Shoulder arthrodesis with plate fixation: Functional outcome analysis

ORIGINAL ARTICLES

Shoulder arthrodesis with plate fixation: Functional outcome analysis Robin R. Richards, MD, FRCS(Cl, Dorcas Beaton, BScOT, and Alan R. Hudson, MB, ChB, FRCS{fd), FRCS(Ct-, T-orento, Ontario, Canada

Fifty-seven patients underwent shoulder arthrodesis with a single plate. The technique utilized included both glenohumeral and acromiohumeral arthrodesis. A 7O-hole plate was used for internal fixation. The position utilized was 3Cf abduction, 3Cf internal rotation, and 3Cf flexion. Forty-six patients underwent the procedure for brachial plexus in;ury, six patients for multidirectional shoulder instability, two patients for osteoarthritis, two patients for failed total shoulder arthroplasties, and one for infection. The patients were independently reviewed by a research clinician and their function assessed according to their ability to perform activities of daily living, sub;ective satisfaction with the procedure, and the degree of pain they felt in their shoulder. Fifty-four shoulders fused within 7Cf of the desired position. Three patients required secondary bone grafting. The complication rate was 74%. Patient satisfaction was highest in those patients undergoing the procedure for brachial plexus in;ury, osteoarthritis, and failed total shoulder arthroplasty (p = 0.0046). Four patients with multidirectional shoulder instability continued to complain of instability in spite of solid arthrodesis (p < 0.07). The diHerence in the cumulative activities of daily living score between the brachial plexus in;ury and poor hand function group and the multidirectional shoulder instability group and patients with other diagnoses was significant (p = 0.0007). Preoperative diagnosis, hand function, and work status were significant determinants of patient satisfaction and the cumulative activities of daily living score (p = 0.0007). (J SHOULDER ELBOW SURG 7993;2:225-39.)

Since the development of shoulder arthroplasty, relatively few indications for glenohumeral arthrodesis remain. 13,26 Previous reports of shoulder arthrodesis primarily concern the relative value of different surgical techniques.* Relatively few previous reports document the incidence of complications and the functional

"References 1,2,4-12,15-24,26,27,31,32,34,38-40. From the Upper Extremity Reconstructive Service, St. Michael's Hospital; and the University of Taranto, Toronto. Reprint requests. Robin R. Richards, MD, 55 Queen St. E., Ste. 800, Toronto, Ontario, Canada, M5C 1R6. Copyright © 1993 by Journal of Shoulder & Elbow Surgery Board of Trustees. 1058-2746/93/$1.00 + .10 32/1149932

outcome that can be expected after the procedure. We have performed shoulder arthrodesis in 57 patients with a standardized surgical technique. We performed a retrospective review to determine the fusion rate after the procedure, the incidence of complications, the functional outcome that can be expected after the operation, and the primary determinants of an individual patient's functional outcome.

MATERIAL The case records and radiographs of 57 consecutive patients who underwent shoulder arthrodesis between June 1980 and June 1991 at our institution were reviewed. All patients were operated on by the same surgeon (RRR) using

225

226

Richards, Beaton, and Hudson

J. Shoulder Elbow Surg. September/October 1993

Figure 1 Radiograph of 28-year-old women with multidirectional Instability This patient had undergone four prevIous procedures In attempt to stabilize her shoulder She continued to have severe pain and shoulder dysfunction This patient was unable to work and required narcotic analgesIc medication There IS severe inferior subluxation of shoulder at rest without traction applied to arm There was no deltoid paralysIs

the same operative technique (see following). There were 51 men and six women. Arthrodesis was performed on the right shoulder in 37 patients and the left shoulder in 20. Mean patient age at the time of surgery was 31 years (range 14 to 60 years). Mean follow-up was 43 months (range 13 to 132 months). The preoperative diagnosis was brachial plexus injury (BPI) in 46, failed total shoulder arthroplasty in two (loosening one, loosening and infection one), osteoarthritis in two, and sepsis after open treatment of a proximal humeral fracture in one patient. Six patients had multidirectional instability refractory to treatment by multiple operative procedures by different surgeons. These patients were specifically referred to undergo arthrodesis because of the severity of their shoulder dysfunction (Figure 1). Forty-two patients with BPI had undergone brachial plexus exploration before undergoing shoulder arthrodesis. Twenty-seven of the patients with BPI underwent 29 elbow f1exorplasties in addition to shoulder arthrodesis. Sixteen patients had Steindler procedures, 12 patients had pectoralis major ten-

don transfers, and one patient had a triceps flexorplasty.28 Two patients had both Steindler and pectoralis major f1exorplasties. All of the patients with BPI had flail shoulders «MRC Grade 2 deltoid power) before the operation. The indication for surgery in the patients with BPI was inability to position the extremity and not pain or instability. Shoulder arthrodesis is not indicated for the patient with BPI if the patient has enough active motion that the hand can be effectively positioned in space. The patients undergoing shoulder arthrodesis who did not have flail shoulders had severe, unremitting pain that interfered with their activities of daily living and negatively affected their quality of life. Eleven patients were receiving worker's compensation board (WCB) benefits. Operative Technique. 30 The patients were given a general anesthetic and placed in the semisitting position. 3 The arm was draped free. The incision extended from the spine of the scapula to the anterior acromion and down the anterior aspect of the shaft of the humerus. The deltoid muscle was detached from the anterior

