Update on the Vitek partial and total temporomandibular joint systems

LINICAL UPDATE J Oral Maxillolac Surg 51:408-415,1993

Update on the Vitek Partial and Total Temporomandibular Joint Systems JOHN N. KENT, DDS,· MICHAEL S. BLOCK, DMD,t JEFF HALPERN, DDS,:f: AND MARK G. FONTENOT, DDS, MENG§ A retrospective recall study was done on 262 VK I (N = 138) and VK II (N = 124) (Vitek, Inc, Houston, TX) partial and total temporomandibular joints placed between 1982 and 1990. The cumulative success rate of VK I total joints observed for up to 10 years was approximately 20%, whereas the success rate of VK II total joints observed up to 6 years was approximately 80%. At the 5- to 6-year interval for each, these rates were 44% and 79%, respectively. Wear of the Teflon fluorinated ethylene propylene polymer surface was the primary reason for VK I failure; there was no material failure of the VK II prostheses. Slightly better pain relief, increase in interincisal opening, improvement in diet, and greater overall satisfaction were noted with the use of VK II. A highly significant improvement in success data was found if no surgery had been performed before either VK I or VK II total joint placement. Rib grafts were not particularly helpful after removal of total joint prostheses, particularly if the patient had a history of multiple surgeries . Total temporomandibular joint surgery must be reserved for patients in whom alternative surgical methods have failed or are no longer indicated. All total joint implants, particularly the VK I, must be observed closely with clinical examination and imaging and removed at the earliest sign of material failure.

From 1982 to 1990, Vitek, Inc, Houston, TX, manufactured two total temporomandibular joint (TMJ) systems (VK I, /982-1986; VK II, 1986-1990). The VK I fossa prosthesis had a superior surface ofProplast II (Vitek, Inc), which was adapted to the fossa and zygomatic arch, and Teflon fluorinated ethylene propylene (FEP) as an inferior articulating surface. It was used as a partial joint against a natural condyle or as a total joint with a metal condyle. The condyle had an anatomic articulating head made of chrome-cobaltmolybdenum with a long ramus shank, 45 to 53 mm, secured to the ramus. The VK II fossa had a superior surface of Proplast hydroxylapatite and an inferior surface of ultra-high-molecular-weight polyethylene

(UHMWPE) and the condyle was modified slightly with a head more laterally extended (5 mm). Both condylar prostheses were stabilized to the ramus with selftapping screws and a bolt. In '1986, we reported on our experience with 53 VK I fossae as a partial joint and 110 fossae with metal condyles as a total joint (l37 cases treated in New Orleans and 55 in Houston).' The cumulative success rate through 36 to 48 months of follow-up was more than 90%. The VK I prosthesis was discontinued in 1986 because of surface wear of the Teflon FEP polymer. Continued follow-up results of 182 joints (126 VK I and 56 VK II) reconstructed only in New Orleans up to 1988 varied as to prosthesis type , failure reason, and length of follow-up," VK I partial and total joint procedures had cumulative success rates of 42.4% and 57.95 %, respectively, with follow-ups to 5 or 6 years. VK II partial and total joint procedures had cumulative success rates of 100% and 88.95%, respectively, but follow-up was less than 3 years . The increased success was due to the improved wear characteristic of the UHMWPE fossa. Approximately 50% of the VK I failures were due to articular surface wear of the Teflon FEP polymer, but there were no VK II failures due to UHMWPE wear. Because of the decrease in successful

Received from the Department ofOral and Maxillofacial Surgery, LSU Medical Center, New Orleans, LA. • Boyd Professor and Head . t Associate Professor. Resident. § Dentist Scientist Award Cand idate. Address correspondence and reprint requests to Dr Kent: Department of Oral and Maxillofacial Surgery, LSU Medical Center, New Orleans, LA 70119.

*

© 1993 American Association of Oral and Maxillofacial Surgeons 0278-2391/93/5104.0013$3.00/0

408

409

KENT ET AL

Table 1.

Outcome of VK I FEP Fossae Through January 1992

Years

No. Entering Interval

No. of Failures

42 36 34 30 23 18 IO 4

6 I 4 7 5 6

0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 0-9

No. Lost to Follow-up or Deceas ed

o

o o o o o o o

6

3

42 35 34 30 23 18 IO 4

I

o I 36

42

No. Available for Evaluation

3 41

I

results between the 1986 and 1988 studies, another recall study was undertaken. This report discusses these results and total TMJ development.

