J Shoulder Elbow Surg (2015) 24, 647-654

www.elsevier.com/locate/ymse

Box-loop ligament reconstruction of the elbow for medial and lateral instability Patrick R. Finkbone, MD, Shawn W. O’Driscoll, MD, PhD* Department of Orthopedic Surgery, Mayo Clinic, Rochester, MN, USA Background: Elbow instability remains a challenging surgical problem. Most commonly, isolated reconstructions of the medial collateral ligament or lateral collateral ligament are performed; however, on occasion, there can be deficiency that requires reconstruction of both ligaments. The senior author has developed a method to reconstruct both the medial and lateral collateral ligaments using 1 graft. This technique uses a ‘‘box-loop’’ design, whereby the donor tendon is passed through the humerus and ulna and tied back to itself, creating a loop. Materials and methods: Fourteen cases with mean follow-up of 64 months were reviewed. Nine patients returned to the clinic and were evaluated both clinically and radiographically. An additional 5 patients participated by phone questionnaire. Results: Average follow-up time was 64 months (range, 19-109 months). According to the Summary Outcome Determination given by the patients, 7 elbows were normal or nearly normal, 4 were greatly improved, 2 were improved, and 1 was worse compared with before surgery. The Summary Outcome Determination score average was 7 (range, 2 to 10). American Shoulder and Elbow Surgeons scores (including both clinic patients and phone questionnaire patients) ranged from 36 to 100, with an average of 81; of 14 patients, 8 had an American Shoulder and Elbow Surgeons self-satisfaction score of 10. The average Quick Disabilities of the Arm, Shoulder, and Hand score was 13 (range, 0-64). The average Mayo Elbow Performance Index score was 88 (range, 60-100), with 4 excellent (90-100), 3 good (75-89), and 3 fair (60-74) results and no poor results. Discussion: This technique was found to have excellent midterm results. Compared with separate medialand lateral-sided reconstruction, there is simplification of bone tunnel formation as well as graft fixation. Level of evidence: Level IV, Case Series, Treatment Study. Ó 2015 Journal of Shoulder and Elbow Surgery Board of Trustees. Keywords: Elbow instability; box-loop reconstruction; medial collateral ligament; lateral collateral ligament

The Mayo Clinic Institutional Review Board that convened on October 22, 2009, approved this project: Protocol No. 09-001432. Consent forms were signed by all participating patients. *Reprint requests: Shawn W. O’Driscoll, PhD, MD, Department of Orthopedic Surgery, Mayo Clinic, Rochester, MN 55905, USA. E-mail address: [email protected] (S.W. O’Driscoll).

Reconstructive techniques have been described to address each of the forms of instability, including the medial collateral ligament (MCL) or the lateral collateral ligament (LCL) complex.7-11,14,17 Gross instability of the elbow requires reconstruction of both ligaments.15,18,19 A technique has been described by van Riet et al18,19 to

1058-2746/$ - see front matter Ó 2015 Journal of Shoulder and Elbow Surgery Board of Trustees. http://dx.doi.org/10.1016/j.jse.2014.12.008

648

Table I

Patient demographics

Case Age at Side surgery

Job

Comorbidities

28

Left

Merchant None Marine officer Right) Flight student None (U.S. Navy)

2

25

3

48

Right) Nurse

4

68

Left

Homemaker

5

56

Left

6

38

Left

Livestock producer/ loan officer Registered nurse

7

50

Left

8

18

Left

9

49

10

26

Right) Manual laborer

None

11

51

Right) Farmer

None

12

20

Left

Sales

None

13

18

Left

None

14

32

Left

Student (athlete) Teacher

Engineer

Material handler Right) Accountant

Prior surgeries

Contracture, HO after closed head injury Chronic instability with deficient coronoid and radial head