J. Shoulder Elbow Surg. Volume 2, Number 5

acromion, and its fibers were split distally. The rotator cuff was resected. The glenoid fossa, the undersurface of the acromion, and the humeral head were decorticated. An attempt was made to obtain arthrodesis of both the glenohumeral and acromiohumeral articulations, because the glenoid fossa offers such a small area for fusion with the humeral head. Decortication of the undersurface of the acromion increased the potential fusion area. A 1O-hole 4.5 mm pelvic reconstruction plate (Synthes, Mississauga, Ontario, Canada, and Wayne, Pa.) was used for internal fixation. Before 1985, a 1O-hole 4.5 mm dynamic compression plate was used. The pelvic reconstruction plate is currently used because it has adequate strength and is more easily contoured in the operating room. After resection of the rotator cuff and decortication of the joint surfaces the shoulder was supported in 30° flexion, 30° abduction, and 30° internal rotation. Abduction was measured from the side of the body. Although this method of measurement does not accommodate for individual variations in muscle mass or body fat, clinical experience has shown it to be accurate within 10° in any plane. The humeral head was brought proximally to appose the decorticated undersurface of the acromion. When the humerus is obducted and flexed 30° the humeral head apposes both the undersurface of the acromion and the glenoid fossa. This position was maintained by supporting the arm with sterile folded sheets. Bending irons were used to contour the plate along the spine of the scapula, over the acromion, and down onto the shaft of the humerus. The malleable nature of the plate allowed precise intraoperative contouring of the implant to the specific local anatomy in any given patient. The plate was bent over the acromion and twisted just distal to the bend to appose the shaft of the humerus. The reconstruction plate has holes similar to the dynamic compression plate that allow angulation of the screws as they are passed through the plate. Two or three screws passing through the plate and the humeral head into the glenoid fossa were inserted first. These screws compress the arthrodesis site. Another cancellous screw was placed across the acromiohumeral fusion site, and the remaining holes of the plate were secured with cortical screws (Figure 2). The ac-

Richards, Beaton, and Hudson

227

Figure 2 Postoperative radiograph after shoulder arthrodesis. Ten-hole 4.5 mm pelvic reconstruction plate has been contoured to shoulder. Screws coursing from plate, through humeral head into glenoid are inserted first. These screws compress arthrodesis site.

romion was not osteotomized because it was used to augment fixation of the scapula to the humerus. Autogenous bone graft was not used. The arm of the patient was supported after the operation with a pillow and swathe. A thermoplastic thoracobrachial orthosis was applied on the first postoperative day. Patients were instructed to wear the orthosis at all times except to bathe for the first 6 weeks after the operation. If there was no radiographic sign of loosening of the internal fixation 6 weeks after the operation, the patient's arm was placed in a sling. Gentle range-of-motion exercises were allowed until radiographic union was achieved. It was difficult to be certain when fusion occurred radiographically, because the fixation wos sufficiently rigid that very little callus formed around the arthrodesis site. Return to strenuous activity was delayed for at least 16 weeks after the operation.

228

1. Shoulder Elbow Surg. September/October 1993

Richards, Beaton, and Hudson

METHOD All patients were followed after the operation until their arthrodesis had united. Radiographs were taken at 6 weeks and at 3, 6, 9, and 12 months until the arthrodesis was solid both clinically and radiographically. If union did not progress and/ or there was loosening of the internal fixation, bone grafting was performed. Position of the arthrodesis was assessed at follow-up by placing the scapula in the anatomic position and measuring the position of the arm goniometrically. Before analysis of the data, patients were subclassified according to cause of the problem into one of five groups. BPI, good hand fundion (n = 21). These patients had sustained brachial plexus injuries that did not involve the hand. BPI, abnormal hand fundion (n = 12). These patients had sustained brachial plexus injuries that had involved the hand to the extent that tendon transfer surgery was required to restore tendon balance in the hand and wrist.

BPI, poor hand fundion (n

= 13).

These patients had sustained brachial plexus injuries with hand involvement, and the neurologic loss in the hand and forearm was sufficiently severe that it was not amenable to treatment by tendon transfer. Instability (n = 6). These patients were referred with a diagnosis of multidirectional shoulder instability. They had undergone multiple operative procedures by multiple surgeons and their instability persisted to the extent that they could not work, had constant pain, and were willing to undergo arthrodesis in an attempt to rehabilitate their severely dysfunctional shoulders. Other diagnoses (n = 5). The remaining patients were placed in this group (failed total shoulder arthroplasty, two, osteoarthritis, two, infection, one). It was possible to contact 33 patients for independent review by a research clinician (DEB). Two patients had died, two patients were incarcerated, and the remaining patients could not be contacted because of a variety of reasons, the most common being that they had moved and could not be located or did not respond to requests to return for follow-up. The patients in the final follow-up group were distributed in the following manner: BPI, good hand function, 10; BPI, abnormal hand function, seven; BPI, poor hand function, seven; multi-

