Abstract
Background: Most total elbow prostheses that are currently in use require cement for fixation of each component. We developed a new (type-5) prosthesis that does not need cement for fixation.Methods: The humeral component is made of cobalt-chromium alloy, and its stem is porous-coated with a plasma spray of titanium alloy. There are two options for the ulnar component: an all-polyethylene type and a metal-backed type with a porous-coated stem. Forty-three elbows in thirty-seven patients who had rheumatoid arthritis were treated with total elbow replacement arthroplasty with use of the type-5 prosthesis. The humeral component was implanted without cement in all elbows, whereas the ulnar component was implanted without cement in eleven elbows and was fixed with cement in the remaining thirty-two. The elbows were followed for an average of three years and ten months (range, two years and six months to five years and six months).Results: The clinical results were assessed according to a modification of the Mayo Clinic Performance Index. At the time of the latest follow-up, the overall result was excellent for six elbows, good for thirty-one, and fair for six. All elbows had been rated as poor before the operation. There was almost complete relief of pain in twenty-nine elbows and mild or occasional pain in the remaining fourteen. Flexion increased markedly, from an average of 104 degrees preoperatively to an average of 133 degrees postoperatively; this difference was highly significant (p < 0.001, Student t test). In contrast, extension (flexion contracture) worsened slightly, from an average of 38 degrees preoperatively to an average of 42 degrees postoperatively; this difference was significant (p < 0.05).There was one postoperative dislocation of the elbow, and ectopic bone formed in another, with recurrence of ankylosis. Both elbows had a reoperation, and a good result eventually was obtained. There were no instances of postoperative infection or neuropathy of the ulnar nerve.Radiographically, there were no radiolucent lines at the bone-metal interface of any of the humeral or ulnar stems that had been implanted without cement, suggesting solid fixation by osseointegration.Conclusions: The results of total elbow arthroplasty with use of this prosthesis appear promising. There was a high rate of relief of pain as well as of restoration of adequate function in patients in whom the elbow was severely affected by rheumatoid arthritis.
Currently, most total elbow arthroplasties are performed with use of acrylic cement for fixation of both components of the prosthesis. Although insertion without cement was tried in Japan by Inoue et al.9, who used a prosthesis made of alumina ceramic, solid fixation by osseointegration at the bone-ceramic interface was not achieved. Those authors ultimately abandoned this method and returned to using the conventional technique of fixation with cement. With the exception of our report12 in 1994, there have not been any other studies, to our knowledge, on fixation of total elbow prostheses without cement.
In our 1994 report12, we described the intermediate-term results, after an average follow-up of three years and one month, of total elbow arthroplasty with use of a nonconstrained, titanium-alloy (type-4) implant inserted without cement. Porous coating of the stem with titanium alloy was very effective in achieving osseointegration at the bone-metal interface. However, we also noted that tribological problems at the articulating surfaces between the titanium alloy and the high-density polyethylene sometimes led to profuse production not only of metal debris but also of polyethylene particles, resulting in marked osteolysis at the distal end of the humerus and subsequent breakage of the humeral component at the base of its stem.
We learned two important lessons from our earlier study12. First, we concluded that articulation of titanium alloy against polyethylene should be avoided, even in a non-weight-bearing joint such as the elbow. Second, we determined that a plasma-spray porous coating of titanium alloy is a very effective and reliable means of achieving secure fixation without cement, even in patients who have rheumatoid arthritis, in whom the presence of osteoporosis is inevitable and the use of acrylic cement generally is considered essential because of the osteoporosis.
In 1993, we discontinued use of the type-4 prosthesis and started to use a new (type-5) model in which the humeral component is designed to be inserted without cement. In this report, we present the intermediate-term clinical results of arthroplasty with use of this prosthesis.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
†Section of Orthopaedics, Sagamihara National Hospital, Sakuradai 18-1, Sagamihara City, Kanagawa Prefecture, Japan. E-mail address for Dr. Kudo: hiroshi23.kudo@nifty.ne.jp.