Materials and Methods All patients were recalled in January 1992 by one of us (J.H.) who had never been involved in any prosth esis placement. Because the patients resided throughout the United States and Canada, the recall involved either a clinical examination at Louisiana State University or a telephone interview. Data obtained by either examination or interview included demographic information, diagnosis before any TMJ surgery, .diagnosis on admission for VK prosthesis placement, prosthesis type, length of follow-up, maximum interincisal opening (IIO) with pain, pain evaluation, diet, overall satisfaction, complications, reasons for prosthesis failure or removal, and subsequent reconstruction methods. Success criteria were decrease in pain, increase in function and opening, improve-

Partial Joints (n = 42)

Reason Remo ved

No.

Average Months to Failure

Material failure Biologic failure Patient

22 3 5

56 25 58

Infection

Unknown Total failures

0

6 36

- Significant difference.

VK II

Partial Joints

Total Joints Total Joints (n = 96) (n = 6) (n = 118)

No.

249 IO 4

II 58

Average

Average

Months to Failure

5428213 at 3 mo I at 2 yr

No.

No.

Months to Failure

0 0 0

0128

0 19 17

85.71 83.26 73.47 56.33 44.08 29.39 11.76 11.76 7.84 7.84

Table 3.

Results FEP Partial

FEP Total

UHMWPE Total

Before prosthesis After prosthesis Change DietBefore prosthesis After prosthesis Chan ge Paint Before prosthesis After prosthesis Change

25.25 30.67 +5.42

15.64 22.06 +6.42

21.74 29.54 +7.80

2.II 2.75 +0.64

1.51 2.40 +0.89

1.94 2.68 +0.74

9.64 8:6 -1.Q4

7.35 6.16 -1.19

8.40 5.81 -2.59

Overall effect]

5.09

3.93

5.32

21

0

0 0

85.71 97 .14 88 .24 76 .67 78 .26 66.67 40 .00 100.00 66 .67 14.34

110

VK II VKI

Cumulative Success Rate (%)

ment in occlusion and facial appearance, stability of. the prosthesis, and absence of radiographic evidence of soft or hard tissue pathology. In those cases in which the prosthesis had been removed, the primary reason for failure or removal was noted at our surgery or at surgery by others and was categorized as material, biological, patient, infection, or unknown failure. Material failures were defined as observable material wear or significant alterations of the VK fossa prosthesis even when fibrous or bony ankylosis was present. Biological failures were those that resulted in fibrous or bony ankylosis with ectopic bone formation but with no material wear. Patient failures were defined as cases requiring removal of the prostheses because of chronic pain unrelated to material wear or ankylosis. Psychosocial influences, recognized as a contributing problem with all failures, were not considered a primary reason for failure because of difficulties in objectively measuring them.

Table 2. VK Removal or Failure VKI

Int erval Success Rate (%)

0 21

- Diet was measured using the following criteria : I, liquid; 2, soft; 3, restricted ; and 4, unrestricted. t Pain was measured on a scale of I to IO with I rated least painful and 10 rated most painful. The overall effect of lID, diet, and pain was rated according to the following criteria: 1, much worse; 4, unchanged; 7, much better.

*

410 Table 4.

TMJ RECONSTRUCTION WITH VITEK TOTAL JOINT

Replacement for Failed Prosthesis 36 VK I Partial Joints Removed

58 VKI Total Joints Removed

21 VK II Total Joints Removed

IO 9 6

25 27

15

II

6

Rib Total joint (VK II) Temporalis muscle Other

o

6

o o

Results

Recall data were obtained on 262 joints in 137 patients; data for 14 joints were lost to follow-up or the patients were deceased. Average age was 37.2 years (range, 9 to 81 years) at the time of prosthesis placement. Only one patient was not an adult. Several patients had more than one partial or total joint procedure. Two hundred seventeen joints had had 840 previous surgeries before partial or total joint surgery. The average number of previous surgeries per joint was 3.86 (range, I to 15). Forty-fivejoints had not been previously operated on. More than 90% of the VK prostheses had been placed by one surgeon (J.K.) and nearly 90% of the surgeries before VK prosthesis placement occurred outside of our institution.