directional instability, six; and other diagnoses, three. At review patients responded to a questionnaire. The questionnaire assessed general confidence in the shoulder, whether or not the surgery was definitely beneficial, somewhat beneficial, not beneficial, or had made them worse, the subjective percentage improvement in shoulder function related to the surgery, whether or not the shoulder felt unstable, the ability to perform 16 activities of daily living (ability to perform activities of daily living was assessed on a four-point ordinal scale as follows: 0: unable to perform, 1: able to perform with much difficulty, 2: able to perform with some difficulty, 3: able to perform with no difficulty), and visual analogue assessment of pain. Statistical analyses. Data are reported as mean values ± the standard error of the mean. All analysis was performed with Statistum Application Software (SAS Institute Inc., Cary, N.C) for personal computers. A one-way analysis of variance was used to compare findings among the five diagnostic subgroups. Duncan's post hoc comparison for multiple measures was performed to identify groupings if the Ftest was significant. A comparison of the proportion of patients reporting stability and benefitfrom the surgery between the multidirectional instability group and all other patients was made. This analysis was carried out with a chi-squared analysis with one degree of freedom. Regression analysis was performed to determine the best predictors of outcome. The two outcome measures studied were subjective percentage improvement in shoulder function as a result of surgery and the cumulative activity of daily living score. The variables assessed in the regression analysis included length of followup, preoperative diagnosis, hand function, dominance, WCB claim, mood, pain, age, and sex. The level of statistical significance chosen was p < 0.05.

RESULTS (Table) A comparison of the baseline information demonstrated no significant differences between age and follow-up time among the diagnostic groups. All patients eventually achieved solid arthrodesis. No malunions occurred (all fusions occurred within 10° of the desired positions in each plane). No intraoperative complications occurred.

J. Shoulder Elbow Surg. Volume 2, Number 5

Three patients had glenohumeral nonunion. The first had undergone arthrodesis after an infected total shoulder arthroplasty. His fixation loosened and required revision at 6 months. At that time bone grafting was performed. A second patient with sepsis after treatment of a proximal humeral fracture had loosening of some of the humeral screws. This patient underwent iliac crest bone grafting 6 months after the operation and went on to solid union. A third patient experienced breakage of one of the compression screws inserted into the glenoid. Screw breakage occurred approximately 30 months after surgery. At the time the patient had been working full-time as a welder for 18 months. Tomograms demonstrated solid acromiohumeral arthrodesis with incomplete glenohumeral arthrodesis. This patient's internal fixation was removed, and the glenohumeral arthrodesis site was bone grafted with iliac crest bone. Union occurred uneventfully. One patient experienced a wound infection that required drainage and eventual plate removal. Seven (12%) other patients required plate removal because of the prominence of the internal fixation irritating the overlying soft tissues. Two patients, both with multidirectional instability, had continued pain in spite of solid arthrodesis. Both of these patients sought referral to pain clinics where thoracic outlet syndromes were diagnosed. These diagnoses were made primarily on the basis of symptoms without objective evidence of neurologic deficit and / or vascular obstruction. These patients were treated surgically by first rib resection. Both patients were diagnosed as having reflex sympathetic dystrophy in addition to thoracic outlet syndrome, although this diagnosis was not confirmed in either patient. Their symptoms were ameliorated to some extent, although they both remain symptomatic. Two patients experienced humeral fractures at the distal end of the plate. Both patients sustained their fractures in falls while in an inebriated state. One patient sustained a second humeral fracture after healing of his first humeral fracture when he was arrested. The first and second fracture in this patient were treated by plate fixation. The fracture in the second patient was treated by application of a spica cast for 6 weeks. The total incidence of complications was 8 (14%) of 57.

Richards, Beaton, and Hudson

229

Subiective improvement. Twenty-seven of 33 patients felt generally confident in the use of their upper extremity. Four of the six patients with multidirectional instability did not feel confident. Ten of the 11 patients in the BPI - good hand function group felt that the surgery had definitely improved their upper extremity. One patient in this group felt that function had improved somewhat. All patients in the BPI - abnormal hand function group felt that the surgery had improved their function. All patients except one in the BPI - poor hand function group believed that their function had improved, and all three patients with other diagnoses believed function had been definitely improved. Only one patient with multidirectional instability felt that shoulder arthrodesis had improved their function. One patient in this group was unsure, two believed their function had improved somewhat, and two believed their function had been worsened by the procedure (Figure 3). The difference between the multidirectional instability group and all of the other diagnostic groups was subjected to chi-square testing. The difference was highly significant (p < 0.00003). The percentage subjective improvement in the BPI - good hand function group was 85 ± 5.1. In the BPI - fair hand function group, subjective improvement was 76 ± 4.6. In the BPI - poor hand function group, the improvement in function was 71 ± 10.8. On Iy four of the six patients with multidirectional instability would estimate thei r percent improvement. Three of the patients stated they had not improved at all, and one patient stated they had improved 75% (mean improvement in this group 19%). Mean improvement of the three patients in the other diagnostic group was 90% ± 10.1 % (Figure 4). The differences in the quoted subjective improvement between the diagnostic groups were significant with the multidirectional instability group having a much lower value than the four other groups (ANOVA, Duncan's p = 0.0046). Instability. No patient apart from those in the multidirectional instability group reported subjective shoulder instability. Four of the six patients in the multidirectional instability group reported a continued sensation of shoulder instability in spite of solid arthrodesis of their shoulders (Figure 5). These patients reported that their shoulder felt "as if it was going to dislocate again" and that "it did not feel any stronger or more stable than it did preopera-

