Sunteți pe pagina 1din 14

Although the indications for biceps tenodesis were multiple, 1 of these was a SLAP tear, specifically in

older patients. Although patient outcomes were not assessed in this study, the authors have
decreased the number of SLAP repairs and increased the number of biceps tenodeses because we feel
patients more than 35 years of age who have SLAP repairs become more stiff and have more pain than
those who undergo biceps tenodesis. In the majority of patients, there does not appear to be a clear
indication to perform a SLAP repair over a biceps tenodesis, hence the trends seen in this study.

Although some argue that SLAP repairs restore arm function better than biceps tenodesis, Chalmers
et al. proved this to be inaccurate. The authors evaluated 18 pitchers (7 uninjured controls, 6 after a
SLAP repair, and 5 after a subpectoral biceps tenodesis) and found that pitchers who underwent a
SLAP repair had altered patterns of thoracic rotation compared with the controls and pitchers who
had undergone a biceps tenodesis.

Laughlin et al. similarly found altered mechanics in 13 collegiate and professional pitchers who
underwent SLAP repairs compared with a group of control pitchers. Furthermore, the results of SLAP
repairs in nonoverhead throwing athletes have also been poor, especially in patients older than 40
years of age. Hence, even in high-level athletes, biceps tenodesis is a reliable option compared with
SLAP repair. In a busy subspecialty-based, referral shoulder practice, SLAP repairs currently represent
a little more than 6% of the authors’ surgical practice compared with slightly less than 30% for biceps
tenodeses. Because overtreatment of SLAP tears may result in increased complications such as
stiffness, persistent pain, and need for revision surgery, these data may be helpful for comparative
purposes regarding current indications for SLAP repair. Every patient should be treated on an
individual basis, but the future treatment of SLAP tears will likely see an increase in biceps tenodesis
and a decrease in SLAP repairs based on the outcomes reported in the literature and the high risk of
failure and complications seen with SLAP repairs.

Limitations

The strengths of this study include the use of a database of more than 9,000 patients by 4 busy
fellowship trained orthopaedic surgeons. The limitations of this study include the lack of outcome
variables to determine if patients who underwent SLAP repairs performed better than those who
underwent a debridement, although this was not the focus of this article. Unfortunately, the exact
number of patients who underwent biceps tenodesis for a SLAP tear could not be reliably reported
because of lack of description regarding the status of the biceps labral complex in some patients
undergoing a biceps tenodesis. Although the dates of surgery were scrutinized to ensure multiple
codes used for a single patient used on the same day were not counted as multiple surgeries, there is
a possibility that some were counted twice. The most common shoulder CPT codes were chosen to
create the denominator in determining the rate of SLAP repair versus other shoulder arthroscopic
procedures. Because not all arthroscopic shoulder CPT codes were included, the actual rate of SLAP
repair compared with all shoulder arthroscopies is likely lower.

Conclusions

Over the past 10 years, the total number of biceps tenodesis has increased, whereas the number and
relative percentage of SLAP repairs within our practice have decreased. The average age of patients
undergoing SLAP repair is decreasing, and most SLAP repairs are performed for type II SLAP tears.

THE DIAGNOSIS, CLASSIFICATION, AND TREATMENT OF SLAP LESIONS


SLAP (superior labrum, anterior and posterior) lesions have become increasingly recognized at a
source of pain and instability in the shoulder. Andrews et al described lesions of the biceps tendon in
high-level throwers in 1985. The mechanism of injury was described as a traction overload to the
labrum by the biceps tendon during the deceleration phase of throwing. Snyder et al coined the term
SLAP lesion (superior labrum anterior to posterior) to describe fraying or detachment of the labral
complex at the superior margin of the glenoid and classified these lesions into 4 subtypes.

Clinical Features

The clinical features of SLAP lesions have been described in several studies. The most common clinical
complaint is pain, increased with overhead activity and a painful “catching” or “popping” in the
shoulder. In the throwing athlete, a SLAP lesions can present as “dead arm” syndrome.

Mechanism of Injury

The mechanism of injury resulting in a SLAP lesion is varied. Repetitive throwing, hyperextension, fall
on an outstretched hand, heavy lifting, and direct trauma have all been implicated. The most common
mechanism in the study by Snyder et al was a fall on an outstretched arm, whereas others have
reported that two thirds of patients sustained a traction injury. Burkhart et al have described
contracture of the posterior-inferior glenohumeral ligament inthrowers. The glenohumeral contact
point is shifted posterosuperior, and increased shear forces are placed on the posterior-superior
labrum, generating a peelback effect and the SLAP lesion

Physical Examination

Preoperative diagnosis of SLAP lesions is aided by physical examination. Tenderness to palpation at


the rotator interval can be a helpful diagnostic sign. The most common positive physical findings are
a positive anterior drawer (53%), positive apprehension at 90° of abduction and maximal external
rotation (86%), and positive relocation test (86%). The Speed’s test and O’Brien signs are helpful for
diagnosing anterior lesions, whereas Jobe’s relocation test is positive most frequently with posterior
lesions. Speed’s test and Yergason’s test are not consistent predictors of a SLAP lesion.

Prevalence and Associated Lesions

In 1995, a review of 2375 shoulder arthroscopies performed by Snyder et al showed 140 superior
labrum injuries, representing an incidence of 6% of all shoulder arthroscopies performed at their
institution. In this study, 21% of lesions were Type I, 55% were Type II, 9% were Type III, and 10% were
Type IV, with 5% of lesions being “complex.” An isolated SLAP lesion was seen in 28% of patients, with
associated lesions consisting of partial thickness rotator cuff tears in 29%, full-thickness tears in 11%,
and Bankart lesions in 22%.