Background of the Type-5 Prosthesis
From April 1972 to March 1982, we had used a series of nonconstrained surface-replacement prostheses (types 1 and 2). The humeral component of both prostheses is made of stainless steel, and neither has an intramedullary stem. The ulnar component is made of ultra-high molecular weight polyethylene and has a short stem for intramedullary fixation. An interim report10 as well as a long-term follow-up study11 on the results of arthroplasty with use of these prostheses have been published. The main problem encountered with these prostheses was proximal subsidence of the humeral component, which was due in large part to the absence of the intramedullary stem. To address this problem, we developed the type-3 prosthesis, in 1980. The condylar portion of the humeral component of this prosthesis had almost the same shape as that of the type-2 prosthesis, but an intramedullary stem was added. The type-3 prosthesis was used, with satisfactory clinical results, until 1987; at an average of eight years postoperatively, only two of forty elbows had subsidence of the humeral component.
In 1988, in an attempt to develop an implant that could be inserted without cement, the humeral component was modified to type 4, which was made of titanium alloy with a porous-coated stem. However, the articulation of the titanium alloy against the high-density polyethylene led to tribological problems1,18, such as metallosis and a high rate of polyethylene wear. This led us to modify the humeral component to type 5, which consisted of cobalt-chromium alloy with one-half of the surface of the stem porous-coated with a plasma spray of titanium alloy. The base of the stem was modified slightly to reinforce this area, in the hope of preventing breakage. The ulnar component was the same as that of the type-4 prosthesis, except that an alternative ulnar component with metal backing and a porous-coated stem was designed chiefly for use without cement (Fig. 1).
Patients
During the three-year period from January 1993 through December 1995, forty-eight elbows in forty-two consecutive patients who had rheumatoid arthritis had a total elbow arthroplasty with use of the type-5 prosthesis. Two of the forty-eight elbows had sustained a fracture of the medial condyle of the humerus in an accident preoperatively, and one had sustained the same type of fracture because of technical failure during the operation. These three elbows were excluded from the study because the humeral component had to be inserted with cement in order to maintain the proper position of the fractured medial condyle. An additional two elbows had a revision because of breakage of the stem of the humeral component of a previously implanted type-4 prosthesis. Although these two elbows had successful exchange of the broken humeral component with a type-5 component without use of cement (only the ulnar component was fixed with cement), we also excluded these elbows from the study because the operations were not primary. The clinical as well as the radiographic results of these five elbows were all satisfactory at the time of the latest follow-up; thus, the exclusions did not introduce bias in this study.
After exclusion of these five elbows, forty-three elbows in thirty-seven patients were left for study. (Six patients had a bilateral procedure.) There were five male and thirty-two female patients, who ranged in age from twenty-seven to eighty years (average, fifty-five years). The oldest patient (Case 29) had had rheumatoid arthritis for seventeen years before the operation. All patients had severe rheumatoid arthritis, and all met the diagnostic criteria of the American Rheumatism Association2, which included (1) stiffness for at least one hour in the morning, (2) involvement of at least three joint areas (soft-tissue swelling or fluid observed by a physician), (3) arthritis of the hand or wrist joints (at least one swollen area in a wrist, metacarpophalangeal, or proximal interphalangeal joint), (4) symmetrical arthritis, (5) rheumatoid nodules, (6) serum rheumatoid factor, and (7) radiographic changes typical of rheumatoid arthritis as seen on posteroanterior radiographs of the hand and wrist. For the purpose of classification, a patient was considered to have rheumatoid arthritis if he or she met at least four of these seven criteria. (The first four criteria had to have been met for at least six weeks.)
The duration of follow-up ranged from two years and six months to five years and six months (average, three years and ten months). One man (Case 6) died of rheumatoid pulmonary complications of interstitial pneumonitis four years after the operation, but he was enrolled in the study because he had been followed until just before his death. There were no other deaths in the series.
Both the humeral and the ulnar component were implanted without cement in eleven elbows, whereas the humeral component was implanted without cement and the ulnar component was implanted with cement in the remaining thirty-two. Thus, the humeral component was implanted without cement in all forty-three elbows.
Of the thirty-two ulnar components that were implanted with cement, eight had a metal backing and twenty-four were all-polyethylene. Three elbows in two patients had had a synovectomy more than six years before the index operation.
