- Email: [email protected]
Early “simple” release of posttraumatic elbow contracture associated with heterotopic ossification
Early “Simple” Release of Posttraumatic Elbow Contracture Associated With Heterotopic Ossification Randall W. Viola, MD, Douglas P. Hanel, MD, Seattle, WA “Simple” elbow release in the setting of heterotopic ossification is defined as excision of ectopic bone and removal of restricting soft tissues without associated articular procedures. In the past, such procedures were postponed until bone scans were quiescent, serum alkaline phosphatase was normal, and the ectopic bone was mature. Postoperative management sometimes included radiation therapy, prolonged nonsteroidal anti-inflammatory agents, and intensive physiotherapy. We believe that delayed treatment beyond the time of fracture healing is unnecessary to obtain results comparable to those of previous studies. Similarly, we propose that radiation therapy is not necessary after excision of heterotopic ossification. Fourteen patients (15 elbows) were prospectively managed with early excision of posttraumatic heterotopic ossification, immediate postoperative mobilization, and a 5-day course of indomethacin. The average time from injury to release was 23 weeks. The mean preoperative arc of flexion/extension was 43°; that of pronation/supination was 79°. After 2 years, the corresponding values were 120° and 152°. Cubital tunnel syndrome, present in 5 patients, resolved after surgery. Three postoperative complications occurred in 2 patients. There were no recurrent contractures or loss of motion. (J Hand Surg 1999;24A:370 –380. Copyright © 1999 by the American Society for Surgery of the Hand.) Key words: Elbow joint, surgery, contracture, ossification, heterotopic.
Elbow stiffness commonly complicates elbow trauma.1 Regan and Reilly1 outlined 3 factors that cause posttraumatic elbow stiffness: the high degree of articular congruity and conformity of the elbow joint predispose the elbow to limited motion after articular injury; the brachialis covers the anterior capsule, predisposing it to posttraumatic periarticular
From the Department of Orthopaedic Surgery, University of Washington, Harborview Medical Center, Seattle, WA. Received for publication February 26, 1998; accepted in revised form September 29, 1998. 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. Reprint requests: Douglas P. Hanel, MD, Department of Orthopaedic Surgery, University of Washington, Harborview Medical Center, Box 359798, 325 9th Ave, Seattle, WA 98104. Copyright © 1999 by the American Society for Surgery of the Hand 0363-5023/99/24A02-0014$3.00/0
370 The Journal of Hand Surgery
ossification; and mobilization after elbow injury is often delayed because it is difficult to achieve rigid internal fixation of comminuted elbow fractures. Furthermore, elbow stiffness may occur in spite of both early aggressive motion and prophylactic measures and periarticular ossification may lead to cubital tunnel syndrome.2– 4 Elbow stiffness is disabling because a wide range of motion (ROM) is necessary for basic everyday activities. Morrey et al’s5 electrogoniometer analysis demonstrated that an arc of elbow motion of 100° (range, 30° to 130°) and an arc of forearm rotation of 110° (range, 55° pronation to 55° supination) is necessary for an individual to perform 90% of his or her normal daily activities. A variety of treatments have been used for posttraumatic stiff elbows. Elbow arthrodesis causes severe functional impairment and should be considered
The Journal of Hand Surgery / Vol. 24A No. 2 March 1999 371
only as a last resort. Turnbuckle splinting does not significantly improve motion.6 Total elbow arthroplasty has a high perioperative complication rate and an unacceptable durability in patients younger than 60 years.7 Interposition arthroplasty has inconsistent results.8 Distraction arthroplasty, a technically demanding procedure, has produced good results but is unnecessary if articular surfaces are preserved.9 “Simple” elbow contracture release, described by Morrey9 as the removal of the causative abnormality without performing any procedures on the elbow articular surfaces, may be used to restore motion to the posttraumatic stiff elbow due to elbow capsule contracture and/or heterotopic ossification (HO). The morbidity associated with elbow release and HO excision is high. Complications include triceps rupture, elbow instability, and neurologic injury.9 Although simple elbow contracture release has good documented results, the timing of this procedure remains controversial.9,10 Many investigators advocate waiting 1 to 2 years before release.11 The desire to delay surgery until heterotopic bone has become metabolically quiescent must be balanced against the risks of progressive soft tissue contracture, potential articular cartilage destruction, and prolonged infirmity.11 In the past, surgical release of the posttraumatic stiff elbow was not performed until the patient’s bone scan and serum alkaline phosphatase were normal and the HO, as seen on plain radiographs, appeared mature. These prerequisites are typically met at 1 to 2 years after injury, with earlier intervention thought to predispose to HO recurrence. Recent reports suggest that good results may be obtained with early surgical release combined with postrelease nonsteroidal anti-inflammatory drugs or radiation therapy.10 We describe the results of a prospective treatment protocol involving early operative intervention (simple elbow contracture release), a short course of nonsteroidal anti-inflammatory drugs, and intensive therapy in patients with posttraumatic stiff elbows.
Materials and Methods Patient Selection From 1992 to 1995, 16 patients (18 elbows) with posttraumatic elbow stiffness associated with HO fulfilled the criteria for simple contracture release as described by Morrey.9 Fourteen patients (15 elbows) were treated within 32 weeks of the original injury; these cases were designated early and are the subject of this study. Elbows 1 and 2 (Tables 1 to 4) were
from a single patient. All patients were managed prospectively according to the contracture release protocol described below.
Indications Our criteria for elbow contracture release included (1) function-limiting elbow stiffness resulting from musculoskeletal trauma (limitation was defined as subjective complaints regarding the activities of daily living and at least 1 of 3 objective findings: a flexion/extension arc ,100°, a pronation/supination arc ,110°, or a flexion contracture .45°12), (2) radiographic evidence of union of fractures, (3) radiographic evidence of intact ulnohumeral joint articular surfaces, (4) HO in any stage of maturity on plain radiographs, and (5) stabilization of traumatic brain injury. Contractures following thermal injury were excluded.
Patient Demographics Patient characteristics are listed in Table 1. Most of the patients were young adult males (average age, 28 years) who sustained elbow trauma, were righthanded, and were treated without HO prophylaxis. Seven patients (53%) sustained traumatic brain injuries; of these patients, 6 were in a coma for at least 1 week (range, 1 to 5 weeks). Table 2 lists patient characteristics at the time of simple contracture release. Nine elbows had normal extremity neurologic examinations. Five elbows had cubital tunnel syndrome. In each of these cases, the diagnosis was established by subjective complaints and confirmed by Semmes-Weinstein monofilament sensibility testing (a minimum of 1-thickness difference in the ring and small finger compared with the radial 3 digits) and visible atrophy of the first dorsal interosseous muscle, when present. Electrodiagnostic studies were not routinely performed. Ulnar nerve lesions were classified according to McGowan13: 2 patients were grade I (decreased sensibility without ulnar intrinsic weakness), 3 were grade II (decreased sensibility with ulnar weakness and atrophy), and none were grade III (decreased sensation with ulnar intrinsic paralysis. One patient (no. 10) sustained an avulsion of the ulnar nerve resulting in a 10-cm defect at the time of injury. The 7 patients (8 elbows) who had sustained traumatic brain injuries were classified according to Garland et al14 at the time of contracture release. These investigators address both cognitive and physical impairment related to traumatic brain injury. In classifying our patients, elbow
372 Viola and Hanel / Release of Elbow Contracture Associated With HO
Table 1. Patient Demographics at the Time of Initial Injury Hand Dominance
Traumatic Brain Injury
Elbow
Age (yr)
Gender
Coma (Duration)
1
29
M
R
Yes
No
L Elbow DL, radial head FX, MCL rupture
2
29
M
R
Yes
No
Radial head FX, DRUJ DL
3
32
M
R
No
No
Radial head FX
4
21
M
R
Yes
Yes (1 wk)
5
38
M
R
Yes
Yes (4 wk)
6
35
M
R
Yes
Yes (1 wk)
7
28
M
R
Yes
Yes (3 wk)
8
20
M
R
No
No
Distal humerus and olecranon FX No orthopedic diagnosis, severe CHI Segmental ulnar FX Distal humerus, radial head, olecranon FX Elbow DL and olecranon FX
9
27
M
R
Yes
Yes (2 wk)
10
22
M
R
No
No
11
19
F
R
Yes
Yes (5 wk)
12 13
26 42
M M
R R
No No
No No
14
33
M
R
No
No
15 Summary/ average
21 28
M M (14) F (1)
R All R
No 53% yes
No No (8 patients) Coma (6 patients)
Initial Injury
Proximal radius and ulnar FX Medial elbow degloving with loss of ECU, MCL, and ulnar nerve; elbow dislocation Distal humerus FX Elbow DL Olecranon, coronoid, and radial head FX Type II Monteggia FX/ DL; type III radial head FX Olecranon FX Multiple diagnosis
Initial Management
HO Prophylaxis
Closed reduction, MCL repair, radial head prosthesis Radial head prosthesis, DRUJ pinning Radial head prosthesis ORIF of humerus and olecranon
No
None
No
ORIF of ulna
No
ORIF of humerus, radius, olecranon Closed reduction and ORIF of olecranon ORIF of radius and ulna Lateral free flap and static elbow external fixator; MCL reconstruction
No
ORIF of distal humerus Closed reduction ORIF of olecranon, coronoid, and radial head ORIF of ulna; radial head excision
No
ORIF Variable
No No (14) Indomethacin (1)
No No No
No No Yes (indocin 3 6 wk)
No No
No
DL, dislocation; FX, fracture; MCL, medial collateral ligament; DRUJ, distal radioulnar joint; CHI, closed head injury; ECU, extensor carpi ulnaris; ORIF, open reduction, internal fixation.