In patients under 40 years old who showed signs and symptoms of instability after a history of acute
trauma, repetitive injury, fall on an outstretched arm, or an injury from heavy lifting, SLAP lesions were
found in 52% of patients. Isolated SLAP lesions were found in 19%, an associated Bankart lesion was
found in 21%, full-thickness rotator cuff tears in 3%, undersurface rotator cuff tear in 3%, Bankart and
undersurface rotator cuff tear in 4%, and grade II impingement in 4%. No patients had a Type I lesion
and 93% of the SLAP lesions were Type II. More recently, Kim and coworkers found SLAP lesions
according to the Snyder classification system in 26% of 544 consecutive shoulder arthroscopies.
Ofthese shoulders, 74% were Type I, 21% were Type II, 0.7% were Type III, and 4% were Type IV. Type
II SLAP lesions in patients under 40 years old were associated with a Bankart lesion, whereas SLAP
lesions in patients over 40 were associated with a tear of the supraspinatus tendon and osteoarthritis
of the humeral head. Maffet et al believed that Type I lesions, or superior labral fraying, represented
normal aging and not likely to lead to shoulder dysfunction.

Normal Anatomy In identifying a SLAP lesion, it is essential to understand the normal anatomy and
variation. Cooper et al described the anatomy of the biceps anchor. Hyaline cartilage covers the
superior rim of the glenoid. The biceps tendon inserts directly into superior labrum. There is a
sublabral sulcus at the labral-bicipital junction that is normal in 73% of patients and should not be
misdiagnosed as a traumatic lesion. Essential to every arthroscopic examination ofthe shoulder is
reflection ofthe biceps attachment at the superior glenoid (Fig. 1A). This maneuver can reveal
otherwise unsuspected trauma at the attachment of the biceps to the labrum. Fissuring in the
sublabral recess, cracks in the articular cartilage of the superior glenoid, and impaction fractures of
the articular cartilage of the superior humeral head are all suggestive of a SLAP lesion. In addition,
widening of the rotator interval (Fig. 1B) and a positive drive-through sign can be clinical features of
the diagnostic arthroscopy that may suggest an associated SLAP lesion.

Important variations in the normal anatomy (Fig. 2) of the labrum have been identified. Three distinct
variations occur in over 10% of patients: an isolated sublabral foramen (Fig. 3), a sublabral foramen
with a cord-like middle glenohumeral ligament (Fig. 4), and a cordlike middle glenohumeral ligament
without tissue at the anterosuperior labrum (Fig. 5). Recognition of these normal variants can help to
avoid misdiagnosing labral lesions.

Classification

Snyder et al initially classified SLAP lesions into 4 categories. Type I is characterized by fraying and a
degenerative appearance of the superior labrum and is an uncommon source of clinical symptoms.
(Fig. 6). Type II is a detachment of the superior labrum from the supraglenoid tubercle (Fig. 7) Type III
is a bucket handle tear that displaced in to the joint while the biceps root remains stable (Fig. 8). Type
IV is a buckethandle tear where the tear propagates into the biceps tendon (Fig. 9). Three subtypes of
Type II SLAP lesions have been identified.6Of 102 Type II lesions observed, 37% were anterior (Fig. 10),
31% posterior (Fig. 11), and 31% combined anterior and posterior (Fig. 12). Maffet and coworkers
described further types of SLAP lesions. A Type V lesion is described as an anterior-inferior Bankart
lesion that propagates superiorly to the biceps tendon (Fig. 13). Type VI SLAP lesion is an unstable flap
tear of the labrum with separation of the biceps anchor (Fig. 14). A Type VII lesion is a superior biceps-
labral detachment that extends anteriorly beneath the middle glenohumeral ligament (Fig. 15).
Recently Nord and Ryu have added several previously unclassified lesions to the schema. A Type VIII
SLAP lesion is a SLAP extension along the posterior glenoid labrum as far as 6 o’clock (Fig. 16). A Type
IX lesion is a pan-labral SLAP injury extending the entire circumference of the glenoid (Fig. 17). A Type
X lesion is a superior labral tear associated with posteriorinferior labral tear (reverse Bankart lesion)
(Fig. 18).

Treatment

Isolated SLAP lesions are best treated with the patient in the lateral decubitus position. Anesthesia
can be either a general anesthetic or an interscalene block. We have had success with interscalene
blocks and favor it in all patients. The block is typically done in the preoperative holding area. After
sedation with 2 mL midazolam (Versed; Baxter, Deerfield, IL), a nerve stimulator is used to localize the
brachial plexus. Twenty milliliters of lidocaine 1% are injected first, followed by 20 mL mepivicaine
(Sensorcaine; Astra-Zeneca, Wilmington, DE) 0.5% with epinephrine and bicarbonate added to
increase the duration and shorten the onset of the anesthesia. A propofol (Diprivan; Astra-Zeneca)
drip can be used intraoperatively to maintain sedation, or a general anesthetic can be combined with
the interscalene block. In the lateral decubitus position, the patient is stabilized with a bean bag and
kidney rests. A pillow is placed between the legs and a second pillow beneath the leg on the table just
distal to the fibular head to avoid injury to the peroneal nerve. The shoulder traction device is placed
on the ipsilateral side of the involved extremity at 30° of abduction and 0° to 20° of forward flexion,
depending on the location of the lesion. A sterile bump can achieve further abduction as needed. The
extremity is placed in balanced suspension with 10 lb of traction. The procedure begins with palpation
of the bony landmarks about the shoulder. The posterolateral corner of the acromion is identified and
the tip of the coracoid and anterolateral corner of the acromion anteriorly. A standard posterior
incision is made 2 cm inferior and 1 cm medial to the posterolateral corner of the acromion. With a
finger palpating the coracoid tip, the obturator and cannula are directed anterior and superior toward
the tip of the coracoid. The elastic resistance of the posterior capsule is felt at the tip of the obturator
and overcome with manual pressure, yielding a pop as the tissue is penetrated. A thorough
examination of the glenohumeral joint is undertaken with attention to a widened rotator interval and
positive drive-through sign, tears of the subscapularis, biceps tendon subluxation, Bankart lesions,
impaction fractures of the cartilage of the humeral head, and posterior Bankart lesions. Care is taken
to externally rotate the extremity to examine for partial-thickness undersurface rotator cuff tears and
full-thickness tears. In evaluation for a SLAP lesion, it is essential to probe the biceps tendon anchor
to examine for laxity, fissuring, and separation of the tendon from the superior rim of the glenoid.
Pathology to the biceps anchor can easily be missed without probing the tendon. An 18-G spinal
needle is placed outside in from a point 1 cm medial to the anterolateral edge of the acromion at an
angle of 45° to the glenoid. The needle enters the glenohumeral joint through the rotator interval and
is used to identify anterior superior cannula placement. The cannula and obturator can be used to lift
the biceps tendon superior from its anchor on the supraglenoid tubercle (Fig. 19A,B). Fissuring of the
articular cartilage adjacent to the biceps-labral complex, fraying of the undersurface of the
attachment, and frank traumatic avulsion of the root can be revealed by this simple maneuver to what
appears to be a normal biceps attachment. After anterior portal placement, the obturator is replaced
into the arthroscope cannula posteriorly, and the arthroscope is placed through the anterior portal to
view posterior capsulolabral structures to complete the examination.

