Punctuated equilibrium and the reverse total shoulder replacement

Like many professionals living in a progressive world a fair portion of my time is spent poddling within a narrow field of specialisation. The frivolous comment that specialisation is about “knowing more and more about less and less” is also, depressingly, rather astute. (of course, if we take that observation to its ultimate conclusion, we reach that absurd singularity of knowing everything about nothing. But just as nature abhors a vacuum it also abhors the infinite: the infinitely small, the infinitely large, the infinite extrapolation. We live in a world of quanta - more accurately, we observe a world of quanta - but this is going rather beyond the point of this post). For those of us that pleasure our egos within a limited scope of endeavour we like to think of specialisation as the knowledge and execution of detail. At its most refined it is the art of nuance. And there’s plenty to be smug about here: a modern, sophisticated world needs its specialists. But humans have limited capabilities and time is finite. A life focussed on detail in one area can expose a broad ignorance about the goings-on in the rest of the world.

It matters because the world can change rather fast.

via http://en.wikipedia.org/wiki/Punctuated_equilibrium 

Punctuated equilibrium, as proposed by Niles Eldredge and Stephen Jay Gould, is a theory of evolutionary biology with a very specific application. But through the popular writings of Mr Gould, and for want of a better term, it is often misrepresented as the embodiment of a neat but much older concept (if you seek clarification here, you are going to be disappointed). The idea is that, for most of the time, um, not much really happens: ie stasis, status quo, same-same. But, on occasion, a significant change occurs. And it often appears to arrive rather suddenly.

To state the blindingly obvious: change happens all the time. Have a yawn. Take a stretch. Yada yada yada. But it mostly happens on a scale that we barely notice it. Sudden change appears to arise because the concept is novel (think the incandescent light globe). Or because the application is novel (think Facebook). Or simply because, as busy  individuals in a complex world, we remain detached from the process and somehow manage to miss all the bits in the middle (additive manufacturing and drone technology for those of us without engineering background nor inclination). A casual reflection on these examples suggests that it is often a medley of all these factors. 

Another marvellous but much less appreciated example is the reverse total shoulder replacement.

In general, the practice of orthopaedics has evolved towards the restoration of anatomy. That is to say, as surgical techniques and prosthetic designs improve, the concept of determining what is within acceptable limits (ie what we can get away with) has morphed into a conviction to restore normal anatomy (ie accepting that normal anatomy is the ideal and making all attempts to achieve this). This is particularly relevant in a rich, developed country with long-lived individuals demanding a high level of function for an extended period of time. In many cases this is not possible.

But that does not necessarily mean that there isn’t a solution. And the reverse shoulder replacement is one example of this. Although this form of shoulder replacement has been kicking around for about 20 years or so it continues to surprise me how few doctors outside the field know about it. And, indeed, how few patients appreciate the significance of it’s development. If you will allow me an indulgence: the reverse shoulder replacement is the “game-changer” or “disruptive technology” that revolutionised the clinical outcomes of a group of patients with shoulder arthritis.

To understand this we will (unfortunately) need to start with some basic anatomy. 

The shoulder is a ball and socket joint. The archetypal ball and socket joint is the hip joint which looks much like a ball mount and coupling used to hitch a trailer or caravan.

The shoulder joint is different in that the cup (socket or glenoid) is much smaller than the ball (the humeral head).