stiffness-related disability was not included as a parameter for the Garland et al classification because it resulted from musculoskeletal rather than neurologic injury. Five patients (6 elbows) were Garland grade I (no or mild cognitive and physical impairment), 1
was grade II (minimal cognitive and moderate physical disability), and 1 was grade IV (moderate to severe cognitive deficits with minimal to moderate physical disability).14 Fourteen elbows had no spasticity; 1 patient had mild spasticity. Patients were
The Journal of Hand Surgery / Vol. 24A No. 2 March 1999 373
categorized according to both the functional classification of elbow stiffness (Table 2) and the anatomic description of elbow HO of Hastings and Graham.11 The functional classification of Hastings and Graham stratifies patients according to the plane(s) of elbow stiffness: grade I (no patients), no functional limitation; grade IIA (6 patients), limited flexion/extension; grade IIB (1 patient), limited pronation/supination; grade IIC (5 patients), limited flexion/extension and pronation/supination; and grade III (3 patients), elbow ankylosis preventing either flexion/extension and/or pronation/supination. Hastings and Graham11 describe the location of HO as posterior elbow, anterior elbow, medial elbow, and lateral elbow. Among our patients, posterior involvement occurred in 12 elbows, anterior involvement occurred in 10, medial involvement occurred in 7, and lateral involvement occurred in 9. Preoperative pain was evaluated according to pain medication use at the time of contracture release. Each patient was asked to describe the use of pain medications to quantify his or her preoperative pain and to classify it as none, mild, moderate, or severe. If pain was controlled with nonopiate pain medications, the patient was classified as having mild pain. Patients with regular opiate use were classified as severe. The integrity of the ulnohumeral joint articular surfaces was evaluated by assessing the ulnohumeral joint space on plain anteroposterior and lateral elbow radiographs and was confirmed by visual inspection at the time of contracture release. Based on radiographs, the patient’s ulnohumeral articulations were classified as either normal or minimally altered. Minimally altered ulnohumeral articulations were those with joint space loss of less than 50% and/or articular incongruity less than 2 mm. Patients with more significant ulnohumeral articular abnormality were considered for reconstructive procedures rather than simple elbow contracture release. Radiographically, 12 elbows had normal ulnohumeral articular surfaces and 3 had minimal articular abnormalities. At surgery, all patients had normal or near-normal elbow articular surfaces except for patient 7, who had lost 50% of the distal humeral trochlear articular surface. This patient is included in the study because he met the radiographic criteria for inclusion and because his ulnar, radial head, and capitellar articular surfaces appeared normal. Each patient’s active elbow ROM before contracture release is listed in Table 2. The average preoperative ROM values as a function of Hastings and Graham’s11 classification are listed in Table 3.
The average time from injury to contracture release was 23 weeks (range, 10 –31 weeks). At the time of release, overall average preoperative flexion and extension were 99° (range, 85° to 135°) and 56° (range, 30° to 90°), respectively, and overall average preoperative pronation and supination were 41° (range, 0° to 90°) and 38° (range, 0° to 80°), respectively. Concomitant surgical procedures performed at the time of simple contracture release are listed in Table 2.
Technique The essential steps of the procedure are: 1. Remove all HO from the brachialis, triceps, and adjacent to the collateral ligaments. 2. Excise the anterior and posterior joint capsule, leaving the collateral ligaments intact (if possible). 3. Clear the coronoid and olecranon fossa of fat and synovium. 4. Decompress the ulnar nerve in the presence of cubital tunnel syndrome. 5. Correct any elbow instability that exists after resection.
The focus of each procedure was removal of HO and contracted capsule until normal elbow ROM was restored. Contracture release was considered complete only after a flexion/extension arc of 140°/10° and a pronation/supination arc of 80°/80° had been achieved. The triceps tendon was not detached from its insertion. The incision used to achieve the steps listed above was individualized based on the following factors: (1) location of previous elbow incision(s), (2) need for ulnar nerve decompression, and (3) plane(s) of elbow contracture. In cases in which a previous approach to the elbow had been performed, that approach was used as long as it allowed all the surgical objectives to be achieved. If ulnar nerve decompression was necessary, either a medial or posterior approach was used. In patients with restricted forearm rotation but preserved elbow flexion and extension, a lateral approach was favored. In patients with restricted flexion and extension but preserved forearm rotation, a medial approach was favored. In patients with loss of elbow motion in both planes, a posterior approach was favored. The above factors were considered in determining the best approach to use for each patient. When forearm pronation and supination was limited, a lateral extensile Kocher approach was used.15 The radial head and neck were exposed, the orbicu-
374 Viola and Hanel / Release of Elbow Contracture Associated With HO
Table 2. Patient Characteristics at the Time of Simple Contracture Release
Elbow
Ulnar Nerve Status
McGowan Class
TBI Status
HO Grade
HO Location
Preoperative Elbow Pain
Articular Surfaces
Time (wk)
1
Intact
NA
I
IIC
AE, LE
NA
Normal
27
2
Intact
NA
I
IIB
AE, LE
NA
Normal
27
3
Intact
NA
NA
IIC
AE, LE
Mild
Normal
17
4
Intact
NA
I
III
PE, AE, LE
Mild
Mild
20
5
Late palsy
I
IV
IIA
PE, ME
None
Normal
21
6
Late palsy
II
I
IIA
PE, AE, ME
None
Normal
26
7
Intact
NA
I
III
PE, AE, ME, LE
Mild
50% Distal humerus injury
22
8
Late palsy
II
NA
IIC
PE, AE, ME
NA
Normal
19
9
Late palsy
II
I
IIA
PE, ME
None
Normal
21
NA
NA
IIC
PE, AE, LE
Severe
Normal
10
11 12
Traumatic neurotomesis Intact Intact
NA NA
II NA
IIA IIA
None None
Normal Normal
31 18
13
Late palsy
I
NA
III
PE PE, AE, ME, LE PE, ME, LE
Mild
Mild
28
14
Intact
NA
NA
IIA
PE
Mild
Normal
31
15
Intact
NA
NA
IIC
AE, LE, PE
Mild
Normal
26
IIA (6) IIB (1) IIC (5) III (3)
PE (12) AE (10) ME (7) LE (9)
None (5) Severe (1) Mild (6)
Normal (12) Mild (3)
23
10
Summary/ Late palsies (5) Grade I (2) average Neurotomesis (1) Grade II (3) Intact (9) Grade III (0)
Grade I (6) Grade II (1) Grade IV (1)
Grade Grade Grade Grade
NA, not applicable; TBI, traumatic brain injury; AE, anterior elbow; LE, lateral elbow; PE, posterior elbow; ME, medial elbow; HW, hardware; ORIF, open reduction, internal fixation; BG, bone graft; MCL, medial collateral ligament; CPM, continuous passive motion.