1dAnterior capsulectomy and HO and capsular excision, ulnar nerve HO excision transposition Revision radial head arthroplasty, 4dORIF with radial head coronoid osteotomy and ORIF, replacement, radial head implant ulnar nerve neurolysis removal and capsular release, ulnar nerve transposition and MUA, MCL reconstruction 1dORIF of medial epicondyle Excision of medial epicondyle nonunion nonunion, ulnar nerve decompression 1dRadial head replacement Revision radial head arthroplasty

None

Chronic instability with medial epicondyle nonunion, ulnar neuropathy Chronic instability, failed radial Smoker, head replacement fibromyalgia, history of lung cancer None Chronic instability with coronoid nonunion Diabetes mellitus None

None None

None

Chronic instability, overstuffed radial head prosthesis with erosion of capitellum Chronic instability after terrible triad injury, ulnar neuropathy Chronic instability, capitellar nonunion Contracture, radioulnar synostosis after proximal radius and ulna fracture, ulnar neuropathy Chronic instability with medial epicondylar avulsion, ulnar neuropathy Chronic varus and valgus instability, prior ulnar neuropathy, distal radius malunion Chronic instability after multiple instability episodes Chronic instability with coronoid nonunion Chronic instability with medial epicondylar nonunion

Concomitant surgery at time of box-loop reconstruction

0

None

2dRadial head replacement, LCL repair and capsular imbrication

Removal of radial head prosthesis, ulnar nerve decompression

1dORIF of radial head and LCL repair, external fixator placement (in place  6 weeks) 0

ORIF coronoid nonunion, HO and capsular excision, ulnar nerve transposition Hemicap reconstruction of capitellum

2dORIF ulna, delayed skin grafting

HO and capsular excision, radial head arthroplasty, ulnar nerve transposition Ulnar nerve decompression

0

2dUlnar nerve transposition, joint debridement and MCL repair

Distal radius osteotomy/correction

0

None

0

ORIF coronoid nonunion, ulnar nerve transposition Excision of medial epicondyle nonunion, ulnar nerve decompression

0

HO, heterotopic ossification; ORIF, open reduction and internal fixation; MUA, manipulation under anesthesia; MCL, medial collateral ligament; LCL, lateral collateral ligament. ) Dominant side.

P.R. Finkbone, S.W. O’Driscoll

1

Preoperative diagnosis

Box-loop ligament reconstruction

649

Figure 1 Box-loop technique. After a 3.2-mm drill hole at the origin of the ligaments (isometric point at the axis of rotation) was created, this process was repeated on the ulna, making the insertions in the middle of the sublime tubercle and the tubercle on the supinator crest. An allograft (most commonly plantaris) is passed through the humerus, looped through the ulna, and again through the humerus and ulna once again (A). After all passes have been completed, the capsule and residual ligamentous tissue are then closed, followed by tensioning and tying of the graft (B). Lateral (C) and medial (D) views demonstrate the placement of the drill holes and the graft. (By permission of Mayo Foundation for Medical Education and Research. All rights reserved.)

reconstruct both the MCL and LCL using 1 graft. The senior author has also independently developed a similar technique. This technique uses a ‘‘box-loop’’ design, whereby the donor tendon is passed through the humerus and ulna and tied back on itself, creating 1 continuous loop. The purpose of the present study is to describe the box-loop technique and to report early to midterm results with the technique.

Materials and methods This study is a retrospective case series of 18 box-loop reconstructions performed by the senior author (S.O.D.) between December 2003 and May 2009. Patients were contacted by phone and asked to participate in the study either by returning to the clinic or completing questionnaires by phone or mail-in. In all, 14

patients agreed to participate (2 were lost to follow-up and 2 declined). Nine patients agreed to come to the clinic, whereas 5 were willing to participate only by phone and correspondence. Patients were included in the study regardless of additional simultaneous surgery performed (i.e., radial head replacement, fracture fixation) or prior diagnoses, surgeries, or injuries. Twelve of the patients had persistent instability after trauma. The other 2 patients had severe contractures with heterotopic ossification (HO), and the planned operation to regain motion would cause gross instability of the elbow. Several of the patients had prior surgery on the elbow (Table I).

