Instability of the shoulder – a neurological disease

Stability: definition

Stability of any articulation (defined as asymptomatic normal mechanical behaviour at rest and in motion) depends on the retention of structural integrity, intact neural control systems (afferent, efferent and neuromuscular connectivity), and the absence of strain exceeding the yield point for the whole system. For the glenohumeral joint (GHJ) stability is therefore the product of a functionally (not necessarily anatomically) intact rotator cuff, competent capsular and labral structures, a sufficient surface arc (or surface area) of contact between humeral head and glenoid, and an intact neuromuscular system comprising central and peripheral connections, with functioning mechanoreceptors clustered in the capsule and labrum, and in the tendons of the rotator cuff (RC) and long head of biceps (LHBT).

The essence of stability and the stability equation

Motion about the centroid of rotation and ʻcontainmentʼ of the humeral head on the size-mismatched glenoid is the result of concavity compression, generated by the centralizing joint reaction force (CJRF) of the rotator cuff (including LHBT), and the generation of negative intra-articular pressure, facilitated by the conformity of the glenoid/labral surface. Rotator cuff competence is a function of centrally driven activation via feed-forward systems, and the strength and inertia of the cuff muscles. Shoulder joint position sense and motion are the products of right parietal cortical programming (the parietal reach region), and visual, vestibular and cervical afferent inputs: it appears that visual cues are more important than hand position in determining the control of shoulder motion. Maintenance of this activity is achieved through feedback mechanisms derived through mechanoreceptors in the capsular and labral structures, rotator cuff tendons and muscles, and the surfaces of the shoulder gliding planes (the acromio-coraco-deltoid-scapulothoracic space, and the glenohumeral joint space), and affected by the inertia of the rotator cuff tendino-fibrous ʻskeletonʼ and muscles, and fascial and cutaneous afferents. The relationship between the dominant factors in creating stability in the GHJ can be illustrated by a simple diagram ( Fig. 6.3.1 ), and described by an equation:

(where CJRF is the centralizing joint reaction force, and the elements RC = rotator cuff, SAC = surface arc (area) of contact, CLMNS = capsulolabral mechano-nociceptive system, NS = central and peripheral neural control system elements, and a–d are factors describing the weight of each element in the overall mechanism of stability for that individual at a specific point in time).

The relative importance of each factor varies according to the individual case, but several points are common.

First, the anatomy of the rotator cuff is relatively invariant (although subscapularis can vary in the ʻheightʼ of its attachment to the lesser tuberosity). The number of muscle spindles (MS, predominantly stretch receptors) in rotator cuff muscles has been studied in marsupials and is approximately 1.5 times the number of Golgi tendon organs (GTO; predominantly contraction receptors) ( , ). GTOs are situated predominantly in the musculotendinous junctions and at the regions of confluence between the tendon and capsule of the joint. Both muscle spindles and GTOs have large myelinated, fast conducting afferents. The GHJ will remain stable in the partial absence of medial capsulolabral structures (for instance after surgical release for frozen shoulder syndrome, or complete capsulotomy for total shoulder replacement), provided that the rotator cuff function is normal. Thus the dominant proprioceptive input for glenohumeral stability would appear to be from the muscles and the tendinous endoskeleton of the rotator cuff. The LHBT has a special role in linking rotator cuff activity, humeral head position and overall upper limb position, particularly in activities requiring optimal close-packing of the shoulder and elbow. Anomalies of the intra-articular tendon and isolated rupture of the tendon do not appear to contribute to humeral head instability, provided the closely associated components of the rotator cuff remain intact. By contrast, a rotator cuff rupture in an elderly patient presents as superior GHJ instability.

Second, the SAC can vary. Thus, segmental dysplasia of the glenoid, congenital glenoid version anomalies and humeral head defects (Broca or Hill–Sachs, and McLaughlin lesions) all contribute to a decrement in the SAC, which predisposes to failure of the CJRF vector to pass through the glenoid during all arcs of motion.

Third, the structure of the CLMNS can vary considerably. The labrum can be absent, hypoplastic, or dysplastic, and ligament anomalies are well recognized (and common) in subjects with no instability. The incidence of ligamentous anomalies in patients with atraumatic structural instability is the same as that in the general population: lax individuals do not appear to have a greater prevalence of anomalies of ligament morphology, although the molecular structure is clearly altered. The labrum contains numerous free nerve endings, with a frequency about ten times greater than any other corpuscular receptor type ( ), which are readily distorted by motion.

The number of corpuscular mechanoreceptors in the capsule of a marsupial is equivalent to that of muscle spindles, with about 87% being small lamellated corpuscles (rapid-response stretch signalling in a local region) having relatively smaller myelinated afferents than either MSs or GTOs. Pacinian corpuscles (rapid-response stretch signalling over a wider field) are found in the outer fibrous laminae of the capsule, predominantly in the axillary region, while Ruffini corpuscles, which are uncommon, are found in the dense fibrous layer of the capsule in the axilla, and the dermis of the skin. When articular afferents are stimulated in the cat shoulder there is a very short latency (3 ms) to contraction in biceps and deltoid, suggesting a monosynaptic reflex at spinal cord level ( ). There therefore appears to exist a gradient of receptor speed, locality and density from within to without the joint. As is apparent from the preceding text, data is largely derived from non-human sources, but there is increasing clarity about the role of neural pathways in generating and maintaining shoulder joint stability ( ). It is not known whether variations occur in the densities or function of mechanoreceptors in the capsule and labrum in patients with atraumatic structural and muscle-patterning.

Fourth, neural control systems are readily deranged, by central or peripheral neurological conditions: aberrant muscle activation or suppression contributes to GHJ instability ( ), and diseases of the cerebellum and basal ganglia may present with shoulder instability, particularly of the scapulothoracic joint.