laris and lateral collateral ligaments were preserved, and the HO bridging the proximal radioulnar joint was removed. If there was an associated loss of elbow flexion and extension, the exposure was carried proximally. The origin of the brachioradialis and the wrist extensors was elevated, the anterior joint capsule was excised, and the HO within the brachialis muscle was removed. Hypertrophic fat or synovium found in the coronoid fossa was then removed. If there was continued limitation in motion, the tri-
ceps muscle was elevated from the humerus, the posterior capsule was excised, and the fat or synovium was removed from the olecranon fossa. When fracture callus caused impingement of the olecranon, 1.5 cm of the tip of the olecranon was removed. If there was continued impingement, the olecranon fossa was widened with a ball tip burr. When forearm motion was normal and elbow flexion and extension were restricted, or in those cases in which cubital tunnel syndrome was present, the el-
The Journal of Hand Surgery / Vol. 24A No. 2 March 1999 375
Preoperative (°) Flexion Extension Pronation Supination 120
35
50
30
120
35
45
20
135
35
15
35
95
90
0
0
92
70
80
70
90
30
45
80
85
75
0
0
100
75
45
45
100
85
65
60
90
50
25
0
65 80
40 55
80 90
80 80
110
30
0
0
110
45
80
80
100
90
0
0
99
56
41
38
Procedure(s) at Time of Release Excision of HO, implant removal Excision of HO, implant removal Excision of HO, implant removal Excision of HO, HW removal, ORIF/BG ulna nonunion Excision HO, ulnar nerve transposition Excision HO, ulnar nerve transposition Excision HO, HW removal, ulnar nerve transposition, annular ligament reconstruction Excision HO, HW removal, ulnar nerve transposition Excision HO, ulnar nerve transposition Excision HO, MCL reconstruction Excision HO Excision HO, ulnar nerve transposition Excision HO, HW removal, ulnar nerve transposition Excision HO, removal of radial head implant Excision HO, HW removal, ulnar nerve transposition Excision of HO (15) Implant removal (7) Ulnar nerve transposition (8)
bow was approached medially.16 Previous surgical incisions were used to mobilize the skin until the medial epicondylar region was exposed. The ulnar nerve was identified in the mid-arm and dissected distally into the proximal forearm. In 4 cases, the ulnar nerve was completely encased in HO at the cubital tunnel. The nerve was released by removing the HO with ronguers. The medial intermuscular septum of the arm was removed and the origin of the flexor–pronator mass was elevated from the medial
Fixator
Surgical HO Prophylaxis Postoperative Approach(es) After Release CPM
None
Lateral
Indocin 3 5 d
None
None
Lateral
Indocin 3 5 d
None
None
Lateral
Indocin 3 5 d
6 wk
None
Posterior
Indocin 3 5 d
6 wk
None
Medial
Indocin 3 5 d
1d
None
Medial
Indocin 3 5 d
4d
6 wk
Medial and posterior
Indocin 3 5 d
6 wk
None
Posterior
Indocin 3 5 d
5d
None
Medial
Indocin 3 5 d
5d
None
Medial
Indocin 3 6 wk
None
None None
Lateral Medial
Indocin 3 5 d Indocin 3 5 d
6 wk 5d
None
Posterior
Indocin 3 5 d
5d
None
Lateral
None
None
None
Medial
None
None
Fixator (1) None (14)
14 Single 1 Double
None (2) 5 d (12) 6 wk (1)
Median 55d
epicondyle of the humerus. This provided direct vision of the brachialis muscle and the anterior joint capsule. Similar to the lateral approach, the anterior capsule was excised and the heterotopic bone removed. While retracting the ulnar nerve with a 1-inch Penrose drain, the posterior joint capsule was entered through a subperiosteal dissection. Once inside the joint, the capsule was removed and the olecranon fossa cleared of fat, synovium, and compromising fracture callus. After the surgical release
376 Viola and Hanel / Release of Elbow Contracture Associated With HO
Table 3. Preoperative and Postoperative Average Ranges of Motion and Average Arcs of Motion as a Function of Hastings and Graham’s Classification11 Classification Grade IIA (F/E , 100, P/S . 110) Grade IIB (F/E . 100, P/S , 110) Grade IIC (F/E , 100, P/S , 110) Grade III (F/E and/or P/S ankylosis) Total
No. of Patients
Flexion
Extension
Pronation
Supination
F/E Arcs
P/S Arcs
6
85/144
56/9
72/86
74/85
29/135
146/171
1
135/140
35/25
45/45
20/90
100/115
65/135
5
106/139
49/21
34/80
28/73
57/118
62/153
3 15
97/150 98/140
65/24 54/20
0/72 41/77
0/63 38/75
32/126 44/120*
0/135 79/152*
Data are presented as average preoperative/postoperative values, in degrees. F/E, flexion/extension; P/S, pronation/supination. *p , .05.
had been completed, elbow stability was assessed. Isolated medial or lateral ligamentous instability was treated with autograft reconstruction as described by Conway et al17 or Nestor et al,18 respectively. One patient with both varus and valgus instability was stabilized with a hinge fixator for 6 weeks.1 In those cases approached medially, an anterior submuscular transposition of the ulnar nerve was completed. Submuscular, rather than subcutaneous, transposition was performed because nearly all the necessary steps, including flexor–pronator mass elevation, were already completed during HO and capsule resection. Care was taken to identify and excise any fascial bands that would compromise a smooth passage of the nerve into the forearm. The origin of the flexor–pronator mass was reconstructed with stout nonabsorbable sutures passed through drill holes in the medial epicondyle. Suction drains were placed anterior and posterior to the elbow joint and the skin was closed with interrupted sutures.
Postoperative Care One patient required stabilization with the Brigham distraction fixator. Thirteen patients were treated with immediate postoperative continuous passive mobilization (CPM). Of these 13 patients, 6 were treated with CPM for 4 days postoperatively, 4 patients continued CPM at home for a total of 6 weeks, and 3 patients did not tolerate CPM beyond the immediate time in recovery room because of pain and inconvenience. The latter 3 patients were treated with active and passive elbow motion 4 times per day. The median number of days of CPM use was 5 days.
After surgery, pain was controlled with constant perfusion of regional anesthesia through an axillary catheter and patient-controlled analgesic pumps for 3 days. This was followed by oral analgesics. Patients were instructed in elbow ROM exercises and were encouraged to perform them at least 4 times a day for 6 months. Gentle active and passive exercises were instituted on the first postoperative day. Aggressive strengthening exercises were postponed until the third week after surgery. Splints, slings, and other forms of immobilization were strongly discouraged while the patients were awake. Patients often lost elbow extension early in their postoperative course; such cases were treated with night extension splinting for 6 weeks. Indomethacin (25 mg orally or rectally 3 times per day) and sulcralfate (1 g orally twice a day) were prescribed for 5 postoperative days. Radiation was not used.
Patient Evaluation Each patient was examined weekly for 3 weeks and then every 12 weeks for at least 2 years. At each visit, active elbow ROM, Semmes-Weinstein sensibility, and gross motor strength were documented. Each case was analyzed by review of medical records, radiographs, physical examinations, and patient interviews.
Statistical Analysis Statistical analyses of the preoperative and postoperative flexion/extension and pronation/supination ROM arcs were compared using the paired Student’s
The Journal of Hand Surgery / Vol. 24A No. 2 March 1999 377
t-test. Analysis was performed using Systat 7.0 (SPSS, Inc, Evanston, IL). The null hypothesis was that the preoperative and postoperative ROM arcs would be similar. Statistical significance was established at p , .05.
Results Postoperative results are summarized in Tables 3 and 4. Table 4 also lists the results and postoperative patient characteristics. The average duration of follow-up was 115 weeks (range, 79 –156 weeks). Average postoperative flexion and extension were 140° (range, 130° to 145°) and 20° (range, 0° to 37°), respectively. The flexion/extension arc improved from an average of 43° before surgery to an average of 120° at 2 years (p , .05) (Table 3). Average postoperative pronation and supination were 77° (range, 45° to 90°) and 75° (range, 25° to 90°), respectively (Table 4). The pronation/supination arc improved from an average of 79° before surgery to an average of 152° at 2 years (p , .05) (Table 3). Three patients (nos. 7, 14, and 15) who did not achieve the ROM thought to be necessary for functional elbow motion (100° flexion/extension arc and 110° pronation/supination arc) were all very close to these goals. All 3 of these patients had a flexion/ extension arc of 95°. Only 1 patient had limitation of pronation/supination. Patient 7 had an 85° pronation supination arc. There were no intraoperative complications. Three postoperative complications occurred in 2 patients. The first patient (no. 4) had tenuous elbow soft tissue coverage at the time of contracture release. With increased activity after the contracture release, the soft tissue envelope became increasingly unstable with multiple episodes of skin ulceration. This was treated with a free flap soft tissue reconstruction 15 months after contracture release. The second patient (no. 7) had 2 complications. First, he developed an olecranon pin track infection 3 weeks after the contracture release (he had been treated with a hinge fixator). This resolved after 10 days of oral antibiotics. The pin was not exchanged. Second, serial radiographs during the first postoperative year showed radiographic resorption of the capitellum. The patient’s white blood cell count and erythrocyte sedimentation rate were normal. As a result of this bone loss, 2 K-wires, which had previously been buried in the capitellum, became loose within the elbow joint. An arthrotomy was required to retrieve these wires (intraoperative cultures were negative). The condi-
tion of the elbow was clinically compatible with Charcot arthropathy without instability. He has pain with heavy lifting but continues to work as a carpenter. Hand dominance was unchanged in 13 patients; 1 patient (no. 10) became ambidextrous to compensate for the permanent ulnar nerve lesion he sustained at the time of the original injury. There were no nerve injuries or nerve palsies resulting from the contracture release. Semmes-Weinstein monofilament sensibility of 2.83 to 3.22 was found in 14 elbows, including those in the 5 patients with preoperative cubital tunnel syndrome. Patient 10 had a permanent ulnar nerve injury as previously described. Six months after elbow release he underwent tendon transfers to compensate for intrinsic motor deficits and debulking of a latissimus dorsi free flap, which provided primary coverage of the open elbow fracture dislocation.