Portal placement for treatment of a SLAP lesion is dictated by the pathology discovered in the
glenohumeral joint. For the most common SLAP lesion, the Type II lesion, 3 portals are commonly
used. The standard posterior viewing portal is initially established. A high rotator interval portal is
created with an outside in technique using an 18-G spinal needle beginning 1 cm medial to the
anterolateral corner of the acromion, through the rotator interval tissue adjacent to the biceps’ exit
to the intertubercular groove. From this high angle, excellent purchase of a suture anchor into bone
can be achieved. If the angle to the glenoid is too low, the anchor can either skive off bone or enter
the glenoid in the osteochondral junction resulting in poor purchase of the anchor and delayed
fragmentation of the cartilage.

The third portal for the isolated Type II SLAP lesion is a midlateral portal, a transrotator cuff portal, or
a posterolateral portal (Port of Wilmington). The Port of Wilmington portal is created with an outside-
in technique using an 18-G spinal needle. The needle enters the skin just off the lateral edge of the
acromion at the midpoint of its anterior posterior dimension at a 45° angle to the acromion. The
needle passes through the musculotendinous junction of the supraspinatus or infraspinatus and into
the glenohumeral joint just posterior to the biceps root. As the needle enters the joint visualized from
the posterior viewing portal, it is important to be certain the needle enters the joint as close as
possible to the humeral head. This allows the cannula to be placed at a distance from the tissue that
is being repaired and out of the way of the viewing portal. Nord et al have described the use of the
Neviaser portal for SLAP repair. The portal is a superior medial portal bordered by the
acromioclavicular joint, the clavicle, and the spine of the scapula and can be used to access the
superior biceps-labral tissue by introducing a penetrating suture grasper to perforate the superior
labrum at a 90° angle and pass the suture from the anchor through the labrum. The suture is then
retrieved out the superior medial cannula and then tied.

Mobilization of the biceps-labral tissue is essential for achieving a stable repair. The Type II SLAP lesion
must often be completed with liberators to free the biceps from scar tissue from initial injury or from
an incomplete avulsion secondary to microtrauma (Fig. 20A,B). Preparation of the bony bed is
achieved with a shaver followed by a rasp (Fig. 20C). Some authors use a burr to remove the leading
edge of the articular cartilage, whereas others prefer a basket or biter to expose the bony edge of the
superior glenoid (Fig. 21A). A bed of bleeding bone is essential to creating a stable repair.

Burkhart et al have emphasized the need to place anchors at the corner of the glenoid at the articular
cartilage interface to restore the normal configuration of the biceps anchor (Fig. 21B,C). Suture-anchor
techniques vary according to author. Biodegradable anchors, PEEK or metal anchors may be selected.
The anchor may be placed first followed by passage of a limb of suture, or the suture may be passed
first and then threaded through the anchor and the anchor placed second. Nam and Snyder have
advocated a single-anchor, double-suture technique for SLAP repair. One anchor is placed at the root
of the biceps into bone with a suture passed posterior and a second one anteriorly over the biceps
root.