VKI Thirty-six ofthe 42 VK I partial joint prostheses had been removed, resulting in a 14.34% interval (0 to 9 years) success rate, or a cumulative success rate of 7.84% through the 8- to 9-year interval (Table 1). The most common reason for removal was material failure ofthe Teflon FEP articular surface with corresponding condylar degeneration (n = 22) (Table 2). The average gain in 110 was 5.42 mm, the average pain relief (measured on a lO-point scale) was 1.04, but the diet texture improvement was only 0.64 points (Table 3). The most Table 5.

common reoperation involved either a rib graft (n = 10), VK II total joint (n = 9), or a temporalis muscle flap (n = 6) (Table 4). Fifty-eight of the 96 VK I total joint reconstructions had been removed, resulting in a 39.68% interval (0 to 9 years) success rate or a cumulative success rate of 19.83%through the 8- to 9-year interval (Table 5). Material failure of the Teflon FEP articular surface (n = 24) was the most common reason for revision surgery. Biological failures, or ankylosis, occurred in 9 patients; there were 10 patient-type failures associated with unexplained pain, but no material wear or ankylosis. Eleven total joints were removed outside of our institution and the reason for failure was unknown to us (Table 2). The average gain in 110 for the VK I total joint patients was 6.42 mm. Pain reliefwas l.I9 points, and improvement in diet texture (measured on a fourpoint scale) was 0.89 points (Table 3). After removal of the prosthesis, these joints were most often reconstructed with VK II total joints (n = 27) or ribs (n = 25) (Table 4).

VKII Twenty-one of the 118 VK 11 total joints had been removed, resulting in a 82.2% interval (0 to 9 years) success rate and a 79.3% cumulative success rate through the 5- to 6-year interval (Table 6). The most common reasons for removal were biological or rean- , kylosis (n = 12) or patient-related (n = 8). To date, no VK 11 total joints have been removed because of material failure (Table 2). The average increase in 110 was 7.8 mm, the average pain relief was 2.59 points, and the improvement in diet texture was 0.74 points (Table 3). After removal of the prosthesis, rib grafting was the most common method of reconstruction (Table 4). Only sixjoints received a VK II fossa against a natural condyle; none have been removed. A comparison of VK I prostheses with the Teflon FEP fossa and VK II prostheses with the UHMWPE fossathat were followed up 10 and 6 years, respectively,

Outcome of VK I FEP Total Joints Through January 1992

Years

No. Entering Interval

No. of Failures

0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 0-9

96 86 70 57 44 32 24 15 3 96

IO IO 9 II 9 4 2 2 I 58

No. Lost to Follow-up or Deceased

No. Available for Evaluation

Interval Success Rate (%)

Cumulative Success Rate (%)

0 4 4 0 3 0 0 0 0 II

96 82 66 57 41 32 24 15 3 85

89.58 87.8 86.36 80.7 78.05 87.5 91.67 86.67 66.67 39.68

89.58 78.65 67.92 54.81 42.78 37.43 34.31 29.74 19.83 19.83

411

KENT ET AL

Table 6.

Outcome of VK II UHMWPE Total Joints Through January 1992

Years

No. Entering Interval

No. of Failures

No. Lost to Follow-up or Deceased

No. Available for Evaluation

Interval Success Rate (%)

Cumulative Success Rate (%)

0-1 1-2 2-3 3-4 4-5 5-6 0-6

118 107 85 51 20 3 118

10 3 6 2 0 0 21

I 0 I 0 0 0 2

117 107 84 51 20 3 116

91.45 97.2 92.86 96.08 100 100 82.2

91.45 88.89 82.54 79.3 79.3 79.3 79.3

showed significant differences (Table 7). After 5 years, the VK II total joint system has a significantly better (P < .0 I, Student's t test) cumulative success rate (79.3%) than does the VK I total joint (42.78%) (Tables 5,6). The average success rates of VK I partial and total joints placed in patients with multiple operations were 0% and 25.7%, respectively. If, however, no previous procedures had been performed, these success rates increased dramatically to 75% and 76.9%, respectively. The same data on VK II partial and total joints, however, showed a higher degree of success (100% and 80.7%, respectively), whether or not the patient had previous surgery. Ifno previous surgery was performed, these success rate.. were both 100% (Table 8). Highly significant differences were noted between failure type and prosthesis type. The time to failure of a VK I total joint because of material wear was nearly double the time to failure because of ankylosis or pain (Table 2). Patient type failures were related to chronic pain only, presumably secondary to previous surgeries and soft tissue reaction to silicone rubber (Silastic; Dow Corning, Midland, MI) or Teflon FEP and Proplast particles or to psychological problems. In this group, the 110 and occlusion were satisfactory, and reconstruction with rib grafts or other types of total joints generally did not improve their status. The overall effect reported in Table 3 is based on a combination of increased opening, decreased pain, and