Discussion Like other published series of elbow contracture releases, this report describes a mixed patient population. However, most were young adult males, frequently with traumatic brain injuries, who sustained significant elbow trauma and had their initial injury managed without HO prophylaxis. The time from injury to HO excision was also variable; however, all contractures were released within an average of 23 weeks (range, 24 to 78 weeks), earlier than in other studies.2,9,10,14,19,20 The extent and distribution of HO were determined by plain radiographs. Advanced imaging studies, such as computed tomography and magnetic resonance imaging, were not obtained for HO localization, as we found plain radiographs adequate for this purpose. All patients demonstrated statistically and clinically significant improvement in elbow motion (p , .05) and elbow function after 2 years. Furthermore, those with ulnar nerve palsies resulting from compression recovered full ulnar nerve function. We attribute these results to thorough operative resection of HO and capsular excision, submuscular ulnar nerve transfer, and aggressive, unlimited, postoperative active and passive motion. The focus of each release was complete resection of HO and capsule so that intraoperative ROM was as close to normal as possible. In addition, all elbows were stable after surgery due to either preservation of the collateral and annular ligaments, ligament reconstruction, or hinge fixator stabilization. This stability allowed un-
87 87 87 100 149
93 91 104
97 134 106
104 139 79 156 80 115
1 1 2 3 4
5 6 7
8 9 10
11 12 13 14 15 Summary/ average
140 140 145 140 140 140
140 130 140
145 145 130
140 140 140 135 145
Flexion (°)
NA, not available; R, right.
Follow-up (wk)
Elbow
10 5 0 45 45 20
20 10 12
10 10 35
35 35 25 15 37
Extension (°)
85 85 90 80 80 77
70 80 80
85 80 60
90 90 45 80 50
Pronation (°)
90 75 90 80 45 75
80 80 60
90 80 25
70 70 90 80 80
Supination (°)
Normal Normal Normal Normal Normal All normal
Normal Normal Normal
Normal Normal Normal
Normal Normal Normal Normal Normal
Stability
None None None Mild None 10 None 3 mild 2 NA
NA NA None
None None Mild
None None None None Mild
Pain
None None None None None 2 Patients
None None None None Elbow free flap coverage None None Pin track infection, pin migration, Charcot elbow None None None
Complications
Table 4. Postoperative Results and Characteristics
R R R R R 13 Unchanged 1 Bimanual
R R Bimanual
R R R
R R R R R
Hand Dominance
Intact Intact Intact Intact Intact 14 Intact 1 Unchanged
Intact Intact Unchanged
Intact Intact Intact
Intact Intact Intact Intact Intact
Neurologic Status
None None Lateral flap debulking, tendon transfers None None None None None 3 Patients
None None None None Elbow free flap coverage None None Pin removal
Additional Surgery
378 Viola and Hanel / Release of Elbow Contracture Associated With HO
The Journal of Hand Surgery / Vol. 24A No. 2 March 1999 379
restricted postoperative motion. In a frequently quoted reference, Thompson and Garcia19 assert that passive elbow ROM causes periarticular elbow HO formation and elbow stiffness after elbow injury. A review of this report reveals that their analysis was retrospective and, as such, subject to selection bias. Their conclusion that passive motion during convalescence should never be used is based on the fact that those patients developing stiff elbows “received passive stretching by therapists or by the use of weights.”17 Who else but patients with limited elbow motion would be subjected to passive elbow stretching? Until there is a prospective study comparing patients with elbow injuries treated with passive stretching to those treated without passive stretching, the relationship between passive elbow motion and HO formation will remain unclear. We are unsure of the role of CPM. Our goal was to restore elbow motion. To that end, we used CPM for the initial 5 days after surgery, and in those patients who did not tolerate CPM, we used aggressive active and passive ROM exercises. Our results demonstrate significant gains in motion. The number of patients and the study protocol do not allow statements of statistical significance with regard to the effects of CPM versus exercise alone. The role of prophylactic measures for the prevention or recurrence of HO about the elbow remains anecdotal. While many investigators espouse the utility of indomethacin, the length of its use remains arbitrary.20 –22 We chose to limit indomethacin use to 5 days, that period of time when patient compliance is maximal. Similarly, the role of radiation therapy is unclear. McAuliffe and Wolfson10 recently indicated that radiation is beneficial after early excision but they provided no control group for comparison. It is our belief that their results reflect the attention to operative detail and the rehabilitation efforts necessary for satisfactory results in this difficult group of patients without relation to radiation therapy. This belief is supported by Jupiter and Ring’s23 presentation of excision of ectopic bone to treat synostosis about the proximal radioulnar joint. These investigators had only 1 recurrence among 18 limbs despite the lack of adjuvant radiation therapy or anti-inflammatory medication,23 raising even further the questionable role of indomethacin in our patients. The extent of HO recurrence in our patients is unknown. Since the functional gains of our patients were impressive and maintained, radiographs were not routinely obtained. The complication rate of posttraumatic elbow con-
tracture release is high.9,15 Complications include triceps avulsion, ulnar nerve paresthesias, deep infection, aseptic resorption of the distal humerus, and skin slough.9,15 Of our 14 patients (15 elbows), only 2 had complications. Patient 7 had an unusual complication: a Charcot elbow. Morrey9 described a similar phenomenon. The etiology of this process is unclear. In our case, we believe that this resulted from excessive soft tissue dissection in the presence of a healing capitellum fracture. The soft tissue resection, although necessary for release, appears to have rendered the capitellum avascular. We do not believe that this was caused by a septic process. There was no indication in postoperative blood analysis or joint fluid cultures to indicate otherwise.
References 1. Regan WD, Reilly CD. Distraction arthroplasty of the elbow. Hand Clin 1993;9:719 –728. 2. Garland DE, Blum CE, Waters RL. Periarticular heterotopic ossification in head-injured adults: incidence and location. J Bone Joint Surg 1980;62A:1143–1146. 3. Wainapel SF, Rao PU, Schepsis AA. Ulnar nerve compression by heterotopic ossification in a head-injured patient. Arch Phys Med Rehabil 1985;66:512–514. 4. Keenan MAE, Kauffman DL, Garland DE, Smith C. Late ulnar neuropathy in the brain-injured adult. J Hand Surg 1988;13A:120 –124. 5. Morrey BF, Askew LJ, An KN, Chao EY. A biomechanical study of normal functional elbow motion. J Bone Joint Surg 1981;63A:872– 877. 6. Green DP, McCoy H. Turnbuckle orthotic correction of elbow-flexion contractures after acute injuries. J Bone Joint Surg 1979;61A:1092–1095. 7. Morrey BF, Adams RA, Bryan RS. Total replacement for post-traumatic arthritis of the elbow. J Bone Joint Surg 1991;73B:607– 612. 8. Knight RA, Van Zandt IL. Arthroplasty of the elbow: an end-result study. J Bone Joint Surg 1952;34A:610 – 618. 9. Morrey BF. Post-traumatic contracture of the elbow: operative treatment, including distraction arthroplasty. J Bone Joint Surg 1990;72A:601– 618. 10. McAuliffe JA, Wolfson AH. Early excision of heterotopic ossification about the elbow followed by radiation therapy. J Bone Joint Surg 1997;79A:749 –755. 11. Hastings H II, Graham TJ. The classification and treatment of heterotopic ossification about the elbow and forearm. Hand Clin 1994;10:417– 437. 12. Morrey BF, An KN, Chao EYS. Functional evaluation of the elbow. In Morrey BF, ed. The elbow and its disorders. Philadelphia: WB Saunders, 1985:73–91. 13. McGowan AJ. The results of transposition of the ulnar nerve for traumatic ulnar neuritis. J Bone Joint Surg 1950; 32B:293–301.
380 Viola and Hanel / Release of Elbow Contracture Associated With HO 14. Garland DE, Hanscom DA, Keenan MA, Smith C, Moore T. Resection of heterotopic ossification in the adult with head trauma. J Bone Joint Surg 1985;67A:1261–1269. 15. Husband JB, Hastings H II. The lateral approach for operative release of post-traumatic contracture of the elbow. J Bone Joint Surg 1990;72A:1353–1358. 16. Morrey BF. The elbow and its disorders. 2nd ed. Philadelphia: WB Saunders, 1993. 17. Conway JE, Jobe FW, Glousman RE, Pink M. Medial instability of the elbow in throwing athletes: treatment by repair or reconstruction of the ulnar collateral ligament. J Bone Joint Surg 1992;74A:67– 83. 18. Nestor BJ, O’Driscoll SW, Morrey BF. Ligamentous reconstruction for posterolateral rotatory instability of the elbow. J Bone Joint Surg 1992;74A:1235–1241.