Preoperative Status of the Elbows
In this series of forty-three elbows, roughly three categories of preoperative clinical status could be assigned. Six elbows with severe limitation of motion (fibrous ankylosis) had 20 degrees of motion or less. Fifteen elbows had painful stiffness; the joint had an arc of motion of 25 degrees or more, but the patient was unable to flex it more than 100 degrees. Twenty-two elbows had painful instability; the joint had a functional range of passive motion, but active motion was very limited because of severe instability as well as pain, and the patient was unable to bring the hand or fingers to the face or the head while holding a heavy object in the hand.
Radiographically, all elbows had advanced (grade-3 or 4) changes according to the classification system of Morrey and Adams14. In grade-3 involvement, there are alterations in the architecture of the joint, such as thinning of the olecranon process or resorption of the trochlea or the capitellum. In grade-4 involvement, there is gross destruction of the joint as a result of severe bone resorption, as often is seen in so-called mutilans deformity. Thirty-nine elbows had grade-3 involvement, and the remaining four had grade-4 involvement. No elbow had mild or moderate (grade-1 or 2) radiographic changes. Five elbows that were clinically classified as having severe limitation of motion (fibrous ankylosis) had grade-3 involvement without a radiographic finding of bone ankylosis.
Operative Technique
The basic operative technique was almost the same as that used in the previous series of type-1 and type-2 prostheses10,11. All elbows were operated on with use of Campbell's posterior approach4, in which a triangular or tongue-shaped flap with a distal base is raised in the triceps tendon. The radial head always is excised, and the ulnar collateral ligament, including the tight anterior bundle, is severed completely to allow dislocation and to gain satisfactory access to the interior of the joint.
Because the humeral component of the type-5 prosthesis, like that of the type-4 prosthesis, has a stem and because cement is not used for fixation, several additional operative steps (not included in the previous series) are performed. For example, at the time of insertion of the humeral component, bone chips, produced from the blocks of bone being resected during trimming of bone ends such as the trochlea, both condyles of the humerus, and the radial head, are packed tightly into the central segment of the inner space within the condylar portion of the humeral component; the humeral component then is inserted into the prepared bone end of the humerus without cement. If any empty spaces remain between the bone end and the inner surface of the condyle mold of the humeral component after insertion, bone chips are packed into those spaces as well. In exceptional cases in which a gross defect of bone stock is present on either side of the humeral condyle, a bone block obtained from the iliac wing is used as a strut graft to fill in the defect13. This procedure was done in four elbows in the current series.
The decision as to whether cement should be used for fixation of the ulnar component is very important and usually depends on the surgeon's assessment of the reliability of the initial fixation when a trial component is inserted into the prepared bed of bone in the ulna. If there is any doubt about the reliability of the initial fixation, fixation with cement should be considered. When cement is to be used for fixation of the ulnar component, we prefer the all-polyethylene component to the one with metal backing and a porous-coated stem; the all-polyethylene component is more economical, and data from our previous, long-term study (average duration of follow-up, nine years and six months) of type-1 and type-2 prostheses support its use11. In that series, the rate of loosening of the all-polyethylene ulnar component, which is similar in design to the ulnar component of the type-5 prosthesis, was two (5 percent) of thirty-seven, whereas the rate of subsidence of the humeral component was twenty-six (70 percent) of thirty-seven.
In order to prevent postoperative dislocation or residual instability, the incised ends of the triceps tendon and the incised margins of the dorsal fascial layer on the radial side of the olecranon are sutured meticulously. This is a key factor for the prevention of dislocation. In elbows that have been classified preoperatively as having either severe limitation of motion (fibrous ankylosis) or painful stiffness, the triceps tendon always is lengthened from ten to twenty millimeters in a V-Y fashion when the triceps-tendon flap is sutured back to the incised proximal stump of the tendon. This is done in order to gain a satisfactory range of flexion (more than 120 degrees). The ulnar nerve is released extensively, as far distally as possible beyond the cubital tunnel, during the first stage of the operation, by splitting the muscle of the flexor carpi ulnaris and also by releasing the nerve from the deep flexor pronator aponeurosis7; the aponeurosis prevents free mobilization of the nerve and might later constrict it because of postoperative edema.
In the final stage of the operation, the nerve is transferred anteriorly through the subcutaneous route. Darrach's procedure5 (resection of the distal end of the ulna) is performed occasionally in addition to the arthroplasty in patients in whom rotation of the forearm, particularly supination, is severely limited (to less than 10 degrees) despite the resection of the radial head during the arthroplasty.