Surgical technique The box-loop technique is performed through separate medial and lateral incisions, with care taken to avoid injury to any cutaneous nerve branches. Starting on the medial side, identify the origins

650

P.R. Finkbone, S.W. O’Driscoll

Figure 2

Preoperative (A and C) and postoperative (B and D) radiographs of case 3.

and insertions of the MCL as follows. The origin is on the anterior inferior margin of the epicondyle. With the elbow flexed 90 , the anterior bundle of the MCL travels in a 45 direction from the anterior inferior margin of the epicondyle to the sublime tubercle of the ulna. The origin is the point at which that line intersects the anterior inferior epicondyle. The insertion is on the sublime tubercle of the ulna, which is a bump that can usually be readily palpated immediately anterior to the ulnar nerve. The anterior bundle of the MCL is exposed by splitting the common flexorpronator tendon origin in that line from the anterior inferior margin of the epicondyle to the sublime tubercle. A 3.2-mm drill hole is made, just a few millimeters deep, into the anterior inferior margin of the epicondyle and also into the ulna immediately distal to the sublime tubercle. Do not drill deeper than a few millimeters at this point as the line of drilling must be determined after identifying the corresponding points on the lateral side. The deep dissection on the lateral side is in the anconeus interval, between the extensor carpi ulnaris and anconeus muscletendon units. The tubercle on the supinator crest is exposed by subperiosteal elevation of the anconeus muscle. The interosseous recurrent vessels are cauterized. The LCL complex and capsule are then split longitudinally in a line from the epicondyle toward the tubercle on the supinator crest. The isometric point on the lateral condyle is determined by visually estimating the center of

the circle that is partially defined by the circular shape of the capitellum when it is viewed directly from the lateral side. This is facilitated by placing small Hohmann retractors on the proximal and distal margins of the capitellum, taking the midpoint between the 2. That midpoint should be equidistant from the anterior margin of the capitellum as from the proximal and distal margins. A 3.2-mm drill hole is made, just a few millimeters deep, into the isometric point on the lateral condyle and also into the ulna adjacent and immediately posterior to the tubercle on the supinator crest. Place the targeting end of a targeted drill guide into the previously drilled hole on the medial epicondyle, then sit the drill bit into the hole at the isometric point on the lateral condyle and drill about 60% of the distance from lateral to medial through the humerus. Do the same thing with the ulnar holes, noting that the angle of this tunnel is unexpectedly sharp because of the offset between the ulnar insertions on the medial and lateral sides. Once having drilled 60% of the way across the humerus and the ulna from the lateral side, do the same from the medial side with each hole respectively. It should be possible to advance the drill through the partial tunnel and come out on the lateral side, confirming completion of the tunnels. An allograft is then prepared using a No. 2-0 FiberWire (Arthrex, Naples, FL, USA) stitch in a Krackow fashion on each end. If a small-diameter graft (such as our preferred graft, a

Box-loop ligament reconstruction

651

Figure 3 Subjective Outcome Determination (SOD) score. Clinical outcomes were assessed according to patient-reported SOD score. (By permission of Mayo Foundation for Medical Education and Research. All rights reserved.)

