REVIEW ARTICLE

Shoulder & Elbow

2017, Vol. 9(1) 46–53
! The Author(s) 2016
The treatment of multidirectional Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
instability of the shoulder with a DOI: 10.1177/1758573216652087
sel.sagepub.com
rehabilitation programme: Part 2

Lyn Watson1,2, Sarah Warby1,2, Simon Balster1,

Ross Lenssen1,2 and Tania Pizzari2

Abstract
Background: The most commonly recommended initial treatment for multidirectional instability is a rehabilitation
program. Although there is evidence to support the effect of conservative management on this condition, the published
literature provides little information on the exercise parameters of such programs.
Methods: This paper is the second part of a two-part series that outlines a six-stage rehabilitation program for
multidirectional instability with a focus on scapula control and exercise drills into functional positions. This paper outlines
stages 3 to 6 of this rehabilitation program.
Results and Conclusions: This clinical protocol is currently being tested for efficacy as part of a randomized con-
trolled trial (Australian New Zealand Clinical Trials Registry #ACTRN12613001240730). The information in this paper
and additional online supplementary files will provide therapists with adequate detail to replicate the rehabilitation
program in the clinical setting.

Keywords
exercise, multidirectional instability, scapula, shoulder, rehabilitation
Date received: 19th April 2016; accepted: 26th April 2016

Preliminary research has shown that this program

Background
significantly improves instability specific outcomes,
The most commonly recommended initial treatment shoulder muscle strength and scapular upward rotation
for multidirectional instability (MDI) of the shoulder of patients with MDI4,5 and the program is currently
is a rehabilitation program.1–3 This paper is the being evaluated in a randomised controlled trial (RCT)
second part of a two-part series that outlines a (Australian New Zealand Clinical Trials Registry
rehabilitation program for MDI. Part 1 of this two- #ACTRN12613001240730).
part series outlined stages 1 to 2 of the Watson MDI
Program, which included the assessment of faulty
scapula and humeral head (HH) biomechanics,
rehabilitation to re-establish scapula control (particu-
larly upward rotation) and control of glenohumeral 1
LifeCare Prahran Sports Medicine Centre, Prahran, Victoria, Australia
joint range in lower levels of elevation. The current 2
La Trobe University, College of Science, Health and Engineering, School
paper will present stages 3 to 6 with a focus on estab- of Allied Health, Department of Rehabilitation, Nutrition and Sport,
lishing scapula and HH control and progression of Bundoora, Victoria 3086, Australia
exercises into functional ranges. The successful com-
pletion of the initial stages of the rehabilitation pro- Corresponding author:
Sarah Warby, LifeCare Prahran Sports Medicine Centre, 316 Malvern
gram (part 1) is imperative for the MDI patient to Road, Prahran, Victoria 3181, Australia.
ensure adequate scapula and glenohumeral head con- Tel: þ61 394793906.
trol to progress through stages 3 to 6. Email: [email protected]
 Watson et al. 47

The watson mdi program: intervention

Stage 4: Sagittal plane and coronal plane control
An overview of stages 3 to 6 of the Watson Program is from 45 to 90 of elevation
outlined in Table 1 and detailed flow charts of stage 3 The aim of stage 4 is to progress movement control to
to 6 are provided in the Supporting information. Part 1 90 of elevation in the coronal and sagittal plane. Stage
described the assessment of the patient’s faulty scapula 4 is divided into two phases: the scapula phase and the
and HH biomechanics to determine what scapula and/ arc of motion phase.
or HH position that the patient must retrain and main-
tain throughout the program. Scapula control is often
facilitated with a scapula resistance (SR) band (Figure Scapula phase. The scapula phase involves gaining con-
1). The SR band is placed around the patient’s scapula trol of the scapula in an extension row motion from 45
and can be used to resist upward rotation, elevation to 90 of abduction, from a red to green TheraBandTM
and/or posterior tilt.4,6 and even heavier bands (blue/black) if functionally
required for the patient (Figure 2a).