230

Richards, Beaton, and Hudson

J. Shoulder Elbow Surg. September/ October 1993

Table

I

Age at Date of Previous i BP :1 Hand tiAHededll Diagnosis name operation Sex dominant side WeB operation Follow-up operations operation

BPI-Good hand function 28 CS 37 GO 19 JR 21 JH 56 JB 20 GC KY 33 MP 30 23 MO 45 FP 19 EG 14 JL 44 GA 20 OM 22 GW RF 33 52 AS 53 VC 20 MG NS 36 KO 24 BPI-Abnormal hand 42 NN 30 OS JR 22 34 KV 25 SO 28 CR 53 PS PH 33 20 CB 27 MF 22 TL 40 MP BPI-Poor hand fundion 19 JG 39 MM 24 RO 23 RM 23 KM 25 CB 26 GR 28 TR FQ

34 41 30 23 27

PA RH AH JE Instability 26 1M 28 JS 27 KP 28 JH 34 GO 25 RA Other-(hand normal) 53 SP (TSA) OA (TSA, INF) 36 32 CL (OA) 43 CC (OA) OM (INF) 60

09/80 04/81 09/82 09/82 10/82 07/84 07/84 10/85 12/85 05/86 08/86 03/87 06/87 08/87 01/88 03/89 09/89 09/89 09/89 04/90 05/90

132 19 120 112 26 18 69 72 43 48 30 19 58 34 52 18 33 66 36 24 23

N N N N N

06/80 08/82 09/84 02/85 03/85 09/85 05/86 10/86 05/90 03/90 06/90 12/90

18 96 23 73 39 80 71 62 27 24 24 18

06/80 06/81 01/82 06/82 01/86 04/87 04/87 05/88 04/90 11190 05/91 07/91 09/91

46 20 18 24 58 61 60 47 23

R L R

N N N N N N N N N N Y N N

R R R

R l l

N Y N

R

L

Y

M M

R

R R

Y N

04/86 01/89 06/90 07/90 08/90 03/91

72 40 26 21 24 19

:1

M

R

;:

L L

R R R

N N N N

06/85 01/88 03/89 06/89 07/89

84 50

3

M M M M M M M M M M M M M M M M M M M M M M M M M M F

M M M M M M

M M M M M M M F

M M M M M F

F F M

M

M

M

R R R l

R R R

R R R

L

R R L

R R R L L L R R L R R R

N N N N y

N N N N y

N N N N N N N Y N

R R R R R

R l

l.

N

R L R R R

L R

y

R R L R R R R R R R R R R R R R R

L

R L

R R

R L R R R R R L R L R l

R R R R R R

l

R

--------

y

Y N N N N y

N

0

1

1

1 0 1

Y

Y

y y y

0

N Y

1

Y

Y

y Y y y

Y y y y y

y

Y

]

0

N

Y

y

y Y

Y

y y y

15 16 13

31

y

Y y

N N N N N N

7

4 ')

3 ~

4

n

" --

------"--------------------~----------

BPI, Brachial plexus injury, WeB, worker's compensation benefits. ADL, activities of dad y liVing

y Y N Y Y Y y Y Y y y

22

24 27

N Y Y

N N N N N ._----

J. Shoulder Elbow Surg. Volume 2, Number 5

231

Richards, Beaton, and Hudson

Flexoplasty (type)