19. Thompson HC III, Garcia A. Myositis ossificans: aftermath of elbow injuries. Clin Orthop 1967;50:129 –134. 20. Naraghi FF, DeCoster TA, Moneim MS, Miller RA, Rivero D. Heterotopic ossification. Orthopaedics 1996;19: 145–151. 21. Randelli G, Romano` CL. Prophylaxis with indomethacin for heterotopic ossification after Chiari osteotomy of the pelvis. J Bone Joint Surg 1992;74:1344 –1346. 22. Kjaersgaard-Andersen P, Nafei A, Teichert G, et al. Indomethacin for prevention of heterotopic ossification: a randomized controlled study in 41 hip arthroplasties. Acta Orthop Scand 1993;64:639 – 642. 23. Jupiter JB, Ring D. Operative treatment of post-traumatic proximal radioulnar synostosis. J Bone Joint Surg 1998; 80A:248 –257.
Elbow stiffness commonly complicates elbow trauma.1 Regan and Reilly1 outlined 3 factors that cause posttraumatic elbow stiffness: the high degree of articular congruity and conformity of the elbow joint predispose the elbow to limited motion after articular injury; the brachialis covers the anterior capsule, predisposing it to posttraumatic periarticular
From the Department of Orthopaedic Surgery, University of Washington, Harborview Medical Center, Seattle, WA. Received for publication February 26, 1998; accepted in revised form September 29, 1998. 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. Reprint requests: Douglas P. Hanel, MD, Department of Orthopaedic Surgery, University of Washington, Harborview Medical Center, Box 359798, 325 9th Ave, Seattle, WA 98104. Copyright © 1999 by the American Society for Surgery of the Hand 0363-5023/99/24A02-0014$3.00/0
370 The Journal of Hand Surgery
ossification; and mobilization after elbow injury is often delayed because it is difficult to achieve rigid internal fixation of comminuted elbow fractures. Furthermore, elbow stiffness may occur in spite of both early aggressive motion and prophylactic measures and periarticular ossification may lead to cubital tunnel syndrome.2– 4 Elbow stiffness is disabling because a wide range of motion (ROM) is necessary for basic everyday activities. Morrey et al’s5 electrogoniometer analysis demonstrated that an arc of elbow motion of 100° (range, 30° to 130°) and an arc of forearm rotation of 110° (range, 55° pronation to 55° supination) is necessary for an individual to perform 90% of his or her normal daily activities. A variety of treatments have been used for posttraumatic stiff elbows. Elbow arthrodesis causes severe functional impairment and should be considered
The Journal of Hand Surgery / Vol. 24A No. 2 March 1999 371
only as a last resort. Turnbuckle splinting does not significantly improve motion.6 Total elbow arthroplasty has a high perioperative complication rate and an unacceptable durability in patients younger than 60 years.7 Interposition arthroplasty has inconsistent results.8 Distraction arthroplasty, a technically demanding procedure, has produced good results but is unnecessary if articular surfaces are preserved.9 “Simple” elbow contracture release, described by Morrey9 as the removal of the causative abnormality without performing any procedures on the elbow articular surfaces, may be used to restore motion to the posttraumatic stiff elbow due to elbow capsule contracture and/or heterotopic ossification (HO). The morbidity associated with elbow release and HO excision is high. Complications include triceps rupture, elbow instability, and neurologic injury.9 Although simple elbow contracture release has good documented results, the timing of this procedure remains controversial.9,10 Many investigators advocate waiting 1 to 2 years before release.11 The desire to delay surgery until heterotopic bone has become metabolically quiescent must be balanced against the risks of progressive soft tissue contracture, potential articular cartilage destruction, and prolonged infirmity.11 In the past, surgical release of the posttraumatic stiff elbow was not performed until the patient’s bone scan and serum alkaline phosphatase were normal and the HO, as seen on plain radiographs, appeared mature. These prerequisites are typically met at 1 to 2 years after injury, with earlier intervention thought to predispose to HO recurrence. Recent reports suggest that good results may be obtained with early surgical release combined with postrelease nonsteroidal anti-inflammatory drugs or radiation therapy.10 We describe the results of a prospective treatment protocol involving early operative intervention (simple elbow contracture release), a short course of nonsteroidal anti-inflammatory drugs, and intensive therapy in patients with posttraumatic stiff elbows.
Materials and Methods Patient Selection From 1992 to 1995, 16 patients (18 elbows) with posttraumatic elbow stiffness associated with HO fulfilled the criteria for simple contracture release as described by Morrey.9 Fourteen patients (15 elbows) were treated within 32 weeks of the original injury; these cases were designated early and are the subject of this study. Elbows 1 and 2 (Tables 1 to 4) were
from a single patient. All patients were managed prospectively according to the contracture release protocol described below.
Indications Our criteria for elbow contracture release included (1) function-limiting elbow stiffness resulting from musculoskeletal trauma (limitation was defined as subjective complaints regarding the activities of daily living and at least 1 of 3 objective findings: a flexion/extension arc ,100°, a pronation/supination arc ,110°, or a flexion contracture .45°12), (2) radiographic evidence of union of fractures, (3) radiographic evidence of intact ulnohumeral joint articular surfaces, (4) HO in any stage of maturity on plain radiographs, and (5) stabilization of traumatic brain injury. Contractures following thermal injury were excluded.
Patient Demographics Patient characteristics are listed in Table 1. Most of the patients were young adult males (average age, 28 years) who sustained elbow trauma, were righthanded, and were treated without HO prophylaxis. Seven patients (53%) sustained traumatic brain injuries; of these patients, 6 were in a coma for at least 1 week (range, 1 to 5 weeks). Table 2 lists patient characteristics at the time of simple contracture release. Nine elbows had normal extremity neurologic examinations. Five elbows had cubital tunnel syndrome. In each of these cases, the diagnosis was established by subjective complaints and confirmed by Semmes-Weinstein monofilament sensibility testing (a minimum of 1-thickness difference in the ring and small finger compared with the radial 3 digits) and visible atrophy of the first dorsal interosseous muscle, when present. Electrodiagnostic studies were not routinely performed. Ulnar nerve lesions were classified according to McGowan13: 2 patients were grade I (decreased sensibility without ulnar intrinsic weakness), 3 were grade II (decreased sensibility with ulnar weakness and atrophy), and none were grade III (decreased sensation with ulnar intrinsic paralysis. One patient (no. 10) sustained an avulsion of the ulnar nerve resulting in a 10-cm defect at the time of injury. The 7 patients (8 elbows) who had sustained traumatic brain injuries were classified according to Garland et al14 at the time of contracture release. These investigators address both cognitive and physical impairment related to traumatic brain injury. In classifying our patients, elbow
372 Viola and Hanel / Release of Elbow Contracture Associated With HO
Table 1. Patient Demographics at the Time of Initial Injury Hand Dominance
Traumatic Brain Injury
Elbow
Age (yr)
Gender
Coma (Duration)
1
29
M
R
Yes
No
L Elbow DL, radial head FX, MCL rupture
2
29
M
R
Yes
No
Radial head FX, DRUJ DL
3
32
M
R
No
No
Radial head FX
4
21
M
R
Yes
Yes (1 wk)
5
38
M
R
Yes
Yes (4 wk)
6
35
M
R
Yes
Yes (1 wk)
7
28
M
R
Yes
Yes (3 wk)
8
20
M
R
No
No
Distal humerus and olecranon FX No orthopedic diagnosis, severe CHI Segmental ulnar FX Distal humerus, radial head, olecranon FX Elbow DL and olecranon FX
9
27
M
R
Yes
Yes (2 wk)
10
22
M
R
No
No
11
19
F
R
Yes
Yes (5 wk)
12 13
26 42
M M
R R
No No
No No
14
33
M
R
No
No
15 Summary/ average
21 28
M M (14) F (1)
R All R
No 53% yes
No No (8 patients) Coma (6 patients)
Initial Injury
Proximal radius and ulnar FX Medial elbow degloving with loss of ECU, MCL, and ulnar nerve; elbow dislocation Distal humerus FX Elbow DL Olecranon, coronoid, and radial head FX Type II Monteggia FX/ DL; type III radial head FX Olecranon FX Multiple diagnosis
Initial Management
HO Prophylaxis
Closed reduction, MCL repair, radial head prosthesis Radial head prosthesis, DRUJ pinning Radial head prosthesis ORIF of humerus and olecranon
No
None
No
ORIF of ulna
No
ORIF of humerus, radius, olecranon Closed reduction and ORIF of olecranon ORIF of radius and ulna Lateral free flap and static elbow external fixator; MCL reconstruction
No
ORIF of distal humerus Closed reduction ORIF of olecranon, coronoid, and radial head ORIF of ulna; radial head excision
No
ORIF Variable
No No (14) Indomethacin (1)
No No No
No No Yes (indocin 3 6 wk)
No No
No
DL, dislocation; FX, fracture; MCL, medial collateral ligament; DRUJ, distal radioulnar joint; CHI, closed head injury; ECU, extensor carpi ulnaris; ORIF, open reduction, internal fixation.