Stage 3: Flexion control from 0 to 45 of elevation Arc of motion phase. The arc of motion phase is subdi-
vided into 3 parts: internal rotation (IR) and external
The aim of stage 3 is to achieve control in the sagittal rotation (ER) at 90 of elevation, flexion at 90 of ele-
plane (flexion) from 0 to 45 of elevation. Flexion is vation, and horizontal extension into horizontal
important to introduce because it is a very functional flexion.
motion and can usually commence once a standing
extension row at 45 (to neutral) has been established
IR and ER at 90. The extension row drill at 90 of abduc-
against a green TheraBandTM (Hygenic Corporation,
tion prepares the patients for ER at 90 (Figure 2b).
Akron, OH, USA) for 20 repetitions. Flexion has a
Patients with significant anterior instability may need
high activation of serratus anterior and thus is essential
to initially reduce their arc of ER or bring the drill
for improving serratus anterior strength and therefore
forward into the plane of the scapula, working
upward rotation of the scapula.7 Care must be taken in
around into the coronal plane once control of the HH
patients with a component of posterior instability
is sufficient. IR at 90 is usually commenced once the
because flexion drills can cause an increased translation
patients can perform 20 repetitions of ER at 90 with a
of the HH if adequate posterior buttress and control
red TheraBandTM.
has not been established. Stage 3 is divided into two
phases: the scapula phase and the arc of motion phase.
Flexion at 90 . Flexion drills are performed to 90 of
elevation. Load is determined by the functional and
sporting requirements of the patient.
Scapula setting phase. The scapula setting drill that was
established in stage 1 (Part 1) is progressed in incre-
ments from a standing position in the coronal plane Horizontal extension into horizontal flexion. Horizontal flex-
to a standing position in the sagittal plane. The load ion causes a large degree of posterior HH translation
for this drill may need to recommence with the weight and needs to be controlled for activities such as taking
of the arm and be progressed in 0.5-kg increments as off a tight top, driving a car, and sporting actions such
control is established. as a backhand in tennis. The horizontal extension into
horizontal flexion drill is a progression from the stand-
ing extension row at 90 elevation. The drill requires the
Arc of motion phase. Flexion drills are performed with a patient to perform a horizontal extension movement
TheraBandTM anchored behind the patient when they from a starting position of relative horizontal flexion.
perform a forward punching motion (Figure 1). Short Over a number of weeks, the drill is progressed into a
arcs of motion are commonly performed with a yellow starting position of more horizontal flexion by the
or red TheraBandTM and progressed to a green patient gradually turning their body in increments
TheraBandTM prior to extending the arc. The flexion towards their affected shoulder until they are in a start-
drills begin at approximately 20 to 30 of shoulder ing position with their arm across their chest (Figure 3).
flexion and are progressed to 45 . Patients with a TheraBandTM resistance is slowly increased at each
large component of posterior instability may have to angle of horizontal flexion until a load is achieved
commence flexion drills in the scapula plane and pro- that is comparable with the patient’s functional
gress in increments around to the sagittal plane. requirements.
 48
Table 1. Overview of the Watson MDI Program: Stages 3 to 6.

Stage 3

Scapula Scapula control in Scapula UR in 1 to 3  20, 2  day 2  20 Scapula UR þ/–

phase sagittal plane standing þ /– posterior 0 kgto 0.5 kgto 1 kgto tilt with 1 kg to 2 kg
tilt working from cor- 1.5 kgto 2 kg in sagittal plane
onal plane to sagittal þ/– red–green TB 2  day
plane band for tilt

Arc of motion phase Arc of motion control Flexion in the scapula 1 to 3  20 repetitions, 1 or 2  20 repetitions Palpate the HH during
in sagittal plane plane 0 to 45 eleva- 2  day* flexion drill in the flexion to assess
tion Yellow–red–green TB* sagittal plane to 45 unwanted HH pos-
Flexion in the sagittal with load appropri- terior translation
plane 0 to 4 elevation ate for function* If patient loses scapula
or HH control
regress to stage 2
for more
strengthening

Stage 4

Scapula Scapula control to 90 Standing Ext row from 45 1 to 3  20 repetitions, 1 to 3  20 repetitions
Phase Abd to 90 Abd 2  day* green TB standing
Red–green TB* Ext row 90 Abd