The authors prefer a 2-anchor technique using the Smith and Nephew (Andover, MA) Bioraptor PK 2.3
anchors which are non-absorbable. The guide is placed through the anterior portal and onto the
prepared glenoid. Anchors are placed at a 45° dead man’s angle to the articular surface for best
mechanical fixation. The guide is held in position and the drill hole made with the appropriate drill bit
equipped with a stop to assure accurate depth for anchor fixation. While holding the guide in position
over the drill hole, the drill is removed and the anchor placed through the guide into the drill hole.
Depending on which anchor is used, the device is either screwed or tapped into bone with a mallet.
Gentle traction tests the purchase of the anchor. A suture retriever is placed through the midlateral
portal and the sutures captured and passed out of the midlateral portal (Fig. 22A). Next, a 22° Birdbeak
(Arthrex) is passed through the anterior portal, through the anterior biceps-labral tissue, and opened
to capture the most superior limb of the suture (Fig. 22B,C). As the grasper is backed out of the
cannula, it is important to hold the opposite limb of the suture at the midlateral portal to avoid
unthreading the anchor (Fig. 23A,B). The suture retriever is brought through the anterior portal to
capture the second suture limb from the midlateral portal and brought out the anterior portal (Fig.
23C). With 1 limb of suture passing over the tissue and 1 limb through the tissue, an arthroscopic knot
is ready to be tied. If a mattress stitch configuration is desirable, the suture retriever is simply passed
a second time through the labrum to capture the second suture limb, retrograding the limb through
the labrum as well. One author (SEP) prefers to tie with the 6th Finger Knot Pusher (Arthrex) (Fig. 24).
This device keeps the knot tensioned during knot tying while alternating half hitches. This permits
expedited knot tying because the need to change posts is obviated and the posts can be alternated by
simply applying traction to “flip” the half hitch. This knottying device also allows efficient past pointing
when tying sutures, thereby maximizing knot security. Knot tying should be performed behind the
labrum and away from the glenoid to protect the articular surfaces from injury. The posterior SLAP
repair is a variation of the anterior repair. The anchor is place through a midlateral portal or
posterolateral Port of Wilmington portal (Fig. 25A). A higher-angled Birdbeak is used through the
midlateral portal and placed through the labrum to capture the suture (Fig. 26A,B). Knot tying is then
accomplished as previously described.
Alternatively, the Penetrator (Arthrex) can be used through the Neviaser portal for passing suture
posterior or anterior to the biceps (Fig. 27A-C). Type III lesions can be treated with a simple
debridement of the bucket handle portion of the biceps anchor and superior labrum. In some cases,
if the labrum is not too badly frayed, the bucket-handle lesion can be repaired to the biceps anchor
much like a meniscal repair. Type IV lesions are most commonly treated with debridement of the
bucket handle portion of the tissue and repair of the biceps anchor as needed. If the bucket-handle
component is viable, a repair of the bucket-handle segment can accompany the reattachment of the
biceps anchor. If the tear extending into the substance of the biceps involves greater than 50% of the
diameter of the tendon, biceps tenodesis or tenotomy is recommended. Repair of Type V SLAP lesions
include fixation of the biceps root, as in a Type II lesion, and continuing with anchors to incorporate
the Bankart lesion. Type VI lesions are best treated with debridement of the flap tear and fixation as
in a Type II lesion. Type VII lesions include fixation of the superior labrum as in a Type II lesion and
suturing of the middle glenohumeral ligament (MGHL) through the anterior rotator interval portal.
Type VIII lesions are treated as a Type II SLAP lesion with fixation of the posterior reverse Bankart
lesion. An accessory posterolateral portal, 1 cm lateral to the standard posterior portal, can facilitate
posterior anchor placement. The guide is placed initially with a blunt trochar to the posterior capsule
and exchanged for a sharp trochar to pierce the posterior capsule. The Type IX lesion, or panlabral
lesion, is treated with anchors anterior, posterior, and superior with a capsular shift. Type X lesions
are treated as Type VIII lesions. In the setting in which glenohumeral internal rotation deficit is the
primary cause of the SLAP lesion and a preoperative posterior-inferior capsular stretching program
has not improved internal rotation, a posterior capsular release in conjunction with a Type II SLAP
lesion repair is indicated.

Accessory Portals for SLAP Repair

As SLAP tears progress further posteriorly as seen in Types VIII and IX, the Port of Wilmington is
effective from 12 to 2 o’clock on a left shoulder or 10 to 12 o’clock position on a right shoulder (Fig.
28). For posterior-superior SLAP lesions that extend from the 2 o’clock to 5 o’clock (left) position
posteriorly, or 10 to 7 o’clock (right), an accessory posterior portal is necessary. This can sometimes
be accomplished without a cannula by using a 2-mm incision 1 cm lateral to the standard posterior
portal and introducing a FASTak Spear (Arthrex) after locating the proper angle with a spinal needle
(lateral accessory portal). After the FASTak is inserted, the suture can be passed through the labrum
using a Penetrator through a separate 2-mm incision 1 cm medial to the posterior portal (medial
accessory portal). The angle of retrieval prevents tension on the labrum while the suture is retrieved.
Both ends of the suture are then retrieved through the anterosuperior portal and the knot is tied using
the standard knot-tying device. A mattress suture can be placed by passing the Penetrator a second
time through the medial accessory posterior portal. If visualization is compromised, the camera can
be switched to an anterior portal and a clear plastic cannula placed through the posterior portal will
allow use of the Birdbeak or similar retrieving device. At the most inferior extent of a Type VIII, IX (pan-
labral tear) or X SLAP lesion, a cannula through the lateral accessory posterior portal can facilitate
access to this difficult to reach region.

Rehabilitation Protocol

Surgery is performed on an outpatient basis. Postoperatively, the shoulder joint is injected with 20 mL
0.5% Marcaine if an interscalene block is not used. The patient is placed in a DonJoy Ultrasling II (Vista,
CA) immobilizer for the first 3 weeks. During the first 4 weeks after surgery, the patient begins passive
forward elevation, and full elbow range of motion is permitted. Abduction and external rotation is
avoided. Passive and active range of motion to 90° of flexion is the goal during weeks 4 through 6.
Active forward flexion beyond 90° and forceful active biceps contraction should be avoided for
approximately 6 weeks postoperatively. If a posterior-inferior capsulotomy is performed, sleeper
stretches are started immediately postop to stretch the posterior capsule.

After the sixth week, full range of motion and gradual strengthening is performed. Posterior rotator
cuff strengthening should be emphasized. Overhead activities or strenuous biceps activity should be
avoided for 12 weeks. At 12 to 16 weeks, physical therapy is discontinued and the patient may return
to normal activities. Gentle interval throwing is resumed at 4 to 5 months. Full-velocity throwing is
allowed at 6 months on a level surface and from a mound at 7 months. Return to unrestricted
overhead sports is permitted at 8 to 9 months postoperatively.

Conclusion

SLAP lesions are becoming identified more commonly as a source of pain and instability in the
shoulder. Before the advent of arthroscopy and with it the ability to evaluate the biceps anchor, it is
likely that these lesions went untreated. Now, with a careful arthroscopic examination, well-planned
portals, and meticulous surgical technique, SLAP lesions can be safely and effectively treated with the
reasonable expectation of returning to normal activities, including high-demand throwing.