Postoperatively, the elbow is immobilized in 60 or 90 degrees of flexion in a posterior splint for one week. If the blood circulation of the incised skin edge at the tip of the olecranon seems to be disturbed with the elbow in 90 degrees of flexion, the elbow should be immobilized in a more extended position (approximately 60 degrees of flexion).
Clinical Assessment
The clinical condition of each elbow before and after the operation was assessed according to a slight modification of the scoring system of the 1992 Mayo Clinic Performance Index14 (Table I). We found a few inconsistencies in the table of the original version of the Performance Index; for example, 20 points was assigned for an arc of motion of 100 degrees; 15 points, for an arc of 50 to 100 degrees; and 5 points, for an arc of 2 degrees. In addition, no points were assigned for moderate instability. Therefore, we slightly modified the original table, assigning 20 points for an arc of 100 degrees or more, 15 points for an arc of 50 to 99 degrees, and 5 points for an arc of 49 degrees or less. We also assigned 5 points for moderate instability, which seemed necessary for evaluation of the stability of nonconstrained prostheses. In a 1998 report from the Mayo Clinic8, the inconsistencies in the original version of the Performance Index had been resolved, and that later version of the system is almost the same as the one with our modifications. Unfortunately, this revised version was not available to us at the time of the latest follow-up of the elbows in this series.
Preoperatively, all forty-three elbows had been rated as poor (a score of 59 points or less), whereas at the latest assessment six elbows were rated as excellent (a score of 90 points or more), thirty-one were rated as good (a score of 75 to 89 points), and six were rated as fair (a score of 60 to 74 points) (Table II).
Preoperative pain was moderate (a score of 15 points) or severe (a score of 0 points) in thirty-nine elbows, and it was mild (a score of 30 points) in four. Postoperatively, fourteen elbows had mild or occasional pain, and twenty-nine had no pain (a score of 45 points).
The average angle of extension (flexion contracture) before the operation was 38 degrees, whereas postoperatively it was 42 degrees (Table III); this difference is significant (p < 0.05, Student t test). In contrast to this slight but significant worsening of the flexion contracture, flexion improved markedly, from an average of 104 degrees preoperatively to an average of 133 degrees postoperatively; this increase is highly significant (p < 0.001). Rotation of the forearm also improved after the operation; this improvement was greater for supination than for pronation. The average angle of pronation was 33 degrees preoperatively compared with 41 degrees postoperatively; this increase is significant (p < 0.01). The average angle of supination was 38 degrees preoperatively compared with 65 degrees postoperatively; this increase is highly significant (p < 0.001).
Eight elbows had a Darrach procedure. The average angle of supination in these elbows was 4 degrees preoperatively compared with 54 degrees postoperatively; this difference is highly significant (p < 0.001). The average angle of pronation was 9 degrees preoperatively compared with 13 degrees postoperatively; however, with the numbers available for study, we could not detect a significant difference (p < 0.19).
Postoperatively, one elbow (Case 2) had a decrease in stability from 10 to 5 points, thirty-five elbows had no change, and seven had an improvement. These seven elbows included the four that had had gross instability (0 points) before the operation; these four recovered functional stability (5 points) after the operation (Table III).
With regard to daily function of the upper limb, all elbows had had a score of 5 or 10 points before the operation, whereas after the operation all except one elbow (Case 22) had a score of 15 points or more (one elbow had a score of 25 points; seventeen, a score of 20 points; and twenty-four, a score of 15 points).
Radiographic Assessment
The radiographs of each elbow that were made at the latest follow-up visit were reviewed carefully, with special attention paid to three parameters: displacement (such as subsidence or migration) of each component, the presence or absence of radiolucent lines at the bone-cement and bone-metal interfaces, and bone resorption or osteolysis. Displacement of the components was assessed by comparing the radiographs made at one month after the operation with those made at the latest follow-up visit. The status of the bone-cement and bone-metal interfaces was assessed carefully on the radiographs made at the latest follow-up. We looked for findings suggestive of bone resorption or osteolysis on the latest lateral tangential radiographs, on which the amount of bone mass within the interior space of the condylar portion of the humeral component is well visualized. The bone in this region consists mainly of the bone stock made up of the grafted bone chips. In our previous report on arthroplasty with use of the type-4 prosthesis12, the earliest sign of bone resorption was a reduction of bone mass within that space.