That means that the glenoid cup does not capture or “contain” the ball of the humeral head. In other words, the bony constraints of the shoulder joint (ie the ball and socket) are not inherently stable on their own. To maintain balance a number of soft tissue structures are required: principle among them being the dynamic stabilising effect of a group of tendons called the rotator cuff.

via http://physioworks.com.au/injuries-conditions-1/rotator-cuff-injuries#.UEwtZ47Unx4

In fact, the rotator cuff is so critical to shoulder function that it really deserves a few more images..

front view

side view

back view 

via The Interactive Shoulder

Shoulder arthritis with an intact rotator cuff can be treated with a standard total shoulder replacement. In this construct the normal anatomy of the shoulder is reproduced by replacing the arthritic joint with a metal ball and plastic (polyethylene) liner. 

via http://orthoinfo.aaos.org/topic.cfm?topic=A00094 

The problem is that rotator cuff tendons also degenerate with age: especially so from middle age onwards. When the rotor cuff tendons tear beyond a certain tipping point the ball becomes unbalanced within the socket causing problems which, over time, can lead to the development of a type of wear-and-tear arthritis called cuff-tear arthropathy. A standard joint replacement becomes unsuitable in this instance. Replacing the joint with one made of metal and plastic may tackle the problem of a worn-out, arthritic joint but if the cuff tear cannot be fixed (as is often the case) then the imbalance will persist and the prosthesis will fail. It is this form of cuff-deficient, shoulder arthritis that is treatable with a reverse total shoulder replacement. 

via http://orthoinfo.aaos.org/topic.cfm?topic=A00094 

The history of shoulder replacement surgery spans well over a 100 years and makes for an interesting, convoluted story with many notations, parenthesis, footnotes and dead ends. And to summarise is to diminish the contributions of many. Nonetheless such abbreviation helps to make a point. Modern shoulder replacement surgery started with an American chap by name of Charlie Neer. Neer replaced the ball (humeral head) with one made of metal in the 1950s and then went on to add a plastic liner for the cup (glenoid) in the 1970s. With solid, reproducible results, the 1970s marked the dawn of shoulder replacement surgery for the treatment of arthritis. 

if Péan was the first to publish in 1894 then this was surely the apogee

first illustration of a metal and plastic shoulder joint replacement - the “total” shoulder replacement

But during the late 70s and certainly by the start of the 1980s it became evident that joint replacement in shoulders with a deficient rotator cuff had much poorer outcomes than those with an intact cuff. A number of designs to tackle the difficult problem of joint imbalance and humeral head migration included more-constrained articulations, hooded (extended) glenoids, oversized humeral head replacements, and the reverse shoulder replacement. 

In short: they all failed. 

1970s reverse shoulder replacement via http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1437172/ 

Neer’s original reverse shoulder replacement via http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3148354/?tool=pubmed 

So the biggest problem in shoulder replacement surgery had been identified and isolated. By 1983 Charlie Neer, who had by then achieved a god-like stature in the eyes of other orthopaedic folk, regarded the treatment of cuff tear arthropathy as “limited goals surgery” (medical speak that translates to “epic fail” in the modern lexicon). But the French have a knack of seeing the world a little differently from the rest of us. And it took a Frenchman by name of Paul Grammont to see the problem in a novel way and make the reversed shoulder articulation work.

The reverse shoulder replacement places the ball where the socket (glenoid) of the shoulder normally resides and replaces the ball (humeral head) with a socket. The principle was to constrain an inherently unstable joint with one that looked much like a hip joint. But because the glenoid is too small to support a cup much larger than itself a few brave pioneers decided to reverse the articulation to get around the technicalities of prosthetic fixation in bone.

Grammont’s brilliance and insight was not to go about perfecting designs that already existed but to specifically identify and deal with the problems at hand:

1. Problem: a deficient rotator cuff means an unbalanced joint. Solution: a fully congruent joint with greater contact area improves stability (the shoulder and it’s modern anatomic replacement, unlike the hip, has a slightly mis-matched radius of curvature between ball and socket so as to allow a small degree of translation). 
2. Problem: the deltoid muscle is both the prime destabilizing force and the only muscle that can allow the shoulder to abduct (reach out to the side). Solution: (and Grammont’s epiphany) drop and medialise the centre of rotation. 