Table 7.

improved diet as determined by the patient. Generally, there was a trend for all patients with VK II prostheses to do better than those with than VK I prostheses in all of these categories. However, the clinical improvement in pain, 110, and diet was minimal for the patient with multiple operations. The overall effect after revision surgery with rib grafts for failed VKjoints varied (Table 9). When rib grafts (n = 24) were placed after VK I joint removal, 50% of patients indicated they were better, 42% said there was no change, and 8% said they were worse. When rib grafts (n = 18) were placed after VK II joint removal, 17% said they were better, 39% said there was no change, and 44% said they were worse. We believe the greater overall improvement after VK I removal surgery in comparison with that after VK II removal was because of relief of the pain associated with Teflon FEP wear debris. In most joints, removal of a VK II and reconstruction with a rib graft did not improve the status. The diagnostic profile of the patient who had undergone the total joint procedure most commonly began with an initial diagnosis of degenerative joint disease or internal derangement (Table 10). The admitting diagnosis for those having total joint reconstruction, however, was predominantly ankylosis associated with pain and condylar resorption. The most common surgeries before prosthesis placement were discectomy with interpositional alloplastic implants, condylar shave, and disc repairs (Table 10).

VK System Evaluation (VK I Placed 1982 to 1986; VK II Placed 1986 to 1990)

Prosthesis

Years Followed

No. Entering Interval

No. of Failures

No. Lost to Follow-up or Deceased

No. Available for Evaluation

Interval Success Rate (%)

Cumulative Success Rate (%)

VK I partial VK II partial VI I total VK II total

10 6 10 6

42 6 96 118

36 0 58 21

1 0 11 2

41 6 85 116

14.34 100 39.68 82.2

7.84 100 19.83 79.3

412

TMJ RECONSTRUCfION WITH VITEK TOTAL JOINT

Table 8.

Average Success Rate (%)

VKI Partial History ofprevious surgery (217 joints) No history ofprevious surgery (45 joints)

0 75

VKI Total 25.7 76.9

VK II Partial 100 100

VK II Total 80.9 100

Discussion

SELECfION OF MATERIALS AND TESTING FOR TOTAL JOINT TMJ prosthesis design is a delicate interaction between engineering considerations and principles, surgical technique and requirements, functional demand, anatomic boundary limitations, and biocompatibility. Some of the mechanical and material properties of polymers that must be considered in the design and fabrication ofTMJ devicesare the stiffness, yield stress, creep resistance, and fatigue strength of the polymer. For example, the static load-carrying capacity of any TMJ device is governed by the magnitude of the compressive functional load to be supported. The bearing capacity (mean surface pressure) is determined by dividing the estimated applied compressive functional load by the projected support area of the condyle/fossa, For UHMWPE, the allowable mean surface pressure ranges from 10 to 13 MPa, whereas for fluorocarbons, such as polytetrafluorethylene (PTFE) and FEP, it is on the order of2 MPa. 3 Criteria for selecting Teflon FEP as an alloplasticbearing surfacewere described by Homsy in 1986 when Kent et al published their initial clinical experiences with the VK I Teflon FEP fossa prosthesis. 1 Homsy selectedTeflon FEP polymer because of its intrinsically low coefficient of friction and, in part, on the basis of wear data he accumulated while investigating the rheological properties of sodium carboxymethylcellulose," Wear rates were reported in 1986for Teflon FEP, polyoxymethylene, polymethylmethacrylate (PMMA), Co-Cr-Mo, but not for UHMWPE. 1 The results indicated that Teflon FEP polymer (0.018 mmjIOO,OOO cycles)and polyoxymethylene (0.022 mmj100,000 cycles) had similar penetrative wear rates, with PMMA having a lessdesirable rate, and Co-Cr-Mo undergoing galling.' In 1986, Homsy stated that his 1973 testing stress of 12 kg/ern? corresponded to a 12-kgtotal load across the TMJ, assuming that the effective articular interface between the condyle and fossawas about I em".' Citing from the dental literature, he formulated the following biomechanical scenario: normal bite force in the adult was reported to be 21 kgf,5 resulting in a singlecondylar reaction force of 75% to 80% of the biting force,"