plantaris allograft) is used, it is passed through the humerus looped through the ulna twice. If a larger diameter graft is used, such as a split semitendinosus or gracilis, it is passed through the humerus and ulna only once. After all passes have been completed, the capsule and residual ligamentous tissue are imbricated, followed by tensioning and tying of the graft. Finally, the graft is sutured to the underlying ligamentous tissues (Figs. 1 and 2). Postoperatively, the patients are placed in a cast for 3 weeks. This is followed by 3 weeks in a removable posterior splint, removing it 2 to 5 times per day for gentle overhead active and active-assistive range of motion exercises. The splint is continued for 6 weeks more for protection at night and when the elbow is at risk. Follow-up evaluation for the patients returning to the clinic included taking a history for subjective instability or complications of surgery, physical examination, radiographs, and evaluation by several different performance scores (see later). Examination parameters included range of motion, lateral pivot shift test, posterolateral rotatory drawer test, and varus and valgus stresses in full extension and 30 of flexion.16 These physical examination findings were compared with preoperative physical examination findings. Patient outcomes were assessed by the Mayo Elbow Performance Index (MEPI),13 American Shoulder and Elbow Surgeons (ASES) score,12 Quick Disabilities of the Arm, Shoulder, and Hand (QuickDASH) questionnaire,1 and Summary Outcome Determination (SOD) score2,3,5 (Fig. 3).

The patients who were willing to participate only by phone or mail-in were assessed by the MEPI, ASES, QuickDASH, and SOD scores. The questionnaires were sent to the patients and then either completed over the phone or completed independently by the patients and mailed back in. The MEPI score was reported only if the patients also supplied a picture of their elbow in full extension and full flexion (from a perfect side view) for confirmation of their range of motion.6

Results Average follow-up time was 64 months (range, 19109 months). No patients who returned to the clinic had symptoms or physical examination signs of instability or any radiographic evidence of instability. Curiously, 3 of these patients responded on the MEPI stating that their elbow was ‘‘moderately unstable’’ but denied instability when questioned specifically about it by the examiner, and their elbows were stable on examination. Two of the patients who had completed phone questionnaires claimed to have ‘‘moderate instability’’ of their elbow but denied any frank instability episodes. One patient had slight laxity to both varus and valgus stress at the 3-month follow-up appointment. The

652

Table II Case

Physical examination and radiographic results

Follow-up (months)

Preoperative flexion/ extension

Preoperative pronation/ supination

Postoperative flexion/ extension

Postoperative pronation/ supination

F/E

P/S

F/E

P/S

Increase in F/E arc

Varus

Stability (preoperative/ postoperative)

Posterolateral rotatory

Radiographs

Valgus

1 2 3

66 19 71

85/65 145/10 130/15

0/0 70/75 85/85

135/15 150/0 145/5

90/90 90/90 90/90

100 15 25

0/0 3/0 Pos/0

0/0 3/0 ND/0

4 5 6 7 8 9 10 11 12 13 14

44 46 20 56 29 87 62 109 96 94 95

150/0 135/10 105/35 90/60 155/-5 85/70 140/0 130/25 130/20 135/10 150/0

90/90 90/90 90/90 30/30 90/90 0/0 90/90 25/10 70/70 80/85 85/85

150/0 147/3 135/15 120/33 155/ 5 105/35 150/0

80/90 80/90 80/80 45/30 90/90 80/30

0 19 40 57 0 55 10

1/0 2 (EUA)/0 2 (EUA)/0 Sublux/0 3/0 0/0 1 2 1 (EUA) 2 ND

2/0 2 (EUA)/0 2 (EUA)/0 Sublux/0 3/0 0/0 3 2 1 (EUA) 3 ‘‘Unstable’’

0/0 3/0 3/0 3/0 3/0 Pos/0 Sublux/0 3/0 0/0 1 1 3 3 ‘‘Unstable’’

Moderate DJD Moderate DJD Negative (residual lateral ossicle) Mild DJD Moderate DJD Moderate DJD, HO Moderate DJD Negative Advanced DJD, residual HO

Pos, examination finding documented as positive but not graded; ND, not documented in the medical record; EUA, examination under anesthesia; Sublux, joint was chronically subluxed; DJD, degenerative joint disease; HO, heterotopic ossification. Cases 10 to 14 were not evaluated in the clinic; examination listed is only preoperative examination. Case 10 submitted photographs, which allowed calculation of flexion and extension. Instability grading: 0, none; 1, mild laxity with end point; 2, moderate laxity with end point; 3, gross instability.