TREATMENT FOR SYMPTOMATIC SLAP TEARS IN MIDDLE-AGED PATIENTS COMPARING REPAIR,


BICEPS TENODESIS, AND NONOPERATIVE APPROACHES: A COST-EFFECTIVENESS ANALYSIS

Purpose: To evaluate the cost-effectiveness of nonoperative management, primary SLAP repair, and
primary biceps tenodesis for the treatment of symptomatic isolated type II SLAP tear. Methods: A
microsimulation Markov model was constructed to compare 3 strategies for middle-aged patients
with symptomatic type II SLAP tears: SLAP repair, biceps tenodesis, or nonoperative management. A
failed 6-month trial of nonoperative treatment was assumed. The principal outcome measure was the
incremental cost-effectiveness ratio in 2017 U.S. dollars using a societal perspective over a 10-year
time horizon. Treatment effectiveness was expressed in quality-adjusted life-years (QALY). Model
results were compared with estimates from the published literature and were subjected to sensitivity
analyses to evaluate robustness. Results: Primary biceps tenodesis compared with SLAP repair
conferred an increased effectiveness of 0.06 QALY with cost savings of $1,766. Compared with
nonoperative treatment, both biceps tenodesis and SLAP repair were cost-effective (incremental cost-
effectiveness ratio values of $3,344/QALY gained and $4,289/QALY gained, respectively). Sensitivity
analysis showed that biceps tenodesis was the preferred strategy in most simulations (52%); however,
for SLAP repair to become cost-effective over biceps tenodesis, its probability of failure would have to
be lower than 2.7% or the cost of biceps tenodesis would have to be higher than $14,644. Conclusions:
When compared with primary SLAP repair and nonoperative treatment, primary biceps tenodesis is
the most cost-effective treatment strategy for type II SLAP tears in middle-aged patients. Primary
biceps tenodesis offers increased effectiveness when compared with both primary SLAP repair and
nonoperative treatment and lower costs than primary SLAP repair. Level of Evidence: Level III,
economic decision analysis.
Type II SLAP tears are the most common lesion affecting the shoulder glenoid labrum. This lesion
affects approximately 6% of the U.S. population, and in symptomatic patients, it can cause significant
pain and disability.

Arthroscopic SLAP repair is a commonly performed procedure for patients with type II SLAP tears who
fail to improve with nonoperative management. This procedure has shown significant improvements
in pain and shoulder outcomes in young patients, but several authors have reported a higher failure
rate in older patients (older than 35-40 years). Biceps tenodesis is an alternative surgical treatment
for type II SLAP tears, especially in older patients and in patients with failed SLAP repairs. Biceps
tenotomy, although considered another alternative treatment for SLAP lesions, has been described
for low demand, elderly patients. However, with SLAP tear in a young, active patient biceps tenotomy
may be less acceptable because of weakness, cosmetic deformity (“Popeye sign”) and risk of biceps
cramps. Recently published evidence from a randomized controlled trial by Schrøder et al. comparing
primary SLAP repair versus primary biceps tenodesis for the treatment of symptomatic type II SLAP
tears in middle-aged patients (mean age 40 years) found no significant differences in function, patient
satisfaction, or complications. There continues to be ongoing controversy regarding the optimal
treatment for isolated type II SLAP tears, and there is limited evidence from a health-economic
perspective regarding the different treatment options for SLAP tears. The purpose of this study was
to evaluate the cost-effectiveness of nonoperative management, primary SLAP repair, and primary
biceps tenodesis for the treatment of symptomatic isolated type II SLAP tear. The hypothesis was that
primary biceps tenodesis would be a more cost-effective treatment option compared with primary
SLAP repair and nonoperative management for type II SLAP tears in middle-aged patients.

Methods

Design

A microsimulation Markov model was constructed to compare three treatment strategies: (1) primary
SLAP repair, (2) primary biceps tenodesis, and (3) continued nonoperative management. The base case
was a 40-year-old patient with a symptomatic isolated type II SLAP tear who previously failed a 6-
month trial of nonoperative management. The Markov cycle length was 1 year, derived from mean
time to surgical failure in the published literature. Surgical failure was defined as persistent pain in
patients who were “unsatisfied” for at least 6 months after the index surgical procedure, as previously
defined in the literature. For each of the treatment strategies, health states and simulated transitions
between health states were defined based on literature review and the senior surgeons' (A.S.R.,
M.S.V.) recommendation. Longitudinal health and economic outcomes were evaluated for each
treatment strategy. Health states were assigned quality of life (i.e., utility) and cost values to capture
the cumulative net gain or loss in these measures over a 10-year time horizon.

Model Parameters and General Assumptions The principal outcomes were quality-adjusted life years
(QALY, or each year alive multiplied by the utility, or quality of life, associated with the health state
experienced in that year), costs, and incremental costeffectiveness ratios (ICERs). The ICER represents
the benefit of one treatment strategy over another. It is defined as (C1-C0/E1-E0), where C1 and C0
are the respective costs of competing strategies and E1 and E0 are respective effectiveness in QALY.
The net monetary benefits (NMB) were used for sensitivity analyses. The NMB combines cost,
effectiveness, and willingness-topay (WTP) into one single measurement, such that the strategy with
the highest NMB value is the most cost-effective. In essence, the NMB is used when more than 2
strategies are compared or when the difference in effectiveness between 2 strategies is small, because
minute differences in effectiveness cause instability in the ICER value. The equation for calculating
NMB is NMB = E x WTP - C. The WTP is interpreted as the maximum monetary amount an individual
using a health care service is willing to pay for 1 additional QALY. In this study, a WTP of $100,000 per
QALY was selected, per literature recommendation. A treatment strategy is cost-effective if its ICER is
below the WTP threshold. Societal and health care payer perspectives were used for base case
analyses, and guidelines from the Panel of Cost-Effectiveness Analysis in Health and Medicine were
followed. Future health care system costs and QALY were discounted annually at 3% to represent
preference for money and health now rather than in the future. The following general assumptions
were made: (1) all patients in the model had failed an initial trial of conventional nonoperative
treatment for 6 months; (2) all patients who underwent index surgery (SLAP repair or biceps
tenodesis) and reported “poor satisfaction” would undergo subsequent nonoperative treatment; (3)
patients with failed subsequent nonoperative management after primary SLAP repair or primary
biceps tenodesis who opt for surgery would undergo salvage or revision biceps tenodesis, respectively;
and (4) all patients lived through the 10 years of follow-up.