Table 9. Overall Effect Performance of Rib Grafts Placed After Removal of VK Prosthesis

Results Prosthesis Removal

No. Better (%)

No. Same(%)

No. Wo=(%)

VK I (n = 24) VK II (n = 18)

12 (50) 3 (17)

10 (42) 7 (39)

8 (44)

2 (8)

Homsy concluded that the maximum force through a single condyle, corresponding to a normal bite force, was 12.6 kg, thus correlating well with the test load he used in evaluating Teflon FEP polymer approximately 13 years earlier," Using a TMJ simulator, Fontenot and Kent recently concluded that average linear penetrative wear rates for the Teflon FEP laminated to Proplast II in the form of the Proplast TMJ interpositional implant (PTlPI) and the VK I Teflon FEP fossa prosthesis were approximately 3.0 mmj 100,000 cycles and l.l mm/ 100,000cycles,respectively,"Differenceswere approximately 100-fold when comparing linear penetrative wear rates from this study with those reported by Homsy et al in 1973 report." Specifically, Homsy et al tested only the Teflon FEP polymer, in contrast to Fontenot and Kent, who tested the end-use device, ie, the VK I Teflon FEP fossa prosthesis. In addition, material testing apparatus used by Homsy et al applied a Table 10. Profile of a Patient With Total Joint Reconstruction

Diagnosis before any surgery DJD (36.2%)

Internal derangement (31.9%) Trauma (19.92%) Rheumatoid arthritis (14.2%) Ankylosis (2.8%) Idiopathic condylar resorption (2.8%) Other (2.2%) Admitting diagnosis for partial and total joints Fibrous ankylosis (33.8%) Condylectomy (17.7%) Bony ankylosis (14.6%) DJD (11.9%)

Prosthesis erosion (6.9%) Other (15.1%) No. ofprevious surgeries Discectomy with silicone rubber (129) Condylectomy (condylar shave) (118) Discectomy with Proplast Tenon (79) Disc repair (63) Coronoidectomy (44) Osteotomy (37) Rib graft (20)

History ofmore than three previous surgeries, disc implants, unbearable pain, hypomobility; patients most likely tofail History ofrheumatoid arthritis, trauma, tumor; patients most likely tosucceed Abbreviation: DJD, degenerative joint disease.

413

KENT ET AL

stress of 1.17 MPa, which was below the 2-MPa coldflow limit (permissible mean contact stress allowed before cold flow ensues) of Teflon FEP, whereas the study of Fontenot and Kent approached and exceeded this limit. Before the development ofthe VK system, Charnley and Kamanger had developed an alloplastic hip using Fluon PTFE, a fluorocarbon polymer similar in material properties to Teflon FEP.8 Critical in the development of an alloplastic hip was information relating to the in situ service conditions ofthe hip, such as load conditions and device design. Because this information was not known at that time, Charnley found it necessary to evaluate new device designs by clinical experimentation. In 1958, he implanted an alloplastic hip replacement using stainless steel as the alloplastic stem and femoral head articulating against an acetabular cup fabricated from Fluon PTFE. 8 He selected Fluon PTFE because of its commercial reputation as a material with a low coefficient of friction and chemical inertness," After 300 total hip procedures, the Fluon PTFE acetabular cup was abandoned because of significant clinical problems with pain, limited range of motion, and "teflonomas" resulting from Fluon PTFE particulate wear debris, and Charnley was compelled to remove and revise these hip prostheses.v'? In 1969, Charnley reported on the wear rates of 100 stainless steel/Fluon PTFE total hip prostheses that had been removed from patients," Revision surgeries were necessary because of failed or failing Fluon PTFE acetabular cups generating massive amounts of wear debris and resulting in an undesirable tissue reaction, pain, and limited range of motion. In many of the cases, the Fluon PTFE acetabular components had worn through. Wear testing ofthe UHMWPE in a VK II fossa using a TMJ simulator in 1986 showed much improved results compared with Teflon FEP. II A uniform and constant force of 12 kg/ern? was applied at the condyle/ fossa interface by way of a preloaded spring. The linkage was designed to articulate the condyle against the fossa to a degree that corresponded to approximately 35 to 40 mm of 110. The testing was conducted in 37°C normal saline at 32 cycles/min. Results indicated that wear of the UHMWPE articulating surface was at least eight times better than the Teflon FEP in articulation against a metallic condyle. II Although separated by 20 years, Charnley and Homsy selected the fluorocarbon polymers Fluon PTFE and Teflon FEP, respectively, primarily because of their low coefficients of friction. In both cases, these polymers incurred loading that exceeded the elastic limits of the materials and resulted in mechanical failure and explantation. Ironically, just as Charnley implanted PTFE acetabular components in 1958 8 only to replace it with UHMWPE in 1962 8 because of its