Fusion solid

Complications

Bone graft

Plate oR Noormo

General confident

5 5 5 5 N N 5 P,5 N N P,5 N N N N P P N P 5 N

Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y

N N N N N N N N N # # N N N N N N N N N N

N N N N N N N N N N N N N N N N N N N N N

Y N N N N N N N N Y Y N N N N N N N N N N

Y Y Y Y

N 5 P P N 5 5 N P 5 5

Y Y Y Y Y Y Y Y Y Y Y Y

N Partial N N N N N N N N N N

N Y-36 N N N N N N N N N N

Y N N N N N N N N N N N

N N N 5 T P P P 5 N P N N

Y Y Y Y Y Y Y Y Y Y Y Y Y

N N Infection N N N N N N N N N N

N N N N N N N N N N N N N

N N Y N N N N N N N N N N

N N N N N N

Y Y Y Y Y Y

T05 T05 N N N N

N N N N N N

Y N Y Y N N

Y N Y N N N

N N N N N

Y Y Y Y Y

N Nonunion N N Nonunion

N Y6M N N Y6M

N N N N N

Y Y Y

S

N

Y

Y Y

Y Y Y Y

Y N Y Y Y Y

Y Y Y Y Y Y Y

Cont'd

J. Shoulder Elbow Surg. September/ October 1993

232 Richards, Beaton, and Hudson

Table cont'd Surgery beneficial

I Percent better

Unstable

85

I H~lr I F~ce

Axilla

Wash back

Bathe

Toilet

I pocket Front

Back pocket

o

3

2

o

2

2

o

o

3 3

2 2

o o

3 2

2 2

~j

o

N

o

3

2

o

o

2

2

o

90

N

0

3

2

0

3

3

2

0

Definitely

100

N

3

3

3

3

3

3

3

3

Definitely

70

N

0

3

3

0

3

2

3

0

Some Definitely Definitely

50 100 85

N N N

0 3 0

1

0 1

0 2

0

3

0

3

3

3

2

0 3 2

'1

3

2

0 3 1

Definitely

75

N

0

3

3

3

2

3

0

Definitely Definitely

100 75

N N

0 0

2 2 3

3 3

2

0

Definitely Definitely Definitely Definitely

60 80 70 75

N N N N

3 3

3 3

3

0 0

3 3

Definitely Some Definitely Definitely Definitely Definitely Definitely

45 70 100 100 35 50 100

N

a

N N N N N

Unsure Worse Definitely Some Worse Some

0 0

Definitely Definitely Definitely

Definitely Definitely Definitely Definitely

100 70

N N N N

Definitely

80

Definitely

1

a

2

"-

0 0

1 3

2

3 3

3

0 0

3

3 3 3

1

0

a

I

3 0

3

3 3

0

0

3

a

2 0 0 0 1 0 0

0 1 0 0 0 0

0 0 0 0 0 0 0

1 0 2 0 0 0 1

0 0 0 0 0 0 0

0

0 0 0 0 0 0 0

0 0 0 0 0 0

1

0 1 1 0 3

0 0 0 0 0 0

0 0 0 0 0 0

2

0 75

y N y y y N

2 0 0 2

3 0 3 3 0 0

0 0 0 0 0 3

0 0 0 0 0 0

100 100 70

N N N

0 2

3

3 3 3

1 3 3

0 2 2

3 3 3

0 3 3

3 3 3

0 3 2

5

25

20

6

7

13

23

No. able to perform

N

0 0 0 0 0

1

I 0

3

0 1

0

11

----_.

Richards, Beaton, and Hudson

J. Shoulder Elbow Surg. Volume 2, Number 5

Dressing

.

Button

Work shoulder

o 2

Lift

Work waist

Use knife

bag

3

3

2

.

Eat

.

233

Sleep

Cumulative ADL

3

25

2 3

31 28

3

3

26

5

VA pain

2

o

3 3

3

2

2

2

2

2

2

3

2

2

3

3

3

3

3

3

2

34

5

3

3

2

3

3

3

3

3

47

0

3

2

2

3

3

3

3

3

36

2

2 3 3

2 3 2

2 3 2

3 3 3

2 3 2

3 3 2

0 3 1

3 3 0

20 45 28

4 0 2

3

3

2

3

3

3

3

3

38

2 3

2 2

2 2

3 3

3 3

3 2

3 0

3

30 34

6 7

3 3 3

2 3 3 0

2 2 1

3 3 3 3

2 3 3 2

3 3 3 2

3 3 3 0

39 38 40

3 0 2 0

0 0 0 0 1 3 1

0 0 0 0 0 0

1 0 0 2 1 0

3 0 3 3 0 0

0 0 0 0 0 0

3 0 2 0 0 0 2

0 0 0 0 0 0

16 1 13 8 7 6 9

4 0 3 3 10

0

0

3 0 2 1 2 1

3 1 3

0

0 3

0

5 9 6 7 8 10 0

2

2

0

2

2

0

3

0

2 0 3

0 0

0

1 1 1 1 0 3

2 3 3

3 3 3

1 2 3

3 3 3

3 3 3

29

21

21

29

25

3 1 3 1 1 2

0

3

0

0

2

3

2

0

2 1 0

2

0

2

1 2 3 3 2

2

0 0 1

11

0

1 0

16 4 19 12 4 20

3 3 3

3 3 3

3 3 3

31 45 46

27

17

27

0

0 0 0

3

2

2 2

0

0

234

Richards, Beaton, and Hudson

J. Shoulder Elbow Surg. September/October 7993

BETTER THAN PRE-OP? Percentage reporting: Yes 100

I

[I~---'-------'---i

_

80

ALL OTHERS

_

MOl

p < 0.00003 (Chi square)

60

2/6

2/6 40 20

o

DEFINITELY

SOMEWHAT

SAME

WORSE

Figure 3 Postoperative improvement

PERCENT ---~-~~----------~--

100

-------

------

ANOVA, DUNCAN'S - p 76

80

--

-----

--,

= 0.0046

,i

71

60

*

40

19

20 0 OTHER

BPI-GOOD

BPI-FAIR

BPI-POOR

MOl

Figure 4 Percentage sublective improvement * Indicates statistically significant difference from other diagnostic groups.

tively." The difference between the response of the patients with multidirectional instability and those with other diagnoses was significant (Fisher's exact test, p < 0.001). Activities of daily living. The patients' ability to perform specific activities of daily living were rank-ordered as follows, beginning with the activity that patients (n = 33) were least able to perform after shoulder arthrodesis: doing hair, five; washing back, six; bathing,

seven; reaching back pocket, 11; managing toiletting, 13; eating, 17; reaching opposite axilla, 20; doing up buttons, 21; working at shoulder level, 21; reaching front pocket, 23; washing face, 25; using knife, 25; lifting bag, 25; sleeping, 27; getting dressed, 29; and working at waist level, 29. Ability to perform individual activities of daily living varied depending on the patient's diagnostic group and the specific activity being as-

235

Richards, Beaton, and Hudson

J. Shoulder Elbow Surg. Volume 2, Number 5

% REPORTING: YES 100

4/6

p < 0.001

80

(FISHER'S EXACT)

60 40

0/27

20

o

MOl

OTHER DIAGNOSES

Figure 5 Subjective shoulder instability.