Model Structure

Surgical strategies were followed by 5 health states: (1) success after primary SLAP repair or biceps
tenodesis; (2) success of postsurgical nonoperative management; (3) failure of postsurgical
nonoperative management; (4) success after revision surgery; and (5) poorly functioning, painful
shoulder (Fig 1). The index surgery (SLAP repair or biceps tenodesis) is the initial event from which
patients transition to successful or failed surgical health states. Patients with successful surgical
interventions either remain in that state (health state 1) or proceed to a failed transition state at an
annual probability defined by estimates from the scientific literature. Patients in the failed state
proceed to have postsurgical nonoperative treatment leading to either success (health state 2) or
failure (health state 3). Patients in failed states after postsurgical nonoperative treatment who opt for
surgery then undergo salvage or revision biceps tenodesis, resulting in success (health state 4) or
failure. Patients in failed states after salvage or revision surgery transition to a terminal state of having
a painful, poorly functioning shoulder (health state 5). On the other hand, continued nonoperative
treatment (strategy 3) was defined using 2 health states: (health state 1) success after continued
nonoperative management; and (health state 2) failure after nonoperative management.

Transition Probabilities

A structured review of literature was performed to derive transition probabilities for the model.
Medline and Embase databases were searched using the following inclusion criteria: (1) English
written peer-reviewed articles published within the last 10 years (2007-2017); (2) patients with
diagnosis of an isolated type II SLAP lesion; (3) mean patient age between 34 and 55 years; (4) no
history of prior SLAP repair; and (5) patients were not elite contact-sports athletes. Systematic reviews
from Erickson et al., Gorantla et al., Abraham et al., and more recently published individual studies
were examined based on the inclusion criteria. A total of 16 articles were identified (12 SLAP repair
[25,142 cases], 7 biceps tenodesis [15,299 cases], and 3 nonoperative treatment [124 cases]). The
study population mean age was 43.8 +/- 6.4. Abstracted data from the referenced articles were used
to calculate the annual failure probabilities: SLAP repair (6.27% +/- 5.57%), biceps tenodesis (2.93%
+/- 2.82%), and nonoperative management (9.44% +/- 9.82%). Annual reoperation probabilities were
also abstracted from the included studies and are presented in Table 1. The failure probability of
salvage or revision biceps tenodesis was derived from Gregory et al. series, which reported a 7% failure
rate after revision biceps tenodesis. Based on the series by Gupta et al. and Gregory et al., a
reoperation rate of zero after revision or salvage biceps tenodesis was used. Individual study event
rates were converted to 1-year transition probabilities using the following equation: p = 1 - e -rt, where
p is the probability, r is the rate, and t is the unit of time.

Health-Related Quality of Life (Health Utility)

Health-related quality of life for each health state in the model was incorporated using utility weights
on a scale from 0 (death) to 1 (full health). The utility score is then multiplied by the amount of time
in the specific health state to determine the QALY. Age-related baseline utility values were derived
from the EuroQol 5-dimension (EQ-5D) general-health outcome scores reported in a randomized
controlled trial by Schrøder et al. (Table 1). The EQ-5D instrument, which has been validated for use
in patients with SLAP lesions, measures 5 dimensions of health status (mobility, selfcare, usual
activities, pain/discomfort, and anxiety/ depression) with 3 levels per dimension (no problem, some
problems, and extreme problems). Utility values for patients with failure after primary surgery or
failed continued nonoperative management were assumed to be equal to the preoperative utility
values.

Utility decrements (i.e., disutility) reflected one-time, year-long losses of quality of life associated with
the inconvenience of undergoing surgical procedures or having persistent pain within the first
postoperative year. Because of the lack of evidence for SLAP surgery to inform these parameters,
published evidence from Dornan et al. and Genuario et al. was used to approximate disutility values
of 0.015 and 0.10, to represent first postoperative year and revision or salvage surgery, respectively.
These values were comparatively smaller than those used for rotator cuff surgery in the
aforementioned studies. Based on the recommendation from the senior surgeons (A.S.R., M.S.V.), a
one-time utility decrement of -0.05 was assigned to all patients who reported persistent pain or
stiffness resulting in poor satisfaction at 6 months after surgery, eventually requiring additional
treatment. Patients with successful salvage or revision biceps tenodesis surgery were deemed to have
a 10% decrease in utility compared with successful primary surgeries. The utility value assigned to the
terminal state of pain and poor shoulder function was 0.55, comparatively smaller than preoperative
utility reported by Schrøder et al. (0.60), to account for persistent dissatisfaction after undergoing
unsuccessful operative and nonoperative management. Published evidence from Luo et al. was used
to define the U.S. minimal clinical difference in EQ-5D scores (0.04 +/- 0.026).

Costs

All costs were reported in 2017 U.S. dollars estimates and adjusted for inflation using Consumer Price
Index when necessary. Direct costs were obtained from the Centers for Medicare and Medicaid
Services (CMS) 2017 Outpatient Prospective Payment Systems and Medicare Physician Fee Schedule
Databases. The Outpatient Prospective Payment Systems and Medicare Physician Fee Schedule reflect
Medicare reimbursement to the hospital and the physician, respectively, per specific American
Medical Association Current Procedure Terminology code. The following Current Procedure
Terminology codes were used: 29807 (arthroscopic SLAP repair), 29828 (arthroscopic biceps
tenodesis), and 23430 (open biceps tenodesis). Final direct procedure costs were calculated by adding
Medicare hospital and physician fee for each of the strategies. For biceps tenodesis, where both
arthroscopic and open techniques are commonly used with reported similar outcomes, the average
of the 2 costs for the base case analyses was used. The cost of salvage or revision biceps tenodesis
was estimated multiplying the base cost of the primary index surgery by a factor of 1.05, consistent
with previous cost-effectiveness studies. The hardware cost variability was assumed to be absorbed
by hospitals as parts of operating costs and not directly seen by CMS or private insurance. The cost of
postoperative rehabilitation ($1,902) consisting of 12 physical therapy sessions was derived from prior
published work by Mather et al.39 CMS reimbursement for nonoperative treatment was estimated
based on published evidence by Edwards et al. (18 physical therapy sessions; $2,854). The lost
productivity (work absenteeism) associated with each treatment option was calculated using the
following formula:

Expected cost of productivity loss = (MWDs=TWD

X MI x (1 – UR)

where MWDs are the missed work days, the TWD are the total working days in one calendar year, the
MI is the median U.S. income for workers between 35 and 64 years old, and UR is the unemployment
rate in the United States. For all surgical interventions, a total of 13 missed work days associated with
the rehabilitation protocol were estimated (5 full work days for surgical recovery, 2 days to
accommodate 4 office visits, and 6 days to accommodate 12 physical therapy sessions). Missed work
days associated with nonoperative treatment after surgical failure were approximated at 11 days (2
days to accommodate 4 office visits, and 9 days to accommodate 18 physical therapy sessions). Total
working days in 1 calendar year were assumed to be 240. After inflation adjustment, the weighted
median U.S. income for the study population was $68,401. The U.S. unemployment rate in the
aforementioned age group for year 2017 is 3%. Accordingly, the 1-year loss of productivity for
undergoing surgery and requiring subsequent nonoperative treatment was $3,586.50 and $3,034,
respectively.

Analyses

TreeAge Pro Suite 2017, R2.0 (TreeAge Software) was used to construct the Markov model and
conduct analyses. Excel (Microsoft) was used for basic calculations. The ICER and NMB for each
treatment strategy were calculated with a Monte Carlo analysis using 50,000 first-order
microsimulations, which approximated the incidence of SLAP repair in the United States over the 10-
year follow-up period.

Sensitivity analyses, performed from a societal perspective, were conducted to determine model
robustness and assess the impact of each variable on the base case results. The model was considered
“sensitive” when the optimal treatment decision changed as the variable(s) was varied through a
meaningful range. Medicare and abstracted data from each of the included studies were used to
derive low and high limits for sensitivity analyses (Table 1). In 1-way sensitivity analyses, variables
were considered robust if the base case optimal option (most cost-effective) did not change. Two-way
sensitivity analyses were performed on variables that were found to influence the model. For the
probabilistic sensitivity analysis (PSA), gamma distributions were used for costs, whereas beta
distributions were used for probabilities and known utility values. Direct cost’s standard deviations
were obtained from Medicare data. Productivity losses parameter’s standard deviation (+/- 50%) was
approximated based on the plausible ranges of missed work days. Disutility parameters, modeled with
uniform distributions, were varied by +/- 50%. Finally, PSA was performed assigning 1,000 second-
order samples over 10,000 first-order cycles.

When comparing interventions, an alternative is “dominant” when it is both more effective and less
costly than the comparator; the comparator is then considered “dominated.” The alternative is
“preferred” when it is more effective and below the WTP threshold, which is defined here as
$100,000/QALY gained, that is, the alternative costs less than $100,000 per QALY gained when using
the alternative treatment instead of the comparator.

Results

Base Case Analyses


Microsimulation analysis from the societal perspective showed that continued nonoperative
treatment was the least costly alternative ($5,888), yielding the least QALY (4.80) over 10 years of
follow-up. Biceps tenodesis dominated the primary SLAP repair strategy and is considered cost-
effective with an ICER of $3,344/QALY gained compared with continued nonoperative management.
Biceps tenodesis costs less on average ($13,044) compared with primary SLAP repair ($14,810), and
also yielded more QALY (6.94 QALY) compared with primary SLAP repair (6.88 QALY). Similar results
held when using a health care payer perspective (Table 2).

One- and Two-Way Sensitivity Analyses

Univariate (1-way) uncertainty of all model parameters and assumptions showed that the model was
sensitive to the following variables: (1) probability of failure after primary SLAP repair; (2) probability
of failure after nonoperative management; (3) utility of successful nonoperative management; (4)
utility of successful primary SLAP repair; (5) utility of successful biceps tenodesis; and (6) cost of biceps
tenodesis. Threshold analysis results showed that when the annual probability of failure of SLAP repair
was below 2.7% or when the cost of biceps tenodesis was above $14,644, SLAP repair was the
preferred strategy. Threshold results for all the variables that showed sensitivity are presented in
Table 3.

A 2-way sensitivity analysis was used to compare the utility of success after each of the competing
strategies using the fixed base case utility value of 0.80 for biceps tenodesis and varying the utility of
SLAP repair and nonoperative management over clinically plausible ranges (Fig 2A). Biceps tenodesis
is preferred for most of plausible utility ranges, with primary SLAP repair gaining preference when its
utility score is above 0.813. Lastly, a 2-way sensitivity analysis using the annual probability failure of
SLAP repair and biceps tenodesis showed that biceps tenodesis was the favored option within the
range of clinically plausible probabilities (Fig 2B).

Probabilistic Sensitivity Analysis

Biceps tenodesis was the preferred strategy in 51.8% of simulations, meaning that it conferred
increased effectiveness at an acceptable cost compared with the competing strategies (SLAP repair
and nonoperative management) at a WTP threshold of $100,000/QALY. In 40% of simulations, primary
SLAP repair was considered cost effective at the WTP threshold. The PSA indicated that the optimal
strategy was somewhat sensitive to model inputs; however, most simulations showed that tenodesis
was preferred (Fig 3).

Model Validation

The 10-year model was externally validated by comparing predicted SLAP repair reoperation rates and
direct costs with those reported in the literature but not used for informing the model inputs. The
model predicted that 2.9% of patients would require reoperation after primary SLAP surgery, which is
within the range of published evidence by Taylor et al. (2.5%) and Mollon et al. (3.3%).