Slight subsidence of the ulnar component was seen in only one elbow (Case 2); in the remainder of the elbows, no displacement of the ulnar component was evident. No displacement of the humeral component from its original position was observed in any elbow.
No elbow had a radiolucent line of any noticeable width (one millimeter or more) at the interface around the stem of the humeral component. In the eleven elbows in which the ulnar component had been implanted without cement, no radiolucent line was evident at the interface around the stem of that component. Of the thirty-two elbows in which the ulnar component had been fixed with cement, six had a thin radiolucent line of one millimeter or less at the bone-cement interface; however, the line was present only in the region of the trochlear notch of the ulna and did not extend distally in five elbows. In the remaining elbow (Case 2), it was present throughout the entire region of the bone-cement interface (Figs. 2-A, 2-B, 2-C and 2-D).
All forty-three elbows retained almost the full amount of bone mass within the interior space of the condylar portion of the humeral component after an average duration of follow-up of three years and ten months. The absence of bone resorption or the preservation of bone mass in the packed graft of bone chips was confirmed in all elbows on the lateral tangential radiographs.
In the four elbows (Cases 12, 15, 22, and 23) that had had a gross defect of bone stock on either side of the humeral condyle before the operation, a full-thickness block of bone obtained from the iliac wing was used as a strut graft to fill in the defect at the time of insertion of the humeral component13. In these four elbows, the size of the strut graft seemed well maintained (that is, resorption of the graft appeared minimum) on the radiograph made at the latest follow-up visit (Figs. 3-A, 3-B, 3-C and 3-D).
An unexpected finding, on the radiographs of three elbows (Cases 3, 24, and 25), was tilting or internal rotation of the ulnar component together with the whole bone of the ulna in the horizontal plane. This was seen as widening or gaping of the medial joint space relative to the lateral joint space between the articulating surfaces of the two components and was evident only on the anteroposterior radiograph; on the lateral radiograph, no abnormality such as subluxation was noted (Figs. 4-A, 4-B, 4-C and 4-D).
Another finding was new-bone formation rather than bone resorption around the margin of the articular portion of the ulnar component in the eleven elbows in which both components had been implanted without cement (Figs. 4-C and 4-D).
Postoperative Complications
One elbow (Case 37) had a posterior dislocation soon after the index operation, and because a trial of closed reduction was unsuccessful it was decided to perform an open operation. At the time of the reoperation, the sutures in the soft tissues around the olecranon, including those in the dorsolateral fascial layer, were found to be quite loose, and the loosened tissues again were tightly sutured. The elbow was immobilized in the flexed position (110 degrees) with use of a plaster cast for three weeks after the reoperation. The dislocation did not recur. At the time of the latest follow-up, the result remained satisfactory (a rating of good), with a residual flexion contracture of 60 degrees.
Another elbow (Case 10), which had had severe limitation of motion (fibrous ankylosis) preoperatively, had a progressive decrease in the range of motion in the immediate postoperative period despite an intensive program of exercise and physiotherapy. The radiograph made six weeks after the operation revealed several hazy shadows of ectopic bone in the soft tissues adjacent to the joint. Seven weeks after the first operation, a reoperation was performed to restore the functional arc of motion. At that time, the ectopic bone as well as the dense fibrous scar tissue surrounding the joint were removed as much as possible, and the triceps tendon was lengthened by fifteen millimeters in a V-Y fashion in order to achieve a sufficient range of flexion (120 degrees or more). Soon after the second operation, a triple-medication regimen consisting of prednisolone (ten milligrams a day), etidronate disodium (800 milligrams a day), and indomethacin (seventy-five milligrams a day) was instituted and was maintained at full dose for six weeks; it then was tapered. At the latest follow-up evaluation, four years after the second operation, the range of motion of the elbow was from 65 degrees of extension to 125 degrees of flexion. The patient was very satisfied with the result because much of the function of the upper limb had been restored, with an increase in the rating from 5 to 15 points.
A third elbow (Case 28) had necrosis of the skin edge in a small area in the region of the tip of the olecranon, necessitating free skin-grafting to obtain healing of the wound.