Arc of motion Arc of motion control ER/IR drills: 1 to 3  20 repetitions, 1 to 3  20 repetitions ER/IR can be per-
phase to 90 Abd ER at 90 2  day* ER/IR at 90 red/ formed between 0
IR at 90 Red–green TB* green* and 90 (e.g. 45
Flexion drills: Weights: 0.5 kg to 1 to 3  20 repetitions Abd) if functionally
Flexion to 90 4 kg* flexion to 90 red/ required by the
HE to HF drills: green* patient
Standing Ext row at 90 1 to 3  20 repetitions Abd ER at 90 per-
working around to HF HF red/green* formed initially in
þ/– SR band scapula plane if
anterior instability

Stage 5

Posterior, middle and Posterior: BOR at 0 to 1 to 3  8 to 20 repe- 1 to 3  8 to 20 repe- Bent rows performed
anterior deltoid 45 to 90 Abd titions titions, 0 kg to 4 kg* to neutral Ext
muscle strength Anterior: Flexion with TB 1  day to 3  week* Posterior deltoid drills
(stage3), sitting/ Red–green–blue–black* performed first as

(continued)
Shoulder & Elbow 9(1)
 Table 1. Continued
Watson et al.

standing short lever 0 kg to 0.5 kg to 1 kgto easier to control

flexion with weight. 4 kg* HH, followed by
Middle: Short lever Abd anterior then
30 to 90 with weight middle deltoid drills

Stage 6

Arc of motion Arc of motion con- ER> 90- EROM* Recruitment/endur- 20 ER/IR/flex/in Integration of trunk
trol > 90 Abd/elev IR > 90-EROM* ance (1 or 2  20 range/load to mimic stability and overall
Flex > 90-EROM* repetitions, 2  day) part practice kinetic chain with
Deltoid drills > 90* to strength (3  10 shoulder drills
to 12 repetitions) to needs to be
ballistic (1 or considered
2  10 þ repeti-
tions) *
Yellow–red–green–
blue TB*
Weight for deltoid: 0
kg to 0.5 kg to 1 kg*

Part practice and Part practice of func- Part Practice (example): Part practice dosage Return to sport/occu-
whole practice tion and integration Catch phase of swim over and load needs to pation/function
into sport/func- swiss ball mimic demands of
tional tasks Whole Practice: task
Participation in train- Whole practice pro-
ing/sport/occupation gressed from small
volume to larger
volume

Repetitions of exercises held for 3 seconds to 5 seconds. Abd, abduction; BOR, bent over row; ER, external rotation; EROM, end range of motion; Ext, extension; HE, horizontal extension; HF, horizontal
flexion; HH, humeral head; IR, internal rotation; SR, scapula resistance; TB, TheraBandTM; UR, upward rotation. *Dosage and load can be progressed from a recruitment and endurance dosage to a dosage
and load functionally required by the patient. Exercises may need to be progressed to blue or black bands or heavier weights if functionally required by the patient. For the detailed flow charts for the Watson
MDI Program, see the Supporting information.
49
50 Shoulder & Elbow 9(1)

Stage 5: Isolated deltoid drills drills because they can cause posterior HH translation
The aim of stage 5 is to develop specific strength in the if they are not well controlled. A rolled up towel behind
anterior, middle and posterior deltoid at the same time the humerus can provide additional posterior support.
as maintaining scapula and HH control. Deltoid muscle Middle deltoid drills are performed initially in small
strength contributes to centering of the HH, as well as ranges of abduction with a short lever (Figure 4c),
normal shoulder kinematics.8 Guidelines for stage 5 are working to ranges that are required by the patient.
outlined in Table 1; however, drills may commence in
conjunction with earlier stages of the program for some
patients.
Stage 6: Sports-specific and functional specific stage
Bent over rowing drills for posterior deltoid (stage 2) The aim of stage 6 is to progress arcs of motion beyond
are progressed in load at 0 abduction. Simultaneously, 90 and then into task specific, occupational specific
bent over rowing drills are progressed in increments to and sports-specific drills. Stage 6 is divided into the
a bent over row at 90 abduction (Figure 4a). Anterior arc of motion phase and then the sports-specific/func-
deltoid drills are performed as a flexion action9 in tional specific phase.
supine, sitting or standing depending on the patient
(Figure 4b). Care must be taken prescribing supine
Arc of motion phase. Prior to commencing exercises that
mimic the details of a specific movement (i.e. tennis

Figure 1. Flexion drill with scapula resistance band. Figure 3. Horizontal flexion drill.