MAXIMUM SCORE POSSIBLE - 3 3.5

3 2.5 2 1.5

0.5

o

OTHER

A

BPI-GOOD

BPI-FAIR

BPI-POOR

MOl

TOTAL SCORE

MAXIMUM SCORE POSSIBLE - 3 50

3.5

ANOVA, DUNCAN'S - P • 0.001

45

3

40

2.5

35 30

2

*

25

1.5

20 15 10

0.5

o

B

5 OTHER

BPI-GOOD

BPI-FAIR

BPI-POOR

MOl

o

C

OTHER

BPI-GOOD

BPI-FAIR

BPI-POOR

MOl

Figure 6 A, Mean activities of daily living (ADL) score: doing up buttons. B, Mean ADL score: sleeping on shoulder. C, Cumulative ADL scores. * indicates statistically significant difference from three other diagnostic groups.

sessed. Patients with poor hand function were not able to perform activities requiring dexterity such as doing up buttons (Figure 6, A). Patients with multidirectional instability had an intermediate score for this activity even though they

had no demonstrable limitation of hand function. Poor hand function was not a variable when an activity such as sleeping on the shoulder was assessed (Figure 6, B). Patients with multidirectional instability had a lower score for

236

J. Shoulder Elbow Surg. September/October 1993

Richards, Beaton, and Hudson

0- NO PAIN

10 - PAIN AS BAD AS IT CAN BE

10 8

7.5

ANOVA, DUNCAWS - P • 0.0008

6 4

2

OTHER

BPI-GOOD

BPI-FAIR

BPI-POOR

MOl

Figure 7 Pain: visual analogue scale

this activity than patients with other diagnoses. The cumulative activity of daily living score was 32.0 ± 2.6 in the BPI - good hand function group and 32.6 ± 3.9 in the BPI - abnormal hand function group. Patients with other diagnoses had the highest cumulative ADL score, 40.6 ± 4.9. Patients with BPI - poor hand function had the lowest score 8.6 ± 1.9. Patients with multidirectional instability had an intermediate score of 12.5 ± 2.9 (Figure 6, q. The difference in the cumulative ADL score between the BPI - poor hand function group and the group of patients with multidirectional instability, when compared with the other diagnostic groups, was significant (ANOVA, Duncan's p = 0.001). Pain. The patients with multidirectional instability complained of severe shoulder pain in spite of solid arthrodesis. Their mean pain score (out of a possible 10 for " pa in as bad as it can be") was 7.5 ± 0.8. In contrast, the patients with other diagnoses did not complain of pain (Figure 7). The patients with BPI complained of an indeterminate amount of pain (good hand function 2.4 ± 0.6, fair hand function 2.7 ± 1.2, poor hand function 3.4 ± 1.2). These differences in pain scores were significant (ANOVA, Duncan's p = 0.0008). Regression analysis. Regression analysis revealed etiologic group to be the single best predictor of the ability of patients to perform their activities of daily living. Whether or not the person had an active WCB claim was of bor-

derline significance in the regression analysis, and it is recommended that it be included in the regression equation. Accordingly, the best model for the patient's perceived benefit from the procedure was the combination of etiologic grouping and WCB status. The r 2 for this with perceived benefit as the outcome measure was 0.5135, with the F test for the model having a p value of 0.0046. The significant regressors were the same with cumulative ADL score as the outcome measure (p = 0.0001). In this regression the etiologic group was the strongest predictor accounting for 51 % of the variance. The lack of significance of the other regressors could be that they truly did not contribute to predicting the defined outcomes, but it may also be due to a lack of power in the small sample, or possibly collinearity between the regressors.

DISCUSSION This series represents, to the best of the authors' knowledge, the largest series of patients having undergone shoulder arthrodesis with a single operative technique. Previous reports of shoulder arthrodesis have included patients whose shoulders were fused by various surgical techniques in a variety of positions. One of the strengths of the current study is that the procedures were done by one surgeon utilizing a single operative technique. Therefore surgical technique and the position of the arthrodesis were less important in determining outcome than they would have been had patients been

J. Shoulder Elbow Surg. Volume 2, Number 5

operated on by a number of surgeons utilizing different surgical techniques.* The operative technique utilized in the study led to a surprisingly high rate of fusion. Primary bone grafting was not performed. Secondary bone grafting was required in only three patients. The factors leading to the observed high fusion rate may include the mobility of the scapula, leading to reduced stress on the internal fixation and careful placement of the first two or three screws into the glenoid to obtain compression at the arthrodesis site. The excellent blood supply of the periarticular soft tissue and decortication of the undersurface of the acromion may also increase the fusion rate. In addition, a thermoplastic orthosis was utilized for the first 6 weeks after the operation. In the present series the complication rate was 14%. We believe this rate is acceptable considering the magnitude of the surgery, the severe osteoporosis present in many of the patients, and the relatively unreliable nature of many of the patients with brachial plexus injuries. The incidence of complications in the large series of Cofield and BriggsS was 18%. The operative technique has proved to be reliable in obtaining arthrodesis and in maintaining position of the arm while fusion occurs. Bone grafting on a routine basis does not seem necessary or justified because only three patients ultimately required it. Routine bone grafting should be considered when performing arthrodesis in shoulders after failed total shoulder arthroplasty, because one such patient experienced a nonunion, and severe bone loss is usually present in these patients. The method of review in this study was independent of the operating surgeon. Patients were given questionnaires that were designed in such a way that the patients could answer them with minimal instruction and / or assistance. The pain assessment was performed with a visual analogue scale. I ". 35. 37 This method of pain assessment avoids the creation of artificial categories of pain and allows the patient freedom in quantitating his or her pain perception. The scores obtained in this fashion can then be analyzed statistically to determine whether or not true differences existed between the groups of patients. 36 'References 1, 2, 4- 12, 15-24, 26, 27, 31, 32, 34, 38-40

