopy. Bar = 100 nm

By electron microscopy, the mean collagen fibril diameter for the Osteoarthritis Group was less at any portion than that for the cadavers (Table 4). The percentage of collagen occupancy for the Osteoarthritis Group was smaller at both femoral and tibial sides than for the Control Group (Table 5). These differences did not reach statistical significance (Tables 4, 5). Significant differences could not be detected in collagen diameter or percentage of collage occupancy among the various sections of the ligament in the Osteoarthritis Group (p = 0.99 and p = 0.77, respectively) (Tables 4, 5). The mean collagen diameter for posterior cruciate ligaments that showed moderate or marked degenerative change by light microscopy was 75.1 nm compared with posterior cruciate ligaments with less degenerative change (87.5 nm) if compared regardless of level of section (p = 0.008). Posterior cruciate ligaments with moderate or marked degenerative change by light microscopy also showed a lower percentage of collagen occupancy than posterior cruciate ligaments with less degenerative change (p = 0.019).

TABLE 4: Electron Microscopy of Posterior Cruciate Ligament: Collagen Diameter

TABLE 5: Electron Microscopy of Posterior Cruciate Ligament: Percentage of Collagen Occupancy

When the correlation between clinical information and data from electron microscopy was tested in the Osteoarthritis Group, the authors found some changes related to the age of the patient. The mean collagen diameter at the midportion and tibial side of the posterior cruciate ligament specimens from the patients older than 60 years were 75.1 nm and 81 nm respectively, compared with 96.2 nm and 95.7 nm for the specimens from the patients younger than 60 years (p = 0.027 and 0.025, respectively). The mean collagen occupancy at the tibial side for the posterior cruciate ligaments from patients older than 60 years was 56% compared with 66.2% for the posterior cruciate ligaments from patients younger than 60 years (p = 0.047). The mean knee and patient function scores of the Knee Society clinical rating system were 44.7 points (range, 18–75 points) and 48.9 points (range, 0–90 points), respectively for the Osteoarthritis Group. Knee Society scores were not associated with collagen degeneration of the posterior cruciate ligament as determined by electron microscopy.

In assessing knee alignment using radiographs taken with the patient in the standing position, a significant correlation was not found between tibiofemoral alignment and mean collagen diameter (p = 0.98) or percentage of collagen occupancy (p = 0.54). The mean femoral shaft-transcondylar and tibial plateau-tibial shaft angles were 95° and 84°, respectively. Neither femoral shaft-transcondylar nor tibial plateau-tibial shaft angle was associated with variations of collagen diameter or percentage of collagen occupancy.

In assessing radiographic degeneration and electron microscopic findings, the authors did observe a relationship. Degenerative changes in four of 26 knees were interpreted radiographically as Grade 2, in 12 knees as Grade 3, in nine knees as Grade 4, and in one knee as Grade 5 as determined by the Ahlbäck grading scale. ²⁷ Specimens from more degenerative knees (greater than Grade 3) had less collagen diameter at all section levels than those from the knees of Grade 2 (p = 0.024 at the femoral side, p = 0.001 at the midportion, and p = 0.040 at the tibial side).

DISCUSSION

Total knee arthroplasty has been a standard operative treatment for osteoarthritic knees and has shown excellent long-term results in many studies. ^{3,5,7,9,12,14,18,26,28,29,31,32} However, it still is difficult to reproduce biomechanical properties of the knee perfectly, which in some cases can contribute to implant failure and revision arthroplasty. Many avenues of development have been used in the process of improving implant design, biomaterials, or operative techniques for total knee arthroplasty. For implant design approaches, controversies concerning posterior stabilization and posterior cruciate ligament function persist. ^{7,13,14,16,20,24,26,29,32} Previous studies have shown that clinical results for posterior cruciate ligament-sacrificing implants with posterior-stabilizing mechanisms perform comparably to posterior cruciate ligament-retaining implants. ^{5,9,12,14,26,32,33} Alexiades et al ¹ showed that 85% of posterior cruciate ligament specimens from osteoarthritic knees had degenerative changes, and that histologic findings were associated with age, gender, and knee deformity. Kleinbart et al ¹⁹ concluded that the histologic degenerative changes they identified in the posterior cruciate ligaments from osteoarthritic knees were different from age-related changes of comparable knees from cadavers. With the number of specimens that were available, the current authors found that the posterior cruciate ligaments from osteoarthritic knees had significantly greater deterioration than posterior cruciate ligaments from the knees of cadavers (light microscopy), although subjects in the Control Group were older than the subjects in the Osteoarthritis Group (p = 0.004). However, a significant correlation between the histologic grade and the clinical information (age, gender, functional score, patient functional score, radiographic grade for degenerative change, or knee alignment) was not detected in the Osteoarthritis Group.

In previous studies of posterior cruciate ligaments retrieved from osteoarthritic knees, histologic appearances were assessed only by light microscopy. ^1,19 The authors think that the current study is unique in terms of using electron microscopy to assess the morphologic properties of posterior cruciate ligaments from osteoarthritic knees. Investigation of the collagen fibrils by means of electron microscopy has documented that the biomechanical properties of ligaments may depend on the distribution and the diameter of collagen fibrils. ^6,23,35 The current findings suggest that the mean collagen fibril diameter for the Osteoarthritis Group was less than that for the cadavers. The percentage of collagen occupancy for the Osteoarthritis Group was smaller than that of the Control Group at femoral and tibial junctions although not it was not uniform throughout the ligament, but with the number of specimens available, these differences did not reach statistical significance. Histologic grades by light microscopy were significantly associated with the mean collagen diameter and percentage of collage occupancy in the Osteoarthritis Group (p = 0.008 and p = 0.019, respectively). These findings suggest that degenerative changes in the osteoarthritic knee may affect the collagen properties in the posterior cruciate ligament, and confirm that the histologic grade determined by light microscopy is significantly associated with reduced mean collagen diameter and a smaller percentage of collagen occupancy for osteoarthritic knees as quantified by electron microscopy and histomorphometry. It still is unclear whether the extent of mean collagen diameter or percentage of collagen occupancy is correlated with mechanical properties of the posterior cruciate ligaments from osteoarthritic knees, or whether the collagen properties of posterior cruciate ligaments from osteoarthritic knees are sufficient to function normally. Mechanical testing of retrieved posterior cruciate ligaments from osteoarthritic knees would be desirable. Although the mean collagen diameter and the percentage of collagen occupancy were correlated with age and radiographic grade of degenerative change in the current study, other clinical factors were not. It is unlikely that clinical parameters can predict the histologic status of the collagen of the posterior cruciate ligament in the osteoarthritic knee.

A weakness in the current study is the small number of specimens in the Control Group. Nevertheless, the authors think that the current study identifies an additional aspect of the histologic appearance of the posterior cruciate ligament from osteoarthritic knees.

Posterior cruciate ligaments from osteoarthritic knees showed degenerative changes as determined by light microscopy and electron microscopy. The authors were unable to identify clinical factors that could predict the histologic status of the collagen of posterior cruciate ligaments in osteoarthritic knees which may contribute to unexpected ligament failure or dysfunction after posterior cruciate ligament-retaining total knee arthroplasty. Based on the current results, previous studies showing excellent clinical outcome of posterior cruciate ligament-substituting implants and the authors’ clinical experiences, the authors think that posterior cruciate ligament-substituting total knee arthroplasty for osteoarthritic knees is preferable for most knees, and likely to prove most predictable for most surgeons.