No patient had a superficial or deep wound infection after the operation. Ulnar-nerve neuropathy had been present preoperatively in four elbows; these patients had had dysesthesia as well as hypoesthesia in the area of the ulnar-nerve distribution on the involved side of the upper limb. At the time of the latest follow-up, no patient had symptoms or signs of ulnar-nerve neuropathy, including the four who had had neuropathy preoperatively.
In our previous series of arthroplasties with use of the type-4 prosthesis12, we encountered metallosis and osteolysis, which ultimately led to breakage of the stem of the humeral component. However, through our experience with that series we gained a high level of confidence that fixation of both components without cement can be achieved successfully with porous coating of the stem with titanium alloy, applied by means of the plasma-spray technique, even in elbows with severe rheumatoid arthritis.
In the present series of forty-three elbows, the type-5 prosthesis was used in an attempt to avoid the problems and at the same time to retain the advantages associated with the type-4 prosthesis. As expected, the clinical as well as the radiographic results were very encouraging compared with those of our previous series12. In that series, twenty-five (78 percent) of thirty-two elbows had a satisfactory clinical result, whereas in the present series, thirty-seven (86 percent) of forty-three elbows had a good or excellent (satisfactory) clinical result. Radiographically, only five of thirty-two elbows in the previous series12 retained the full amount of bone mass within the condylar portion of the humeral component, compared with all elbows in the present series, meaning that no appreciable osteolysis had taken place in the current group of elbows.
In our previous series12, breakage of the stem of the humeral component at its basal junction was the most serious late complication, necessitating a revision operation in five elbows. In the present series, no stem had such breakage, probably because of the reinforcement of that portion of the component and also because of the absence of osteolysis and the retention of the bone mass in the region of the humeral condyles; these factors should contribute to supporting the condylar portion of the component against the joint forces, thus preventing excessive stress at the junction of the stem.
Radiographic signs indicating impending loosening of the components were completely absent in the present series; however, because of the intermediate-term duration of follow-up, careful long-term monitoring of the radiographs is necessary before definite conclusions can be reached.
Remarkable relief of pain was the factor that contributed most to the excellent and good clinical results in the current series; twenty-nine elbows were virtually free of pain, and fourteen had mild or occasional pain.
The arc of motion also improved postoperatively. There was a tendency for a loss of extension (an increase in flexion contracture) of some elbows, but the range of flexion was markedly increased in all elbows. Thus, it can be said that, postoperatively, the arc of motion shifted toward a more functional range (an angle of flexion of 120 degrees or more).
As a natural consequence of relief of pain and a more functional arc of motion, activities involving the elbow, such as combing the hair, feeding oneself, and putting on a shirt, became easier for all patients after the operation. However, patients who had a flexion contracture of 50 degrees or more sometimes had difficulty with activities such as putting on shoes and performing personal hygiene.
Despite the nonconstrained nature of the type-5 prosthesis and the absence of the radial head after its excision, all elbows had good (10 points) or fair (5 points) stability postoperatively; there tended to be good stability of elbows that had a residual flexion contracture of 40 degrees or more, which seemed to render the joint more stable in the frontal plane. The configuration or shape of the articulating surfaces of both components also might have provided intrinsic stability. Three ulnar components were seen to be tilted on anteroposterior radiographs, but stability was not adversely affected.
Although we are a little concerned about the worsening, or lack of improvement, of the flexion contracture after the operation, this might be the price that has to be paid for good stability of the joint, as mentioned earlier. Generally, it is difficult to consistently achieve sufficient correction of the flexion contracture with use of a surface-replacement prosthesis6,11,17 such as the type-5 implant. This may be because a variable degree of lengthening of the joint segment occurs after implantation of this prosthesis3. In addition, we suspect that our postoperative rehabilitation program may have overemphasized increasing flexion rather than extension. If this is another reason for the lack of improvement of the flexion contracture, we may need to revise our rehabilitation program and to start range-of-motion exercises for extension earlier.
Another concern is the tilting of the ulnar component in the horizontal plane that was seen in three elbows on anteroposterior radiographs made soon after the operation. Although the same operative approach and technique, including complete release of the ulnar collateral ligament, was used in our previous series of types 1 and 2 implants11 and type-4 implants12, tilting was not observed in those series, suggesting that complete release of the ulnar collateral ligament is not responsible for, or at least is not the only cause of, this finding.