P.R. Finkbone, S.W. O’Driscoll

Box-loop ligament reconstruction elbow was immobilized in a long arm cast for 6 weeks, and the examination at 6-month follow-up showed the elbow to be stable. Range of motion for this patient at that time was reported as full flexion but lacking 15 of full extension. Range of motion improved or was the same (as preoperative) in all patients. The average improvement in range of motion (flexion-extension arc) from preoperative to postoperative was 32 (see Table II for detailed results of range of motion, stability, and postoperative radiographs for each patient). Strength was normal (5 of 5) in all patients for elbow flexion and extension; 1 patient had slight weakness (4 of 5) in pronation and supination. Functional outcomes (as determined by performance scores and questionnaires) were obtained for all 14 patients. This was done by phone questionnaire for the 5 of the 14 patients who were not evaluated in the clinic (Table II). The functional outcomes for the box-loop procedure were exceptional for the majority of patients (Table III). According to the SOD given by the patients, 7 elbows were normal or nearly normal, 4 were greatly improved, 2 were improved, and 1 was worse compared with before surgery. The SOD score average was 7 (range, 2 to 10). ASES scores (including both clinic patients and phone questionnaire patients) ranged from 36 to 100, with an average of 81; of 14 patients, 8 had an ASES self-satisfaction score of 10. The average QuickDASH score was 13 (range, 0-64). The average MEPI score was 88 (range, 60-100), with 4 excellent (90-100), 3 good (75-89), and 3 fair (60-74) results and no poor results (Tables III and IV). Pain had improved in most patients, with 4 patients reporting no pain, 6 patients with mild pain, 4 patients with moderate pain, and no patients with severe pain. One patient with an ASES score of 36 had painful arthritis but thought she had gained significant improvement in function after the surgery. She had been having the arthritic pain before surgery. Another patient thought that her elbow was the same or slightly worse after the surgery; this patient was only able to complete a phone questionnaire and unable to come into the clinic. Radiographs showed that all ulnohumeral joints were congruent in appearance without any signs of instability. There were mild to moderate degenerative changes in 7 of 9 elbows. Mild to moderate HO was seen in 6 of 9 elbows, 2 of which were residual HO in patients undergoing excision of HO. None of the patients required additional operations for instability, range of motion, or arthritis.

Complications One patient developed a delayed-onset ulnar neuropathy necessitating ulnar nerve transposition on postoperative day 12.2,4,5 After transposition, her symptoms resolved during the following 3 to 4 months. This same patient also developed incisional erythema that resolved after administration of oral antibiotics.

653 Table III

Scoring system results

Case

ASES

ASES selfsatisfaction (1-10)

QuickDASH

MEPI

SOD

1 2 3 4 5 6 7 8 9 10 11 12 13 14

100 86 79 87 100 36 85 83 61 99 74 90 89 66

10 10 6 10 10 4 10 10 10 Not answered 8 8 10 0

0 7 23 3 0 64 7 0 43 0 20 2 2 7

100 85 65 85 100 65 95 85 60 100

10 9 5 9 10 3 9 10 6 10 8 7 6 2

MEPI, Mayo Elbow Performance Index; ASES, American Shoulder and Elbow Surgeons elbow assessment; QuickDASH, Quick Disabilities of the Arm, Shoulder, and Hand questionnaire; SOD, Subjective Outcome Determination.

Table IV

Correlation of Outcome Scores

SOD vs. QuickDASH ASES vs. MEPI SOD vs. MEPI MEPI vs. QuickDASH ASES vs. QuickDASH SOD vs. ASES

Spearman rho

P value

0.98 0.87 0.82 0.78 0.75 0.73

Box-loop ligament reconstruction of the elbow for medial and lateral instability.

Elbow instability remains a challenging surgical problem. Most commonly, isolated reconstructions of the medial collateral ligament or lateral collate...
887KB Sizes 0 Downloads 13 Views