better wear performance, the Teflon FEP fossa prosthesis was introduced by Homsy in 1982 only to be replaced in 1986 by UHMWPE because of its better wear performance. It is unfortunate that the extent of functional loading in the TMJ was not appreciated or its effect adequately tested during the early development of the VK system.

Complications and Follow-up Complications with total TMJ prostheses include hematoma, excessive bleeding, motor or sensory damage, ear canal injury, malocclusion, continued pain, and recurrent hypomobility secondary to dense fibrous encapsulation of the prosthesis.'? Failure to perform aggressive excision of dense fibrous tissue and bonebetween the skull base and medial aspect of condyle, sigmoid notch, and coronoid process commonly leads to reankylosis and procedural failure. Vascular injury and infection can occur because of a previously operated, scarred tissue bed. Errors in technique and unique problems in patients with multiple operations lead to poor results. Functional joint anatomy is essential, and improper relationship of artificial joint parts, too rapid or excessive joint loading, or inadequate postoperative physical therapy and/or follow-up can contribute to the failure of the total joint procedure. A normal but limited macrophage response with giant cells is a common finding with the use of any TMJ polymer material.'! However, an extensive and persistent macrophage response with a large aggregation of the histiocytes (foreign body giant cells) and granulomatous tissues indicates significant particulate wear debris from the polymer and requires removal of the device. Since 1990, US Federal Drug Administration (FDA) bulletins have advised health professionals and patients about the serious risk of the foreign body giant cell reaction from PTIPI used as a disc replacement causing bone resorption, chronic pain, hypomobility, and malocclusion. Symptoms and signs are most commonly due to polymer wear debris on both the condyle and fossa sides ofthe implants. Most ofthese same findings have also occurred with the VK I fossa prosthesis used against either a natural condyle or metal condyle. Fossa bone resorption, however, was not observed on removal of most VK I prostheses unless wear through the Teflon FEP surface into Proplast, or mobility of the fossa prosthesis, had occurred. In these cases, resorption through the fossa floor exposing dura mater required grafting along with subsequent reconstruction (other total joints or rib grafts) (J.N. Kent, unpublished data). All VK I joints should be followed closely and removed at the earliest sign of problems because we believe all may eventually fail. MRI is useful to detect a granulomatous tissue reaction. It is known that particulate polymer debris from failed interpositional joint im-

414 plants made of Proplast-Teflon or silicone rubber contribute to chronic pain. 3•14 If, however, polymer debris is retained in the peripheral tissues outside the joint, chronic pain may persist after removal of these implants and placement oftotaljoint prostheses. Although the FDA recommends cr for postoperative evaluation, its use is not adequate for metallic total joints. Tomograms and cephalograms are most helpful and can be compared with immediate baseline films. To date we have observed no significant wear of the UHMWPE surface, nor have we seen fossa resorption from Prop last, through 5 to 6 years of follow-up in either removed prostheses or reentry procedures to release or remove fibrous tissue and bone causing hypomobility and pain. Removal of a VK II total joint because of problems associated with the PTIPI and VK I, both with Teflon FEP as an articulating surface, may not be indicated in the absence ofankylosis or material failure. Our experience with these "biological or patient" failures with unrelenting chronic pain also suggests that VK II total joint removal and replacement with other prostheses or rib grafts has a limited chance ofcausing improvement. Therefore, when there are no wear signs of pathology noted on reentry, the prostheses are retained and only the restricting hard and soft tissues are removed. Patients should be monitored yearly and imaged, if necessary, for comparison with immediate postoperative films. The follow-up data on the VK system and other total joint systems indicate that several variables affect partial or total joint reconstruction. Previous surgery on the TMJ is directly associated with increased risk for failure. In our series of patients, condylar degeneration and excessivefibrosis from multiple procedures, and particularly chronic pain, played major roles in determining the success of total TMJ replacement. Our data further suggest that patients who have had sound surgical procedures of any type accompanied by chronic pain should be told that complete pain relief is not realistic. Surgical revision after total joint prosthesis placement as well as soft and hard tissue grafts, should be resisted in the face of chronic pain that is associated with normal radiographic findings, acceptable mandibular function under general anesthesia, and satisfactory occlusion. Exploration, avoiding major revision surgery, may be justified to rule out suspicion of implant wear and tissue reaction-a difficult problem if patients have had previous failed implants used for disc replacement. In these patients, the condyle often must be removed to adequately inspect and/or excise excessive scar. Successful total TMJ replacement is assessed by relief of pain, and re-establishment of masticatory function and esthetics. Patients who have had less or no previous surgery, such as loss of a condyle from trauma or advanced rheumatoid disease, have better expectations