However, it is certain that an imbalance of tension between the soft tissues on the medial and lateral sides of the joint, with tension being tighter on the lateral side, must be present in these elbows. We have found that, as a result of insertion of the ulnar component, the dorsal portion of the trochlear notch of the ulna tends to become thicker in the sagittal plane because of the added thickness3 of the component itself, and this increased thickness may lead to excessive tightening of the soft-tissue structures on the lateral side during closure of the incised dorsolateral fascial layers, particularly in elbows that had tight and contracted periarticular soft tissues preoperatively. Since we noticed this radiographic finding in a few elbows in the current series, we have made it a rule to place at least two tight silk sutures in order to approximate the stumps of soft tissue (joint capsule and ligament) on the medial side of the joint, thus counterbalancing the tighter tension on the lateral side that is generated by the sutures placed in the dorsolateral fascial layer.
Some surgeons may have concerns about the complete release of the ulnar collateral ligament as it may result in postoperative dislocation or instability of the joint. However, we have been using this technique for more than twenty years without problems. We still believe that this procedure is mandatory in arthroplasties in which the type-5 prosthesis is used, for several reasons. First, in an elbow with severe destruction, as is seen in rheumatoid arthritis, the ulnar collateral ligament is in a state of contracture, and the contracted ligament may hamper free movement of the joint after implantation of a surface-replacement prosthesis such as the type-5 implant. Second, the function of the ulnar collateral ligament in the presence of a flexion contracture could largely be compensated for by the various soft-tissue structures on the medial side of the joint. Finally, the articulating surface of the humeral component of the type-5 prosthesis has a shallow, wide monofacet configuration in the frontal plane, which allows for mediolateral shifting of the ulnar component on the articular surface of the humeral component without the risk of lateral subluxation; thus, one function of the ulnar collateral ligament—to prevent mediolateral movement (lateral subluxation)—is not necessarily mandatory for the type-5 prosthesis.
Although the ulnar collateral ligament, as already mentioned, is routinely released with our approach, the lateral ligament always is carefully preserved to prevent varus instability from occurring at a later date.
Neuropathy of the ulnar nerve is one of the main complications of total elbow arthroplasty, and its frequency has varied6,14-17. In the present series as well as in our previous series of type-4 implants12, no elbow had signs of ulnar-nerve involvement after the operation. Conversely, four elbows in the current series that had had neuropathy due to rheumatoid involvement before the operation had gradual resolution of the symptoms six to twelve months postoperatively. We believe that the absence of neurological complications in the present series can be attributed to the extensive release of the nerve performed at the beginning of the operation to protect the nerve from excessive stretching during the subsequent operative procedures and to avoid exposing it to too much local pressure or entrapment by its anterior transposition. Indeed, we have seen virtually no complications related to the ulnar nerve since the introduction of extensive release of the nerve. However, this finding may need to be verified in another large series in which the operations are performed by different surgeons.
Eight elbows had a Darrach procedure in conjunction with the arthroplasty in order to restore rotation of the forearm, particularly supination, and seven of these elbows had a functional arc of supination at the time of the latest follow-up. This improvement in supination, combined with the increased range of flexion of the elbow, helped the patients to increase function greatly in daily activities involving use of the upper limb.
Although fixation of the humeral component without cement was successful in elbows that had slight or moderate osteoporosis and there was no selection bias in this series, the success of fixation without cement in elbows with very severe osteoporosis is not guaranteed, and we recommend the use of cement in such elbows. We believe that fixation of the ulnar component without cement should be reserved for elbows with good-quality bone and sufficient bone stock for the component to be securely supported by the bone.
Although all eleven metal-backed ulnar components that had been implanted without cement had secure fixation, we now prefer fixation of the all-polyethylene ulnar component with cement to fixation of the metal-backed component with the porous-coated stem without cement. We have much more confidence in the former method because of the long-term results with use of the all-polyethylene ulnar component of types 1 and 211 and type-3 prostheses.