Figure 2. (a) Extension row at 90 of abduction. (b) External rotation at 90 of abduction.
 Watson et al. 51

Figure 4. (a) Bent over row at 90 abduction for posterior deltoid. (b) Flexion for anterior deltoid. (c) Short lever abduction for
middle deltoid.

compared to a person who needs to lift the occasional

heavy weight overhead for work (high load, lower repe-
titions). Depending on the sporting and occupational
demands of the patient, the program can emphasize
concentric and/or eccentric or ballistic (plyometric)
actions. Weight-bearing exercises (progressing from
wall weight bearing drills to full weight bearing drills)
can be utilized if the patient functionally requires them.
Weight bearing drills should not be prescribed when
any component of posterior instability remains.

Whole practice. Once the patient has gained control

of part practice drills, then whole practice execution of
the drills can be emphasized, with gradual return to
training and work, and with gradual increases in
Figure 5. Part practice of the pull phase of a freestyle stroke in volume, as the therapist considers appropriate.
swimming. Part practice of the catch phase and recovery phase
would also be implemented.
Maintenance programme. Once a patient has reached
their goals and completed the program, they are
serve), patients may need further steps of progression. encouraged to continue a maintenance program at
This often involves IR, ER, deltoid and flexion drill two or three times a week of four to eight exercises to
from 120 and up to end of range of abduction with maintain their level of function. This often involves the
the TheraBandTM in varying positions, emphasizing patient performing their weights exercises on one day
end of range upward rotation control. Integration of and band exercise on another day. It is important that
the kinetic chain must also be considered in this the patient understands any limitations that they may
phase if not prior.10 have as a result of their condition. For example, some
more severe cases of MDI require strict activity limita-
Sports-specific and functional specific ranges tions and should be advised not to return to certain
activities (e.g. contact sports) or should be advised to
Part practice. This stage requires the exercises to be limit activities (e.g. some gym exercises, amount weight
functional and should closely mimic the sport/activity. used and/or higher volumes of specific activities).
Breaking down drills into subcomponents can be useful
for the patient to gain control over the entire motion
Parameters
(Figure 5). Dosage at this phase needs to represent what
is functionally required of the patient. Consider the dif- Dosage. Dosage is based on the number or repetitions
ferences between part practice drills for someone need- that the participant can achieve pain free and with good
ing control at a computer (low load, high repetitions), scapula control. Exercises typically commence with a
52 Shoulder & Elbow 9(1)