stiffness-related disability was not included as a parameter for the Garland et al classification because it resulted from musculoskeletal rather than neurologic injury. Five patients (6 elbows) were Garland grade I (no or mild cognitive and physical impairment), 1
was grade II (minimal cognitive and moderate physical disability), and 1 was grade IV (moderate to severe cognitive deficits with minimal to moderate physical disability).14 Fourteen elbows had no spasticity; 1 patient had mild spasticity. Patients were
The Journal of Hand Surgery / Vol. 24A No. 2 March 1999 373
categorized according to both the functional classification of elbow stiffness (Table 2) and the anatomic description of elbow HO of Hastings and Graham.11 The functional classification of Hastings and Graham stratifies patients according to the plane(s) of elbow stiffness: grade I (no patients), no functional limitation; grade IIA (6 patients), limited flexion/extension; grade IIB (1 patient), limited pronation/supination; grade IIC (5 patients), limited flexion/extension and pronation/supination; and grade III (3 patients), elbow ankylosis preventing either flexion/extension and/or pronation/supination. Hastings and Graham11 describe the location of HO as posterior elbow, anterior elbow, medial elbow, and lateral elbow. Among our patients, posterior involvement occurred in 12 elbows, anterior involvement occurred in 10, medial involvement occurred in 7, and lateral involvement occurred in 9. Preoperative pain was evaluated according to pain medication use at the time of contracture release. Each patient was asked to describe the use of pain medications to quantify his or her preoperative pain and to classify it as none, mild, moderate, or severe. If pain was controlled with nonopiate pain medications, the patient was classified as having mild pain. Patients with regular opiate use were classified as severe. The integrity of the ulnohumeral joint articular surfaces was evaluated by assessing the ulnohumeral joint space on plain anteroposterior and lateral elbow radiographs and was confirmed by visual inspection at the time of contracture release. Based on radiographs, the patient’s ulnohumeral articulations were classified as either normal or minimally altered. Minimally altered ulnohumeral articulations were those with joint space loss of less than 50% and/or articular incongruity less than 2 mm. Patients with more significant ulnohumeral articular abnormality were considered for reconstructive procedures rather than simple elbow contracture release. Radiographically, 12 elbows had normal ulnohumeral articular surfaces and 3 had minimal articular abnormalities. At surgery, all patients had normal or near-normal elbow articular surfaces except for patient 7, who had lost 50% of the distal humeral trochlear articular surface. This patient is included in the study because he met the radiographic criteria for inclusion and because his ulnar, radial head, and capitellar articular surfaces appeared normal. Each patient’s active elbow ROM before contracture release is listed in Table 2. The average preoperative ROM values as a function of Hastings and Graham’s11 classification are listed in Table 3.
The average time from injury to contracture release was 23 weeks (range, 10 –31 weeks). At the time of release, overall average preoperative flexion and extension were 99° (range, 85° to 135°) and 56° (range, 30° to 90°), respectively, and overall average preoperative pronation and supination were 41° (range, 0° to 90°) and 38° (range, 0° to 80°), respectively. Concomitant surgical procedures performed at the time of simple contracture release are listed in Table 2.
Technique The essential steps of the procedure are: 1. Remove all HO from the brachialis, triceps, and adjacent to the collateral ligaments. 2. Excise the anterior and posterior joint capsule, leaving the collateral ligaments intact (if possible). 3. Clear the coronoid and olecranon fossa of fat and synovium. 4. Decompress the ulnar nerve in the presence of cubital tunnel syndrome. 5. Correct any elbow instability that exists after resection.
The focus of each procedure was removal of HO and contracted capsule until normal elbow ROM was restored. Contracture release was considered complete only after a flexion/extension arc of 140°/10° and a pronation/supination arc of 80°/80° had been achieved. The triceps tendon was not detached from its insertion. The incision used to achieve the steps listed above was individualized based on the following factors: (1) location of previous elbow incision(s), (2) need for ulnar nerve decompression, and (3) plane(s) of elbow contracture. In cases in which a previous approach to the elbow had been performed, that approach was used as long as it allowed all the surgical objectives to be achieved. If ulnar nerve decompression was necessary, either a medial or posterior approach was used. In patients with restricted forearm rotation but preserved elbow flexion and extension, a lateral approach was favored. In patients with restricted flexion and extension but preserved forearm rotation, a medial approach was favored. In patients with loss of elbow motion in both planes, a posterior approach was favored. The above factors were considered in determining the best approach to use for each patient. When forearm pronation and supination was limited, a lateral extensile Kocher approach was used.15 The radial head and neck were exposed, the orbicu-
374 Viola and Hanel / Release of Elbow Contracture Associated With HO
Table 2. Patient Characteristics at the Time of Simple Contracture Release
Elbow
Ulnar Nerve Status
McGowan Class
TBI Status
HO Grade
HO Location
Preoperative Elbow Pain
Articular Surfaces
Time (wk)
1
Intact
NA
I
IIC
AE, LE
NA
Normal
27
2
Intact
NA
I
IIB
AE, LE
NA
Normal
27
3
Intact
NA
NA
IIC
AE, LE
Mild
Normal
17
4
Intact
NA
I
III
PE, AE, LE
Mild
Mild
20
5
Late palsy
I
IV
IIA
PE, ME
None
Normal
21
6
Late palsy
II
I
IIA
PE, AE, ME
None
Normal
26
7
Intact
NA
I
III
PE, AE, ME, LE
Mild
50% Distal humerus injury
22
8
Late palsy
II
NA
IIC
PE, AE, ME
NA
Normal
19
9
Late palsy
II
I
IIA
PE, ME
None
Normal
21
NA
NA
IIC
PE, AE, LE
Severe
Normal
10
11 12
Traumatic neurotomesis Intact Intact
NA NA
II NA
IIA IIA
None None
Normal Normal
31 18
13
Late palsy
I
NA
III
PE PE, AE, ME, LE PE, ME, LE
Mild
Mild
28
14
Intact
NA
NA
IIA
PE
Mild
Normal
31
15
Intact
NA
NA
IIC
AE, LE, PE
Mild
Normal
26
IIA (6) IIB (1) IIC (5) III (3)
PE (12) AE (10) ME (7) LE (9)
None (5) Severe (1) Mild (6)
Normal (12) Mild (3)
23
10
Summary/ Late palsies (5) Grade I (2) average Neurotomesis (1) Grade II (3) Intact (9) Grade III (0)
Grade I (6) Grade II (1) Grade IV (1)
Grade Grade Grade Grade
NA, not applicable; TBI, traumatic brain injury; AE, anterior elbow; LE, lateral elbow; PE, posterior elbow; ME, medial elbow; HW, hardware; ORIF, open reduction, internal fixation; BG, bone graft; MCL, medial collateral ligament; CPM, continuous passive motion.