Richards, Beaton, and Hudson

237

The method of functional assessment utilized in this study relies heavily on a patient self-assessment questionnaire. This method of functional assessment ignores the measurement of range of motion and strength assessment performed by physicians. Instead, the method concentrates on the patients' function and their subjectively perceived benefit. Although we administered the follow-up questionnaire in person, it can be answered by mail or over the telephone. Because the patients are the consumers of the operation and its result, it is the authors' contention that this method of follow-up is appropriate. In retrospect, if we had asked each patient to complete a questionnaire at their last regular follow-up visit, recall of the patients would not have been necessary. Although the mean follow-up time would have been decreased, we would have had final follow-up data on a higher proportion of patients. An even more reliable method of functional assessment would have been to observe the patients performing specific tasks. Independent observation of the patient would increase the objectivity of the functional outcome assessment and remove patient satisfaction as a factor that might modify the ADL score reported by the patient. The question of pain after shoulder arthrodesis is a troubling one. This phenomenon has been previously observed and reported by our group and other authors. s. 29 Shoulder arthrodesis places significant stress on the periscapular musculature. The loss of shoulder rotation significantly limits shoulder motion and prevents most patients from doing their hair easily, washing their back, bathing, and reaching their back pocket. In the authors' experience it takes most patients at least 6 to 12 months to recover completely from the operation and to maximize their shoulder function. The presence of pain in the patients with brachial plexus injuries is understandable, because patients with peripheral nerve injury frequently do experience pain on an ongoing basis. The severe pain reported by the patients with multidirectional shoulder instability is more difficult to explain. Three of these patients were receiving worker's compensation benefits. All of these patients had undergone multiple operative procedures. The authors can offer no explanation for the patients continued discomfort in spite of solid arthrodesis. We speculate that these patients had

238

.J. Shoulder Elbow Surg. September/October 7993

Richards, Beaton, and Hudson

complex neuromuscular dysfunction involving not only the structures stabilizing the glenohumeral joint but also the periscapular musculature. Absolute stabilization of the glenohumeral joint, by itself, appears to have been insufficient to improve their shoulder function. Although all of the patients in this diagnostic group appeared to be depressed at various points in their clinical course, none of them were diagnosed as having psychiatric disorders, apart from chronic pain syndromes. 33

12. Gill AB. A new operotlon for arthrodeSIS of the shoulder J Bone Joint Surg 1931; 13·287

CONCLUSIONS

13 Hawkins RJ, Neer CS. A functional analySIS at shoulde· fUSions C1in Orthop 1987,223:65-76

The use of single plate fixation, the 30 - 30 30 position, and a 6 week period of immobilization was reliable in terms of maintaining position and encouraging fusion with an acceptable complication rate. Routine use of a bone graft is unwarranted in most cases. Preoperative diagnosis and weB status are the major determinants of perceived benefit and the cumulative activity of daily living score after shoulder arthrodesis. Hand function is a major outcome determinant in patients with BPI who undergo shoulder arthrodesis. Shoulder arthrodesis is a poor therapeutic alternative for patients with multidirectional instability who have had multiple procedures attempting to stabilize their shoulders. Shoulder pain can persist in some patients despite solid arthrodesis of the glenohumeral joint. Shoulder arthrodesis is a good reconstructive procedure for certain specific diagnoses and deserves a continued role in the armamentarium of the reconstructive surgeon. Patient selection is critical to outcome after the procedure.

14. HuskissonEC Meosurementofpom LancerlY74;li21. 31

REFERENCES 1. Barr JS, Freiberg JA, Colonna PC, Pemberton PA A survey of end results on stabilization of the paralysed shoulder. Report ofthe Research Committee of the AmerIcan Orthopaedic Association J Bone Jomt Surg 1942;24.699-707 2. Borton NJ. Arthrodesis of the shoulder for degenerative conditions. J Bone Joint Surg lAm] 1972;54A,8: 1759-64 3. Bateman JE. The shoulder and neck. Ed 2. Philadelphia WD Saunders Co, 1978'273-4 4. Beltran JE, Trilla JC, Barjan R A Simplified compression arthrodesis of the shoulder. J Bone Joint Surg [Am] 1975;57A.4:538-41 5 Brittain HA. Architectural prinCiples In arthrodesIs Ed:3 Edinburgh· E and L Livingstone, 1952 6. Carroll RE Wire loop In arthrodesIs of the shoulder. Clin Orthop 1957;9:185-9.