recruitment dosage (1 to 3 sets of 20 repetitions per- spine posture, poor deep cervical flexor strength, symp-
formed 2 or 3 times a day), followed by an endurance tomatic thoracic outlet syndrome,13 rotator cuff inflam-
dosage (1 to 3  10 to 15 repetitions 1 or 2 times a day), mation, volition instability, and significant levels of
and moving onto a strength dosage in later stages (3 to pain. The treating physiotherapist needs to be aware
4 sets of 8 to 12 repetitions performed every second of these issues and address them accordingly.
day).11 For most exercises, repetitions are held for 3
seconds.
Conclusions
Load. Load with weights typically commences with 0 kg This paper outlines the last four stages of the six-stage
(the weight of the arm) and progresses in 0.5 kg incre- Watson MDI program. The Watson MDI Program
ments to a minimum of 2 kg for most exercises. focuses on developing scapula and HH control prior
TheraBandTM exercises typically commence with to exercises into range, and is completed with func-
yellow and progress to a minimum of red for females tional and sports-specific exercises. It provides therap-
and green for males. The progression of load depends ists with a flexible model on which they can deliver
on the functional and sporting requirements of the treatment, based on assessment and individual patient
patient. Most patients will need to be progressed presentation.
beyond to 2 kg and some patients beyond the green
TB (for load guidelines, see Table 1).
Declaration of Conflicting Interests
Progression of exercises. Progression of exercise drills The author(s) declared the following potential conflicts of
through the program can be achieved by increasing interest with respect to the research, authorship, and/or pub-
the arc of motion, increasing the load, changing the lication of this article: L. Watson has been teaching physio-
dosage or increasing the level of elevation in which therapy shoulder courses for over 25 years. This paper
the patient performs the drill. Table 1 and the flow outlines a rehabilitation program that is often taught on
charts (see the Supporting information) and previous these courses; therefore publication may strengthen the
series guide the therapist through a typical format of rehabilitation programme as a course resource. S. Balster,
S. Warby, T. Pizzari, and R. Lenssen are casually employed
progressions for the MDI patient. Clinical signs that
by L.Watson to assist with her shoulder courses. The authors
indicate a patient is ready to progress include: patients
declare that there are no financial interests in any company or
are able to perform their previously prescribed exercises institution that might benefit from the publication of the sub-
with no symptoms; the physiotherapist observes the mitted article. There are no competing interests relevant to
patient perform their current exercises with good scap- this publication.
ula and HH control; and the patient can maintain good
scapula and HH control when the physiotherapist iso-
metrically loads them in the position that simulates the Funding
proposed new drill. Exercise drills are continued The author(s) received no financial support for the research,
between stages of the program; however, dosage and authorship, and/or publication of this article.
load of earlier exercises may be progressed. For exam-
ple, a patient may be performing stage 4 ER at 90 of References
elevation with a red band and an endurance dosage but 1. Warby SA, Pizzari T, Ford JJ, Hahne AJ and Watson L.
will be continuing ER at 0 with a blue band with a The effect of exercise-based management for multidirec-
strength dosage. tional instability of the glenohumeral joint: a systematic
review. J Shoulder Elbow Surg 2014; 23: 128–42.
2. Bahu MJ, Trentacosta N, Vorys GC, Covey AS and
Pathological limitations Ahmad CS. Multidirectional instability: evaluation and
Patients with MDI may have a primary direction of treatment options. Clin Sports Med 2008; 27: 671–89.
instability.12 This is an important consideration when 3. Guerrero P, Busconi B, Deangelis N and Powers G.
Congenital instability of the shoulder joint: assessment
prescribing exercises. A patient with a predominance of
and treatment options. J Orthop Sports Phys Ther 2009;
posterior instability will load into flexion at a later
39: 124–34.
stage, whereas a patient with a predominance of anter- 4. Watson L. Multidirectional instability of the shoulder: what
ior instability will need to load into abduction and end is it? Does physiotherapy work? What should we be doing?
range ER at a later stage. Therapist assessment must Melbourne: New Moves, Australian Physiotherapy
guide exercise selection. Conference, 2013.
Patients with MDI often present with comorbidities 5. Watson L. Functional and clinical changes in multidirec-
that may need to be addressed prior to commencing tional instability of the shoulder after conservative rehabili-
shoulder rehabilitation. These include poor cervical tation. PhD Thesis: La Trobe University, Bundoora, 2015.
 Watson et al. 53

6. Ganderton C. A novel way to train subscapularis: 17–20 10. Saeterbakken AH, van den Tillaar R and Seiler S. Effect
October 2013. Melbourne: New Moves Australian of core stability training on throwing velocity in female
Physiotherapy Association, 2013. handball players. J Strength Cond Res 2011; 25: 712–8.
7. Decker MJ, Hintermeister RA, Faber KJ and Hawkins 11. Kraemer WJ, Adams K, Cafarelli E, et al. American
RJ. Serratus anterior muscle activity during selected College of Sports Medicine Position Stand. Progression
rehabilitation exercises. Am J Sports Med 1999; 27: models in resistance training for healthy adults. Med Sci
784–91. Sports Exerc 2002; 34: 364–80.
8. Wickham J, Pizzari T, Stansfeld K, Burnside A and 12. Cordasco FA. Understanding multidirectional instability
Watson L. Quantifying ‘normal’ shoulder muscle activity of the shoulder. J Athlet Train 2000; 35: 278–85.
during abduction. J Electromyogr Kinesiol 2010; 20: 13. Watson LA, Pizzari T and Balster S. Thoracic outlet syn-
212–22. drome part 1: Clinical manifestations, differentiation and
9. Kronberg M, Nemeth G and Brostrom LA. Muscle activ- treatment pathways. Manual Ther 2009; 14: 586–95.
ity and coordination in the normal shoulder: an electro-
myographic study. Clin Orthop Rel Res 1990; 257: 76–85.