Agins, H. J.; Alcock, N. W.; Bansal, M.; Salvati, E. A.; Wilson, P. D., Jr.; Pellicci, P. M.; and Bullough, P. G.: Metallic wear in failed titanium-alloy total hip replacements. A histological and quantitative analysis. J. Bone and Joint Surg.,70-A: 347-356, March 1988.70-A347
1988
Arnett, F. C.; Edworthy, S. M.; Bloch, D. A.; McShane, D. J.; Fries, J. F.; Cooper, N. S.; Healey, L. A.; Kaplan, S. R.; Liang, M. H.; Luthra, H. S.; Medsger, T. A., Jr.; Mitchell, D. M.; Neustadt, D. H.; Pinals, R. S.; Schaller, J. G.; Sharp, J. T.; Wilder, R. L.; and Hunder, G. G.: The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthrit. and Rheumat.,31: 315-324, 1988.31315
1988
Blewitt, N., and Pooley, J.: Elbow lengthening after total prosthetic arthroplasty: observations and clinical implications. J. Shoulder and Elbow Surg.,3: 200-206, 1994.3200
1994
Campbell, W. C.: Incision for exposure of the elbow joint. Am. J. Surg.,15: 65-67, 1932.1565
1932
Darrach, W.: Forward dislocation at the inferior radio-ulnar joint, with fracture of the lower third of the shaft of the radius. Ann. Surg.,56: 801-802, 1912.56801
1912
Ewald, F. C.; Simmons, E. D., Jr.; Sullivan, J. A.; Thomas, W. H.; Scott, R. D.; Poss, R.; Thornhill, T. S.; and Sledge, C. B.: Capitellocondylar total elbow replacement in rheumatoid arthritis. Long-term results. J. Bone and Joint Surg.,75-A: 498-507, April 1993.75-A498
1993
Gabel, G. T., and Amadio, P. C.: Reoperation for failed decompression of the ulnar nerve in the region of the elbow. J. Bone and Joint Surg.,72-A: 213-219, Feb. 1990.72-A213
1990
Gill, D. R. J., and Morrey, B. F.: The Coonrad-Morrey total elbow arthroplasty in patients who have rheumatoid arthritis. A ten to fifteen-year follow-up study. J. Bone and Joint Surg.,80-A: 1327-1335, Sept. 1998.80-A1327
1998
Inoue, H.; Hamazaki, J.; Fujiwara, M. Hayashi, M.; Shuh, K.; and Tanabe, G.: [Alumina ceramic elbow prosthesis.]. Seikei Saigaigeka,25: 1207-1212, 1982.251207
1982
Kudo, H.; Iwano, K.; and Watanabe, S.: Total replacement of the rheumatoid elbow with a hingeless prosthesis. J. Bone and Joint Surg.,62-A: 277-285, March 1980.62-A277
1980
Kudo, H., and Iwano, K.: Total elbow arthroplasty with a non-constrained surface-replacement prosthesis in patients who have rheumatoid arthritis. A long-term follow-up study. J. Bone and Joint Surg.,72-A: 355-362, March 1990.72-A355
1990
Kudo, H.; Iwano, K.; and Nishino, J.: Cementless or hybrid total elbow arthroplasty with titanium-alloy implants. A study of interim clinical results and specific complications. J. Arthroplasty,9: 269-278, 1994.9269
1994
[PubMed]
Kudo, H.: Non-constrained elbow arthroplasty for mutilans deformity in rheumatoid arthritis. A report of six cases. J. Bone and Joint Surg.,80-B(2): 234-239, 1998.80-B(2)234
1998
Morrey, B. F., and Adams, R. A.: Semiconstrained arthroplasty for the treatment of rheumatoid arthritis of the elbow. J. Bone and Joint Surg.,74-A: 479-490, April 1992.74-A479
1992
Pöll, R. G., and Rozing, P. M.: Use of the Souter-Strathclyde total elbow prosthesis in patients who have rheumatoid arthritis. J. Bone and Joint Surg.,73-A: 1227-1233, Sept. 1991.73-A1227
1991
Risung, F.: The Norway elbow replacement. Design, technique and results after nine years. J. Bone and Joint Surg.,79-B(3): 394-402, 1997.79-B(3)394
1997
Weiland, A. J.; Weiss, A.-P. C.; Wills, R. P.; and Moore, J. R.: Capitellocondylar total elbow replacement. A long-term follow-up study. J. Bone and Joint Surg.,71-A: 217-222, Feb. 1989.71-A217
1989
Williams, D. F.: Editorial. Titanium: epitome of biocompatibility or cause for concern. J. Bone and Joint Surg.,76-B(3): 348-349, 1994.76-B(3)348
1994