laris and lateral collateral ligaments were preserved, and the HO bridging the proximal radioulnar joint was removed. If there was an associated loss of elbow flexion and extension, the exposure was carried proximally. The origin of the brachioradialis and the wrist extensors was elevated, the anterior joint capsule was excised, and the HO within the brachialis muscle was removed. Hypertrophic fat or synovium found in the coronoid fossa was then removed. If there was continued limitation in motion, the tri-
ceps muscle was elevated from the humerus, the posterior capsule was excised, and the fat or synovium was removed from the olecranon fossa. When fracture callus caused impingement of the olecranon, 1.5 cm of the tip of the olecranon was removed. If there was continued impingement, the olecranon fossa was widened with a ball tip burr. When forearm motion was normal and elbow flexion and extension were restricted, or in those cases in which cubital tunnel syndrome was present, the el-
The Journal of Hand Surgery / Vol. 24A No. 2 March 1999 375
Preoperative (°) Flexion Extension Pronation Supination 120
35
50
30
120
35
45
20
135
35
15
35
95
90
0
0
92
70
80
70
90
30
45
80
85
75
0
0
100
75
45
45
100
85
65
60
90
50
25
0
65 80
40 55
80 90
80 80
110
30
0
0
110
45
80
80
100
90
0
0
99
56
41
38
Procedure(s) at Time of Release Excision of HO, implant removal Excision of HO, implant removal Excision of HO, implant removal Excision of HO, HW removal, ORIF/BG ulna nonunion Excision HO, ulnar nerve transposition Excision HO, ulnar nerve transposition Excision HO, HW removal, ulnar nerve transposition, annular ligament reconstruction Excision HO, HW removal, ulnar nerve transposition Excision HO, ulnar nerve transposition Excision HO, MCL reconstruction Excision HO Excision HO, ulnar nerve transposition Excision HO, HW removal, ulnar nerve transposition Excision HO, removal of radial head implant Excision HO, HW removal, ulnar nerve transposition Excision of HO (15) Implant removal (7) Ulnar nerve transposition (8)
bow was approached medially.16 Previous surgical incisions were used to mobilize the skin until the medial epicondylar region was exposed. The ulnar nerve was identified in the mid-arm and dissected distally into the proximal forearm. In 4 cases, the ulnar nerve was completely encased in HO at the cubital tunnel. The nerve was released by removing the HO with ronguers. The medial intermuscular septum of the arm was removed and the origin of the flexor–pronator mass was elevated from the medial
Fixator
Surgical HO Prophylaxis Postoperative Approach(es) After Release CPM
None
Lateral
Indocin 3 5 d
None
None
Lateral
Indocin 3 5 d
None
None
Lateral
Indocin 3 5 d
6 wk
None
Posterior
Indocin 3 5 d
6 wk
None
Medial
Indocin 3 5 d
1d
None
Medial
Indocin 3 5 d
4d
6 wk
Medial and posterior
Indocin 3 5 d
6 wk
None
Posterior
Indocin 3 5 d
5d
None
Medial
Indocin 3 5 d
5d
None
Medial
Indocin 3 6 wk
None
None None
Lateral Medial
Indocin 3 5 d Indocin 3 5 d
6 wk 5d
None
Posterior
Indocin 3 5 d
5d
None
Lateral
None
None
None
Medial
None
None
Fixator (1) None (14)
14 Single 1 Double
None (2) 5 d (12) 6 wk (1)
Median 55d
epicondyle of the humerus. This provided direct vision of the brachialis muscle and the anterior joint capsule. Similar to the lateral approach, the anterior capsule was excised and the heterotopic bone removed. While retracting the ulnar nerve with a 1-inch Penrose drain, the posterior joint capsule was entered through a subperiosteal dissection. Once inside the joint, the capsule was removed and the olecranon fossa cleared of fat, synovium, and compromising fracture callus. After the surgical release
376 Viola and Hanel / Release of Elbow Contracture Associated With HO
Table 3. Preoperative and Postoperative Average Ranges of Motion and Average Arcs of Motion as a Function of Hastings and Graham’s Classification11 Classification Grade IIA (F/E , 100, P/S . 110) Grade IIB (F/E . 100, P/S , 110) Grade IIC (F/E , 100, P/S , 110) Grade III (F/E and/or P/S ankylosis) Total
No. of Patients
Flexion
Extension
Pronation
Supination
F/E Arcs
P/S Arcs
6
85/144
56/9
72/86
74/85
29/135
146/171
1
135/140
35/25
45/45
20/90
100/115
65/135
5
106/139
49/21
34/80
28/73
57/118
62/153
3 15
97/150 98/140
65/24 54/20
0/72 41/77
0/63 38/75
32/126 44/120*
0/135 79/152*
Data are presented as average preoperative/postoperative values, in degrees. F/E, flexion/extension; P/S, pronation/supination. *p , .05.
had been completed, elbow stability was assessed. Isolated medial or lateral ligamentous instability was treated with autograft reconstruction as described by Conway et al17 or Nestor et al,18 respectively. One patient with both varus and valgus instability was stabilized with a hinge fixator for 6 weeks.1 In those cases approached medially, an anterior submuscular transposition of the ulnar nerve was completed. Submuscular, rather than subcutaneous, transposition was performed because nearly all the necessary steps, including flexor–pronator mass elevation, were already completed during HO and capsule resection. Care was taken to identify and excise any fascial bands that would compromise a smooth passage of the nerve into the forearm. The origin of the flexor–pronator mass was reconstructed with stout nonabsorbable sutures passed through drill holes in the medial epicondyle. Suction drains were placed anterior and posterior to the elbow joint and the skin was closed with interrupted sutures.
Postoperative Care One patient required stabilization with the Brigham distraction fixator. Thirteen patients were treated with immediate postoperative continuous passive mobilization (CPM). Of these 13 patients, 6 were treated with CPM for 4 days postoperatively, 4 patients continued CPM at home for a total of 6 weeks, and 3 patients did not tolerate CPM beyond the immediate time in recovery room because of pain and inconvenience. The latter 3 patients were treated with active and passive elbow motion 4 times per day. The median number of days of CPM use was 5 days.
After surgery, pain was controlled with constant perfusion of regional anesthesia through an axillary catheter and patient-controlled analgesic pumps for 3 days. This was followed by oral analgesics. Patients were instructed in elbow ROM exercises and were encouraged to perform them at least 4 times a day for 6 months. Gentle active and passive exercises were instituted on the first postoperative day. Aggressive strengthening exercises were postponed until the third week after surgery. Splints, slings, and other forms of immobilization were strongly discouraged while the patients were awake. Patients often lost elbow extension early in their postoperative course; such cases were treated with night extension splinting for 6 weeks. Indomethacin (25 mg orally or rectally 3 times per day) and sulcralfate (1 g orally twice a day) were prescribed for 5 postoperative days. Radiation was not used.
Patient Evaluation Each patient was examined weekly for 3 weeks and then every 12 weeks for at least 2 years. At each visit, active elbow ROM, Semmes-Weinstein sensibility, and gross motor strength were documented. Each case was analyzed by review of medical records, radiographs, physical examinations, and patient interviews.
Statistical Analysis Statistical analyses of the preoperative and postoperative flexion/extension and pronation/supination ROM arcs were compared using the paired Student’s
The Journal of Hand Surgery / Vol. 24A No. 2 March 1999 377
t-test. Analysis was performed using Systat 7.0 (SPSS, Inc, Evanston, IL). The null hypothesis was that the preoperative and postoperative ROM arcs would be similar. Statistical significance was established at p , .05.
Results Postoperative results are summarized in Tables 3 and 4. Table 4 also lists the results and postoperative patient characteristics. The average duration of follow-up was 115 weeks (range, 79 –156 weeks). Average postoperative flexion and extension were 140° (range, 130° to 145°) and 20° (range, 0° to 37°), respectively. The flexion/extension arc improved from an average of 43° before surgery to an average of 120° at 2 years (p , .05) (Table 3). Average postoperative pronation and supination were 77° (range, 45° to 90°) and 75° (range, 25° to 90°), respectively (Table 4). The pronation/supination arc improved from an average of 79° before surgery to an average of 152° at 2 years (p , .05) (Table 3). Three patients (nos. 7, 14, and 15) who did not achieve the ROM thought to be necessary for functional elbow motion (100° flexion/extension arc and 110° pronation/supination arc) were all very close to these goals. All 3 of these patients had a flexion/ extension arc of 95°. Only 1 patient had limitation of pronation/supination. Patient 7 had an 85° pronation supination arc. There were no intraoperative complications. Three postoperative complications occurred in 2 patients. The first patient (no. 4) had tenuous elbow soft tissue coverage at the time of contracture release. With increased activity after the contracture release, the soft tissue envelope became increasingly unstable with multiple episodes of skin ulceration. This was treated with a free flap soft tissue reconstruction 15 months after contracture release. The second patient (no. 7) had 2 complications. First, he developed an olecranon pin track infection 3 weeks after the contracture release (he had been treated with a hinge fixator). This resolved after 10 days of oral antibiotics. The pin was not exchanged. Second, serial radiographs during the first postoperative year showed radiographic resorption of the capitellum. The patient’s white blood cell count and erythrocyte sedimentation rate were normal. As a result of this bone loss, 2 K-wires, which had previously been buried in the capitellum, became loose within the elbow joint. An arthrotomy was required to retrieve these wires (intraoperative cultures were negative). The condi-
tion of the elbow was clinically compatible with Charcot arthropathy without instability. He has pain with heavy lifting but continues to work as a carpenter. Hand dominance was unchanged in 13 patients; 1 patient (no. 10) became ambidextrous to compensate for the permanent ulnar nerve lesion he sustained at the time of the original injury. There were no nerve injuries or nerve palsies resulting from the contracture release. Semmes-Weinstein monofilament sensibility of 2.83 to 3.22 was found in 14 elbows, including those in the 5 patients with preoperative cubital tunnel syndrome. Patient 10 had a permanent ulnar nerve injury as previously described. Six months after elbow release he underwent tendon transfers to compensate for intrinsic motor deficits and debulking of a latissimus dorsi free flap, which provided primary coverage of the open elbow fracture dislocation.