TMJ RECONSTRUCfION WITH VITEK TOTAL JOINT

than the chronic pain patient with a history ofmultiple joint procedures. The disadvantages and limitations of any total joint replacement with alloplastic materials also includes the lack of options if the prosthesis fails and there is need for replacement prosthesis. In our experience, replacement with another total joint prosthesis (Synthes, Christensen, Techmedica, or another VK II) has been as successful as rib grafts, but unfortunately results of most ofthese cases are still less than desirable. Selection of materials and proper design are key parameters with total joints, and there also must be in vitro laboratory testing on these materials. However, such testing has been inadequate or perhaps not performed in the past. Therefore, safety and efficacy, as well as in vivo life service predictions, are estimates at best. We do not know if current materials, namely polymers, will provide adequate service life because compelling in vivo and in vitro data are not available. The FDA has warned that VK II total joints may present risks similar to those with PTIPI or VK I total joints because Proplast may not perform well in loadbearing joints. Our observations at surgery indicate that the problem is with the Teflon FEP surface and not the Prop last, which serves as a filler. An endless list of surgical procedures for TMJ reconstruction has been developed by oral and maxillofacial surgeons. However, our review ofthe literature reveals supporting claims for some procedures without specific data, success/failure criteria, and identification ofdiagnosis or conditions leading to failure. Frequently the procedure failed because of pain and/or ankylosis, but what were the underlying factors? If these joints did fail, why and how were they fixed? What is the incidence of reoperation and its degree of success/failure? Should, in fact, certain joints not be continually reconstructed when the success rate ofsubsequent procedures is no better? Most of the literature that we reviewed recommends procedures on the basis of less than 2 years of mean follow-up. Although such new information is promising, longer follow-up is needed. In most reported TMJ surgery series, only interval or average success rates with a mean follow-up are reported. The interpretation of that data is frequently based on the case followed longest. We strongly recommend life table analysis (cumulative success rates), which is more valid to support long-term interpretation.'? Such data would be useful to the FDA, National Institutes of Health, and manufacturer ofdevices. Our profession should demand this form of data reporting for joint repairs and soft tissue and bone grafting procedures as well. Finally, a means of predicting success oftotal joint and other TMJ reconstructive procedures must be developed on the basis ofa description (signs and symptoms) of the TMJ deformity. They include a pain index, degree of myofascial pain dysfunction, ar-

415

ROBERT D. MARCIANI

ticular cartilage/cortex health of the condyle, function (110 and excursions) occlusal abnormalities, ramus vertical dimension changes, number and type of previous surgical procedures, assessment of postoperative management, and presence of psychological or mental disorders.

5. 6. 7. 8.

References

9.

I. Kent IN, Block MS, Homsy CA, et al: Experience with a polymer glenoid fossa prosthesis for partial or total TMJ reconstruction. J Oral Maxillofac Surg 44:520, 1986 2. Kent IN, Block MS: Five year follow-up of polymer glenoid fossa prosthesis for partial and total TMJ reconstruction. J Oral Maxillofac Surg 46:M 15, 1988 (abstr MIS) 3. Wright KWJ: Friction and wear of materials and joint replacement prostheses, in Williams OF (ed): Biocompatibility of Orthopedic Implants, vol 10 Boca Raton, FL, CRC Press, 1982, pp 141-196 4. Homsy CA, Stanley RF, King JW : Pseudosynovial fluids based on sodium carboxymethylcellulose, in Gabelnick HL , Litt M

10. II. 12.

13. 14.