Discussion

This cost-effectiveness analysis shows that primary biceps tenodesis is a cost-effective strategy for
treating symptomatic isolated type II SLAP tears in patients between 34 and 55 years old. Base case
analysis showed dominance (increased effectiveness and decreased costs) of biceps tenodesis when
compared with SLAP repair. Under base case conditions, when compared with nonoperative
treatment, biceps tenodesis was cost-effective when using both societal and health care payer
perspectives.
The reported annual rates of poor outcomes after SLAP repair have ranged from 3.2% to 27%. In this
study, results from sensitivity analyses suggest that an annual failure rate of -2.7% would have to be
achieved for SLAP repair to be preferred over biceps tenodesis. A similar annual failure rate was
reported by Neri et al. with patients over the age of 40 at a 3-year follow-up period. When comparing
biceps tenodesis with continuing nonoperative treatment, sensitivity analysis showed that an annual
failure probability -1.5% would have to be achieved for nonoperative treatment to be more
costeffective than biceps tenodesis. The option of continuing nonoperative treatment has been
studied by Edwards et al., Shin et al., and Schrøder et al., and there is considerable variation in the
reported failure rates after nonoperative treatment (1.3% to 20.8%). Patients' individual
characteristics such as age and activity level (type of work, overhead activity, type of sports, and
athletic requirements) have been shown to affect the outcomes. For example, Shin et al. found that
15% of the patients in their series (mean age of 39, all patients nonoverhead or competitive level
athletes) had unsatisfactory clinical outcomes with nonoperative treatment and subsequently
required surgical treatment. In contrast, Edwards et al. reported that more than half the patients in
their series (mean age of 34 and +50% were overhead athletes) failed nonoperative management and
eventually required surgery.

Although sensitivity analyses showed robustness in annual probability of biceps tenodesis failure
across clinically plausible ranges (0% to 5.4%), the model was sensitive to biceps tenodesis surgical
costs, favoring SLAP repair over biceps tenodesis when biceps tenodesis surgical costs were greater
than $14,644. It is important to note that the aforementioned cost represents the maximum allowable
biceps tenodesis direct surgical costs ([hospital + physician fee]; Medicare median cost: $6,295)
without including physical rehabilitation or productivity loss associated costs. Varying all other cost
parameters across Medicare published evidence and literature derived ranges did not impact the
model conclusions.

Based on the randomized controlled trial by Schrøder et al., all successful competing strategies (SLAP
repair, biceps tenodesis, and continuing nonoperative treatment) were assigned a health utility of
0.80. The uncertainty analysis showed that success after nonoperative management was the preferred
treatment strategy if its utility value was >0.88, whereas primary SLAP repair was preferred if the
utility value of the base case was >0.81 or if the utility of biceps tenodesis decreased to <0.79;
otherwise biceps tenodesis was the preferred treatment.

The aforementioned values are all quite close, indicating that the results are sensitive to utilities of
these health states. Although the 0.08 and 0.02 EQ-5D differences between tenodesis versus
nonoperative treatment and tenodesis versus SLAP repair, respectively, are within the range of
published minimal clinically important difference in EQ-5D scores in the United States (0.04 +/- 0.026),
this finding points to the relevant impact that minute changes in validated quality of life measures can
have in decision analysis models.

Results from the PSA indicate that the general results hold in 52% of simulations and that biceps
tenodesis is the preferred strategy based on the selected WTP threshold of $100,000/QALY gained.
The interpretation of QALY is based on the premise that 1 QALY is equivalent to 365 quality-adjusted
life days; thus a 0.06 increase in QALY (difference between biceps tenodesis and SLAP repair)
translates to approximately 22 quality-adjusted life days. Because the goal of costeffectiveness studies
is to maximize health when deciding between competing strategies, from a QALY perspective, biceps
tenodesis is a “preferred” treatment for the base case scenario. As the U.S. health care system shifts
to value-based medicine, cost-effectiveness analyses can inform decision making so that resources are
allocated optimally, yielding the greatest benefit from treatment options, given their cost. The results
of this study have important clinical implications. There is a growing body of evidence that supports
primary biceps tenodesis as an effective alternative treatment to SLAP repairs in middle-aged and
older patients. This study showed that biceps tenodesis is more cost effective than SLAP repair in this
patient population. The present finding conforms with emerging data revealing high patient
satisfaction and low revision rates in middle-aged patients undergoing primary biceps tenodesis.
Further enhancements to the costeffectiveness model would include updates of input values for
failure probabilities and clinical outcome scores for biceps tenodesis in younger population as new
high level evidence becomes available. The current literature shows that SLAP repair is the most
common treatment in younger age groups despite a less predictable return to play in overhead
athtletes. However, the optimal treatment selection is a shared decisionmaking process by the patient
and physician, and involves multiple considerations including patient characteristics such as age,
activity demands, and patient expectations.

Limitations

The results of this study should be interpreted with the following inherent limitations. First, the data
inputs were derived from Levels II-V evidence, because at the date in which the present study was
completed there were no Level I studies comparing the treatments for isolated type II SLAP tear.
Furthermore, biceps tenotomy, which is a less commonly performed procedure for this study’s target
population, was not included as a treatment option. Second, the study by Gottschalk et al., which was
included as source data for the present model, had patients in their series with concomitant lesions
that required additional procedures. Others have evaluated the effect of addressing associated
pathological abnormalities at the time of SLAP repair and have found a minimal clinical effect on
patient outcomes. Third, the cost estimates represent a national average for treatment costs. Decision
makers may want to geographically adjust costs according to their health care system. Fourth, the
costs of lost productivity may have been underestimated, as patient travel and transportation costs
for receiving treatment were not included. In addition, the missed work days were estimated using
U.S. societal perspective; in Europe a much longer recovery time period after surgery is common.
However, the robust sensitivity analysis conducted around cost variables is believed to have captured
the potential impact of cost variability in the overall study conclusions. Finally, the influence of
additional factors that could potentially affect the outcome after the treatment of SLAP tears like
patient activity level and type, presence of associated shoulder lesion, and workers’ compensation
were not considered.

Conclusions

When compared with primary SLAP repair and nonoperative treatment, primary biceps tenodesis is
the most cost-effective treatment strategy for type II SLAP tears in middle-aged patients. Primary
biceps tenodesis offers increased effectiveness when compared with both primary SLAP repair and
nonoperative treatment and lower costs than primary SLAP repair.

S-ar putea să vă placă și