Discussion Like other published series of elbow contracture releases, this report describes a mixed patient population. However, most were young adult males, frequently with traumatic brain injuries, who sustained significant elbow trauma and had their initial injury managed without HO prophylaxis. The time from injury to HO excision was also variable; however, all contractures were released within an average of 23 weeks (range, 24 to 78 weeks), earlier than in other studies.2,9,10,14,19,20 The extent and distribution of HO were determined by plain radiographs. Advanced imaging studies, such as computed tomography and magnetic resonance imaging, were not obtained for HO localization, as we found plain radiographs adequate for this purpose. All patients demonstrated statistically and clinically significant improvement in elbow motion (p , .05) and elbow function after 2 years. Furthermore, those with ulnar nerve palsies resulting from compression recovered full ulnar nerve function. We attribute these results to thorough operative resection of HO and capsular excision, submuscular ulnar nerve transfer, and aggressive, unlimited, postoperative active and passive motion. The focus of each release was complete resection of HO and capsule so that intraoperative ROM was as close to normal as possible. In addition, all elbows were stable after surgery due to either preservation of the collateral and annular ligaments, ligament reconstruction, or hinge fixator stabilization. This stability allowed un-
87 87 87 100 149
93 91 104
97 134 106
104 139 79 156 80 115
1 1 2 3 4
5 6 7
8 9 10
11 12 13 14 15 Summary/ average
140 140 145 140 140 140
140 130 140
145 145 130
140 140 140 135 145
Flexion (°)
NA, not available; R, right.
Follow-up (wk)
Elbow
10 5 0 45 45 20
20 10 12
10 10 35
35 35 25 15 37
Extension (°)
85 85 90 80 80 77
70 80 80
85 80 60
90 90 45 80 50
Pronation (°)
90 75 90 80 45 75
80 80 60
90 80 25
70 70 90 80 80
Supination (°)
Normal Normal Normal Normal Normal All normal
Normal Normal Normal
Normal Normal Normal
Normal Normal Normal Normal Normal
Stability
None None None Mild None 10 None 3 mild 2 NA
NA NA None
None None Mild
None None None None Mild
Pain
None None None None None 2 Patients
None None None None Elbow free flap coverage None None Pin track infection, pin migration, Charcot elbow None None None
Complications
Table 4. Postoperative Results and Characteristics
R R R R R 13 Unchanged 1 Bimanual
R R Bimanual
R R R
R R R R R
Hand Dominance
Intact Intact Intact Intact Intact 14 Intact 1 Unchanged
Intact Intact Unchanged
Intact Intact Intact
Intact Intact Intact Intact Intact
Neurologic Status
None None Lateral flap debulking, tendon transfers None None None None None 3 Patients
None None None None Elbow free flap coverage None None Pin removal
Additional Surgery
378 Viola and Hanel / Release of Elbow Contracture Associated With HO
The Journal of Hand Surgery / Vol. 24A No. 2 March 1999 379
restricted postoperative motion. In a frequently quoted reference, Thompson and Garcia19 assert that passive elbow ROM causes periarticular elbow HO formation and elbow stiffness after elbow injury. A review of this report reveals that their analysis was retrospective and, as such, subject to selection bias. Their conclusion that passive motion during convalescence should never be used is based on the fact that those patients developing stiff elbows “received passive stretching by therapists or by the use of weights.”17 Who else but patients with limited elbow motion would be subjected to passive elbow stretching? Until there is a prospective study comparing patients with elbow injuries treated with passive stretching to those treated without passive stretching, the relationship between passive elbow motion and HO formation will remain unclear. We are unsure of the role of CPM. Our goal was to restore elbow motion. To that end, we used CPM for the initial 5 days after surgery, and in those patients who did not tolerate CPM, we used aggressive active and passive ROM exercises. Our results demonstrate significant gains in motion. The number of patients and the study protocol do not allow statements of statistical significance with regard to the effects of CPM versus exercise alone. The role of prophylactic measures for the prevention or recurrence of HO about the elbow remains anecdotal. While many investigators espouse the utility of indomethacin, the length of its use remains arbitrary.20 –22 We chose to limit indomethacin use to 5 days, that period of time when patient compliance is maximal. Similarly, the role of radiation therapy is unclear. McAuliffe and Wolfson10 recently indicated that radiation is beneficial after early excision but they provided no control group for comparison. It is our belief that their results reflect the attention to operative detail and the rehabilitation efforts necessary for satisfactory results in this difficult group of patients without relation to radiation therapy. This belief is supported by Jupiter and Ring’s23 presentation of excision of ectopic bone to treat synostosis about the proximal radioulnar joint. These investigators had only 1 recurrence among 18 limbs despite the lack of adjuvant radiation therapy or anti-inflammatory medication,23 raising even further the questionable role of indomethacin in our patients. The extent of HO recurrence in our patients is unknown. Since the functional gains of our patients were impressive and maintained, radiographs were not routinely obtained. The complication rate of posttraumatic elbow con-
tracture release is high.9,15 Complications include triceps avulsion, ulnar nerve paresthesias, deep infection, aseptic resorption of the distal humerus, and skin slough.9,15 Of our 14 patients (15 elbows), only 2 had complications. Patient 7 had an unusual complication: a Charcot elbow. Morrey9 described a similar phenomenon. The etiology of this process is unclear. In our case, we believe that this resulted from excessive soft tissue dissection in the presence of a healing capitellum fracture. The soft tissue resection, although necessary for release, appears to have rendered the capitellum avascular. We do not believe that this was caused by a septic process. There was no indication in postoperative blood analysis or joint fluid cultures to indicate otherwise.
References 1. Regan WD, Reilly CD. Distraction arthroplasty of the elbow. Hand Clin 1993;9:719 –728. 2. Garland DE, Blum CE, Waters RL. Periarticular heterotopic ossification in head-injured adults: incidence and location. J Bone Joint Surg 1980;62A:1143–1146. 3. Wainapel SF, Rao PU, Schepsis AA. Ulnar nerve compression by heterotopic ossification in a head-injured patient. Arch Phys Med Rehabil 1985;66:512–514. 4. Keenan MAE, Kauffman DL, Garland DE, Smith C. Late ulnar neuropathy in the brain-injured adult. J Hand Surg 1988;13A:120 –124. 5. Morrey BF, Askew LJ, An KN, Chao EY. A biomechanical study of normal functional elbow motion. J Bone Joint Surg 1981;63A:872– 877. 6. Green DP, McCoy H. Turnbuckle orthotic correction of elbow-flexion contractures after acute injuries. J Bone Joint Surg 1979;61A:1092–1095. 7. Morrey BF, Adams RA, Bryan RS. Total replacement for post-traumatic arthritis of the elbow. J Bone Joint Surg 1991;73B:607– 612. 8. Knight RA, Van Zandt IL. Arthroplasty of the elbow: an end-result study. J Bone Joint Surg 1952;34A:610 – 618. 9. Morrey BF. Post-traumatic contracture of the elbow: operative treatment, including distraction arthroplasty. J Bone Joint Surg 1990;72A:601– 618. 10. McAuliffe JA, Wolfson AH. Early excision of heterotopic ossification about the elbow followed by radiation therapy. J Bone Joint Surg 1997;79A:749 –755. 11. Hastings H II, Graham TJ. The classification and treatment of heterotopic ossification about the elbow and forearm. Hand Clin 1994;10:417– 437. 12. Morrey BF, An KN, Chao EYS. Functional evaluation of the elbow. In Morrey BF, ed. The elbow and its disorders. Philadelphia: WB Saunders, 1985:73–91. 13. McGowan AJ. The results of transposition of the ulnar nerve for traumatic ulnar neuritis. J Bone Joint Surg 1950; 32B:293–301.
380 Viola and Hanel / Release of Elbow Contracture Associated With HO 14. Garland DE, Hanscom DA, Keenan MA, Smith C, Moore T. Resection of heterotopic ossification in the adult with head trauma. J Bone Joint Surg 1985;67A:1261–1269. 15. Husband JB, Hastings H II. The lateral approach for operative release of post-traumatic contracture of the elbow. J Bone Joint Surg 1990;72A:1353–1358. 16. Morrey BF. The elbow and its disorders. 2nd ed. Philadelphia: WB Saunders, 1993. 17. Conway JE, Jobe FW, Glousman RE, Pink M. Medial instability of the elbow in throwing athletes: treatment by repair or reconstruction of the ulnar collateral ligament. J Bone Joint Surg 1992;74A:67– 83. 18. Nestor BJ, O’Driscoll SW, Morrey BF. Ligamentous reconstruction for posterolateral rotatory instability of the elbow. J Bone Joint Surg 1992;74A:1235–1241.
19. Thompson HC III, Garcia A. Myositis ossificans: aftermath of elbow injuries. Clin Orthop 1967;50:129 –134. 20. Naraghi FF, DeCoster TA, Moneim MS, Miller RA, Rivero D. Heterotopic ossification. Orthopaedics 1996;19: 145–151. 21. Randelli G, Romano` CL. Prophylaxis with indomethacin for heterotopic ossification after Chiari osteotomy of the pelvis. J Bone Joint Surg 1992;74:1344 –1346. 22. Kjaersgaard-Andersen P, Nafei A, Teichert G, et al. Indomethacin for prevention of heterotopic ossification: a randomized controlled study in 41 hip arthroplasties. Acta Orthop Scand 1993;64:639 – 642. 23. Jupiter JB, Ring D. Operative treatment of post-traumatic proximal radioulnar synostosis. J Bone Joint Surg 1998; 80A:248 –257.