It is self-evident that we are not aware of our shoulder if it has not been injured or affected by a disease in some way, nor do we regard it during gestural and stereotactic activities. In other words, the presentation of our shoulder is unconscious: all activity is predicted, planned and executed without cognition. This lack of conscious awareness of the shoulder (or any other joint) seems likely to be a function of the entirety of the afferent data about the shoulder. Different combinations of afferent inputs must be produced during varying positions and actions involving the joint. However, providing the total afferent input remains within a range of normality for that joint at that particular time (as predicted by the purpose for which the joint action has been generated), there is no conscious awareness of the joint. If injured, there is awareness (i.e. a conscious, therefore cortical, event occurs, one usually signifying potential danger) and, if the injury is sufficient, what is clinically defined as ‘apprehension’ (i.e. a conscious, therefore cortical, event appreciated as a fear of actual danger) often accompanied by a reflex activation of muscles of the shoulder girdle region. If the injury is more deleterious still, and tissue disruption of a critical ‘amount’ occurs, then pain will occur (in biological evolutionary terms, the incentive to remove oneself from the source of the injury). The central pathways of awareness, apprehension and pain are similar. The total afferent input from the shoulder joint has been changed in some way, and if there has been a disruption of tissues then that region of disruption will likely be de-afferented. A change in overall afferent input, either an absolute loss or an alteration in the ratio of inputs from the various sources, results in both cortical and spinal level reflex (protective) activity. Since the central nervous system works on the principle of inhibition, the ‘permission’ of spinal activity activating a shoulder girdle response (‘apprehension’) is produced by an inhibition of the normal inhibitory pathways of motor activation patterns. The individual becomes aware (now at cortical level) of actions, motions or positions of the arm in space which will both generate pain (through distortion of the injured afferent neural tissue) and unpleasant sensations perceived as potential or incipient displacement of the articular surfaces. Restoration of the afferent input, by the healing of the disrupted tissues (whether by natural means or by surgery) results in less pain and less apprehension. That some patients often remark that their shoulder ‘feels’ better immediately after restorative surgery (if undertaken soon after the injury) suggests that the afferent pathways from spinal level to cortical level remain intact and non-adapted. That other patients, also undergoing restorative surgery, particularly after recurrent episodes of dislocation or failed healing, take a long time to recover to the point of no apprehension suggests not only that re-afferentation is relatively less well organized and quantitatively diminished, but also that central re-adaptation has to occur. That rehabilitation programmes designed to restore stability take roughly the same period as programmes for integrating tendon transfers suggests similar central neural mechanisms are at work. Muscle-patterning instability, the so-called Stanmore type III instability (see below), is a large spectrum of disorders of shoulder stability, characterized by a lack of structural abnormalities, likely to be a failure (by inhibition) of the normal inhibitory pathways of muscle control.

The increased risk of a traumatic GHJ dislocation in the first-degree relatives of patients with traumatic (i.e. structural) GHJ instability suggests a predisposition to the condition, even where the structure of the joint was previously apparently normal. Whether this is located in the afferent, central, or efferent components of the neural control mechanisms is uncertain, but the part of the ‘system’ that could most readily be perturbed is the afferent, in terms of the number, density and distribution of mechanoreceptors.

Instability: a clinical definition

Instability is usefully defined as the condition of symptomatic abnormal motion of the joint. This universal definition combines the mechanistic view and the clinical perspective, and distinguishes the pathological state from the constitutional state of laxity, in which anatomical anomalies may exist, but do not necessarily cause impairment of function. Since it is a symptom, the term implies recognition (a cerebral cortical event) of the abnormal motion (a difference in position, motion direction or motion velocity has been identified by the patient): ‘instability’ cannot be determined by examining the shoulder under anaesthetic (EUA). Laxity (looseness) of the GHJ can be determined by EUA either by comparison of the perceived abnormal with the perceived normal side or by reference to what is considered a normal range of laxity for the population of which the patient is representative.

The relationship between the dominant factors in creating instability in the GHJ can be described by a second equation, derived from the first:

where the elements RC , SAC , CLMNS and NS are as before (see above) and a–d are factors describing the weight of each element in the overall mechanism of instability for that individual at a specific point in time. P are patient-specific factors (occupational and sports, lifestyle, motivation in rehabilitation, etc.), and t represents the probability that pathology will evolve over time.

The pathologies causing instability comprise structural ( RC , SAC ) and non-structural ( NS ) elements. Many authors hold that the labrum has a structural role, particularly in anterior glenohumeral instability: by virtue of its content of noci-mechanoreceptor clusters and arborizations of free nerve endings, it clearly has a role in neural information loops and so also represents a non-structural stabilizing element. The structural elements may be developmentally abnormal, contain abnormal collagen, or acquire microtraumatic lesions over time (collectively defined as atraumatic structural elements) or damaged by extrinsic force (defined as traumatic structural elements). The non-structural elements can be developmentally abnormal (e.g. neuromuscular dyspraxias) or can be acquired over time as perturbations of neuromuscular control, particularly at periods of skeletal growth ( Fig. 6.3.2 ).

The shoulder in hypermobility syndromes

Laxity (or ‘looseness’) is defined as asymptomatic motion in a shoulder through a range that would not be considered normal for the patient’s age or gender. Laxity varies with age. Adolescents have a greater overall range of motion than adults, and stiffness (reduced compliance of rotator cuff and capsular tissues) increases with age. Shoulder laxity can exist as part of the generalized benign hypermobility syndromes, or as part of an upper limb dominant form of hypermobility, or in isolation. In the latter form, if one shoulder alone is affected then congenital bony architectural anomalies should be considered, whereas if bilateral isolated shoulder hypermobility is present then soft tissue anomalies are more prevalent.