A reversed shoulder articulation just so happened to allow Grammont to tackle all these issues at the same time. But also introduced the now well-recognised problem of component loosening at the glenoid. Grammont argued that the problem of loosening could be addressed by placing the centre of rotation (COR) at the bone-prosthesis interface. Previous reverse shoulder replacements placed the COR close to the natural anatomic position which inadvertently introduced a toggle (moment arm) at the glenoid bone-component interface (see the examples above). This so-called “rocking-horse effect” magnifies small, repetitive forces and plays a significant role in the loosening of orthopaedic implants.

depiction of normal shoulder with deltoid muscle vertical and rotator cuff muscle horizontal 

(X = point of shoulder to top of humerus, Y = top of humerus to deltoid insertion, X+Y = resting length of deltoid)

depiction of shoulder with significant cuff tear and humeral head migration

depiction of reversed shoulder articulation with increased deltoid lever arm by medialisation of centre of rotation (M) and restoration of deltoid resting length 

via http://www.boneandjoint.org.uk (also note that the COR sits at the bone-prosthesis interface)

Grammont’s first reversed articulation design was this:

“Trompette” reverse prosthesis (1985) via http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3148374/?tool=pubmed 

Just 6 years later the design evolved to have all the major elements seen in current reverse total shoulder replacements:

“Delta-III” (1991) via http://www.maitrise-orthop.com/viewPage.do?id=806 

Although Grammont’s research was well-published in French scientific journals, his brilliant if non-anatomic solution was all but unknown to the outside world (honestly, with the exception of hopeless romantics and the French themselves, who the fcuk reads French?!). In 1993 Grammont published for the first time in the English literature (“Delta shoulder prosthesis for rotator cuff rupture” in Orthopaedics 1993 Jan;16(1):65-8). But, being weird-looking and superficially similar to a failed design (and not to mention being French), it took quite some time for English-speaking world to warm to the Delta-III reverse shoulder prosthesis. Ten years later, in the light of extensive use and impressive results coming out of Europe, the American FDA (Food & Drug Administration), a regulatory body which by necessity has a guarded approach to innovation, proved quick to grant approval for the use of the Delta-III prosthesis in the United States in November 2003. 

And since that time the use of reversed shoulder prosthesis has exploded. At a major shoulder conference in Nice (France) earlier this year it was estimated that, amongst the experienced shoulder surgeons attending, approximately 50% of all total shoulder replacements inserted were of the reversed variety.

via http://www.scholarpedia.org/article/Punctuated_equilibria 

I was quite excited when I first became aware of punctuated equilibrium. Not so much it’s specific (and only correct) application to explain findings in the fossil record, but as a concept of rapid change: a sudden trajectory diverging off a path of gradual progress. When I see it (or rather, when I think I see it) something in the back of my mind tingles. 

Yes, change happens all the time. And whether change occurs suddenly or gradually is often a matter of scale. To take a paleontological example: most would agree that the Cambrian Explosion happened pretty suddenly when taken on a geological time scale. But in human terms it lasted many, many, many generations. A time span so long that any single human observer of the period would have recorded the events unfolding as: “same-same” (if he wasn’t some primordial creepy-crawly unable to differentiate mouth from anus). But if you were Gleise 581 (aka Proxima Centuri, a red dwarf star in the nearby constellation of Centaurus) and been hanging around since the Big Kahuna you’d probably be thinking “boy, that went quick”. 

In the context of the human experience, sudden change appears to arise because the concept is novel. Or because the application is novel. Or because we remain detached from the process and somehow manage to miss all the bits in the middle. It is often a combination of all these factors.

The development of the reverse total shoulder replacement is, of course, more complicated than what I have attempted to describe here. But as far as disruptive technology goes it is as good an example (misrepresentation) of punctuated equilibrium as any. A hundred thousand years in the future when the human race is extinct some alien life form may well dig up our remains and see fossils with metal and plastic shoulder replacements. Rather suddenly, around 2000 AD, a shoulder replacement with a reversed orientation appeared.

My mind tingles.