(eds): Rheology ofBiological Systems . Sprin gfield. II.., Thomas, 1973. pp 278-298 Hylander WL: The human mandible: Lever or link? Am J Phys Anthropo143:227, 1975 Smith RJ: Mandibular biomechanics and temporomandibular joint function in primates. Am J Phys Anthropol49:34l, 1978 Fontenot MG, Kent IN: In vitro wear performance of proplast TMJ disc implants. J Oral Maxillofac Surg 50:133,1992 Charnley J, Kamangar A: The optimum size of prosthetic heads in relation to wear of plastic sockets in total replacement of the hip . Med Bioi Eng Comput 7:31, 1969 Eftekhar NS: The life and work of John Charnley. Clin Orthop 211:10-22 ,1986 Older J : A tribute to Sir John Charnley. Clin Orthop 211:23-29, 1986 Vitek Advertisement ofVK II prostheses, September 1986 Kent IN, Misiek OJ: Biomaterials for cranial, facial, mandibular, and TMJ reconstruction, in Fonseca R, Walker R, (eds): Oral and Maxillofacial Trauma. Philadelphia, PA, Saunders, Ch 29,2:781-1026, 1991 Kaplan EL, Meier P: Nonparametric estimation from incomplete' observations. J Am Stat Assoc 53:457-481,1958 Spagnoli 0, Kent J: Multicenter retro spective analysis of Proplast II-Teflon disc implants. Oral Surg Oral Med Oral Pathol 74: 411, 1992

J Oral Maxillolac Surg 51:415-416.1993

Discussion Update on the Vitek Partial and Total Temporomandibular Joint Systems Robert D. Marciani, DMD Veterans Administration and University ofKentucky, Lexington Kent et al have presented a retrospective clinical study of temporomandibular joint (TMJ) replacement that reflects a unique insight into the care of the mutilated joint. In many cases their experience represents "end of the line" care for patients who have had multiple previous surgeries, complex psychosocial conditions, and failure of alloplastic joint replacement material. Assessing the results of surgical manipulation of the TMJ is a complex undertaking that inevitably precludes compelling scientific methodology. Sophisticated facial pain investigators recognize the inadequacies of retrospecti ve studies (lack of controls, poor standardization ofevaluation mechanisms, and inadequate provision for a host of confounding variables) and wistfully design prospective TMJ research protocols. The results of this study are diminished in significance by the limitations inherently present in a retrospective review. The authors do not report how many patients received a clinical examination at Louisiana State University or were interviewed by telephone, how many were female, and what the criteria were for surgical success or failure. Techniques of clinical examination and subjective data gathering are not outlined; therefore, jaw mobility measurements and function data are inconclusive. Kent and et al l •2 have accepted the difficult challenge of analyzing the biomechanical forces of the TMJ. Biological, mechanical, parafunctionaI, and implant design factors come together in a variably responding model to diminish the op-

port unity for conclusive scientific findings. The concave configuration of the glenoid fossa, buttressed by the slope of the articular eminence, houses a convex condyle that has considerable rotational and translational mo vement when compared with the hip joint. Wear characteristics of ajoint prosthesis made of similar materials cannot be expected to produce the same outcome in dissimilar joints. Extrapolating the utility of a prosthetic, relatively nonmovable joint, placed in a geriatric patient is at best questionable when applied to the bilaterally active, functioning jaw joint in a young person. Consistent failure of the VK I partial and total joints is a clear indictment of the mechanical characteristics of fluorinated ethylene propylene. Laboratory tests conducted by Kent et al 3 have identified the poor wear characteristics of this material when used as a partial or total joint replacement or a TMJ interpositional implant. Through the 5- to 6-year interval, approximately 80% of the YK II total joints were still functioning. Laboratory tests of the ultrahigh-molecular-weight PE indicate much more favorable mechanical wear characteristics, which may account for the greater success of the YK II prosthesis. The average success rate of the VK II total joint, however, was reduced in the category of patients with multiple operations. Success of the YK II total joint for periods substantially longer than for the YK I prosthesis would be extremely beneficial to patients with severe joint destruction and associated dentofacial deformities. Ongoing clinical and radiographic observation of the YK II prosthesis, as suggested by the authors, is reasonable and encouraged. What are the characteristics of the patients who are likely to receive TMJ implants? Are the mechanical and biological needs ofthe TMJ comparable with those ofother body joints, allowing biomaterials and -principles of joint surgery to be universally applied? Are the indications for placement and