7 Charnley J, Houston JK Compression urthrodesls ot the shoulder J Bone Joint Surg [Br] 1964,46-B 4,614-20 8 Cofield RH, Briggs BT Glenohumeral arthrodesIs J BO["1(' Jomt Surg 1979;61 A 668-77 9. DavIs JB, Cottrell GW A technique tor shoulder arthro desls J Bone JOint Surg [Am] 1962;44A'657-61 lODe Velasco PG, Cardoso MA. ArthrodesIs of the shaUl' der. C1in Orthop 1973;90'178-82 11

Freidman RJ, Ewald FC Arthroplasty of the Ipsilateroi shoulder and elbow m patients who hove rheumatOid orthrltis J Bone JOint Surg [Am] 1987;69A661-6

15 Johnson CA, Healy WL, Brooker AF Jr, I\rockow K!\ Externol fixation shoulder arthrodeSIS On Orthop 1986,211219-23 16 KOCialkowskl A, Wallace WA Shoulder arthrodeSIS uSing an external fixator J Bone JOint Surg [Br] 1991 ;73B 180-1 17. Kostulk JP, Schotzker J Shoulder arthrodeSIS - AO technique In' Bateman JE, Welsh RP, eds Surgery of the shoulder Philadelphia CV Mosby, 1984207-10 18 Leffert RD BrachlOl plexus InlUries New York Churchl;. LIVingstone, 1985: 193-21 0 19 Moh JY, Hall JE ArthrodeSIS of the shoulder m chridrer' J Bone Jomt Surg [Am] 1990,72A 582-6 20 Makin M. Early arthrodeSIS for 0 flail shoulder In young children J Bone JOint Surg [Am] 1977,59A 317-21 21

May VR Shoulder fUSion a review of fourteen cases J Bone JOint Surg [Am] 1962,44A 1 65-76

22 Mosely HF ArthrodeSIS of the shoulder Orthop 1961,20 156-62

In

the adult Clin

23. Muller ME, Allgower AM, Willenegger H Manual at lr, ternal flxotion Ed 2 Berlin Springer-Verlag, 1979 24 Nagano A, Oklnaga S, OchiOi N, Kurokawa T Shoulder arthrodeSIS by externol fixotion Clin Orthop 1989. 24797-100 25 Neer CS II, Watson KC, Stanton FJ Recent experience m total shoulder replacement J Bone Jomt Surg [Am! 1982,64A 319-37 26 PUttl V. ArthrodeSIS for tuberculOSIS of the knee ana shoulder Chlr Organl Mov 1933;18217 27 Raunlo P ArthrodesIs of the shoulder lomt In rheumatolci arthritis Reconstr Surg Traumato11981;18 48-54 28 Richards RR Operative treatment for Irreparable leSions of the brachial plexus In. Gelberman RH Operative nerve repair and reconstruction Philadelphia JB Llppln· cott Co, 1991.1303-28 29 Richards RR, Waddell JP, Hudson AR Shoulder arthro desis for the treatment of brachial plexus palsy C1ln Orthop 1985,198.250-8 30 Richards RR, Sherman RMP, Hudson AR, Waddell Jr' Shoulder arthrodeSIS usmg a modified pelVIC reconstruc tlon plate. a review of eleven cases J Bone JOint Surg [AmJ1988;70A416-21 31

Riggins RS Shoulder tuslon Without external flxatJon J Bone Jomt Surg [Am] 1976,58A:7-1007-8

1. Shoulder Elbow Surg. Volume 2, Number 5

Richards, Beaton, and Hudson

32 Rowe CR. Re-evaluatlon of the position of the arm In arthrodesIs of the shoulder in the adult. J Bone Joint Surg [Am] 1974;56A.5:913-22. 33. Rowe CR, Pierce DS, Clark JG. Voluntary dislocation of the shoulder. a preliminary report on a clinical, electromyographic and psychiatric study of twenty-six patients. J Bone Joint Surg [Am] 1973;55A 445-60 34. Schrader HA, Frandsen PA. External compression arthrodesis of the shoulder loint. Acta Orthop Scand 1983;54592-5. 35 Scott J, Husklsson EC Graphic representation of pain Pain 1976;2'175-84. 36. Scott J, Huskisson EC. Measurement of functional ca-

paCity with visual analogue scales 1977,16.257-9.

239

Rheum Rehabd

37. Sriwatanakui K, Kelvle W, Lasagna L, Cahmihm JF, Weis OF, Mehta G. Studies with different types of visual analog scales for measurement of pain. C1in Pharmacol Ther 1983;34.2:234-9. 38. Uematsu A. ArthrodeSIS of the shoulder: posterior approach Clin Orthop 1979; 139.169-73 39. Watson-Jones R. Extra-articular arthrodeSIS of the shoulder. J Bone JOint Surg 1933,15:862-71 40. Wilde AH, Brems JJ, Boumphrey FR. ArthrodeSIS of the shoulder: current indications and operative technique Orthop C1in North Am 1987,18'463-72

Bound Volumes Available to Subscribers Bound volumes of the 1993 issues of JOURNAL OF SHOULDER AND ELBOW SURGERY are available to subscribers (only) from the publisher, at a cost of $46.50 for domestic, $55.76 for Canadian, and $52.50 for international subscribers, for Vol. 1 (January-December); shipping charges are included. Each bound volume contains subject and author indexes, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram, with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby, Subscription Services, 11830 Westline Industrial Dr., St. Louis, MO 63146-3318, USA; phone (314)453-4351 or (800)453-4124.

Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular subscription.