Are we asking the right questions about failed total shoulder arthroplasty?

A recent publication The Rising Incidence and Future Trends of Revision Total Shoulder Arthroplasty sought to determine the epidemiology of revision TSA using a national database. The authors found that the incidence of revision TSA increased from 223 to 1247 cases per 100,000 person-years between 2015 to 2024, a 5.6-fold increase.

These data point to the steadily increasing number of patients having revision for arthroplasty failure, the economic burden of these revisions, and the need for an increasing number of surgeons who are equipped to manage these failures. Of importance is that these data reflect the incidence of revision per 100,000 people (not the rate of revisions for patients having shoulder arthroplasty). In that the number of shoulder arthroplasties performed per 100,000 people is steadily increasing each year, it is intuitive that the number of revisions per 100,000 people would also increase each year. 

It seems that the questions of greatest importance are not addressed in this study:

1. Are patients undergoing TSA today at greater risk for complications than patients a decade ago? If so, is this a consequence of applying shoulder arthroplasty to younger patients, patients with more complex diagnoses, and patients with poorer overall health? Or is this a consequence of less experienced surgeons performing an increasing number of these procedures?

2. What are the specific failure modes driving revisions, and are those modes changing over time? For anatomic TSA, are the rates of glenoid component loosening, instability, rotator cuff failure, and infection getting higher, lower or staying the same? For reverse TSA how are the rates of instability and dislocation, acromial and scapular spine fracture, infection, notching with component loosening, and periprosthetic fracture changing with time?

3. Which patient and surgical factors are associated with each distinct failure mode? Risk factors differ by failure mode; we need to understand these associations as we strive to reduce the risk of each type of failure.  Aggregating across all causes of revision in a single regression model — as most administrative database studies must do — destroys the signal. The predictors of infection-related revision are nutritional status, glycemic control, obesity, prior surgery, and prophylaxis adequacy. The predictors of instability-related revision in rTSA involve the surgeon's choices of component design, version and inclination and approaches to soft tissue tensioning. The predictors of acromial fracture involve bone mineral density, scapular morphology, type of arthritis, prior surgery, rotator cuff status, component lateralization, and deltoid tensioning. 

In that the surgeon is the method, surgeon knowledge, training, and experience deserve special study across all failure modes. These may be the most important modifiable risk factors — yet they receives minimal attention in a literature preoccupied with implant selection and surgical technology.

4. What might be done differently to prevent each type of failure? This is the most important question and the hardest to study, because it requires honest counterfactual analysis rather than statistical association. It asks not merely which factors correlate with failure, but which specific decisions — patient selection, perioperative optimization, surgical technique, rehabilitation protocol — were modifiable and, if modified, would plausibly have prevented the need for revision.

Answering this question at scale will be difficult, requiring mandatory registries with standardized failure mode coding and linkage between primary and revision procedures.

Recognizing that failure modes differ by surgeon and practice, an immediately applicable and practical approach is analogous to aviation crash analysis. Each surgeon analyzes their failures by comparing each revision to matched non-revised controls from their practice. (See How a surgeon can learn from their own adverse outcomes - an example of intrapractice analysis in reverse shoulder arthroplasty.)

A final question: is the most cost-effective approach to reducing the risk of arthroplasty complications and revisions (A) more expensive technology or (B) better surgeon training (see How much precision do we need to pay for in shoulder arthroplasty?)

A risk factor for small bird adverse outcomes in my backyard.

Cooper's Hawk

2021

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/

Here are some videos that are of shoulder interest
Shoulder arthritis - what you need to know (see this link).
How to x-ray the shoulder (see this link).
The ream and run procedure (see this link).
The total shoulder arthroplasty (see this link).
The cuff tear arthropathy arthroplasty (see this link).
The reverse total shoulder arthroplasty (see this link).
The smooth and move procedure for irreparable rotator cuff tears (see this link)
Shoulder rehabilitation exercises (see this link).

Mission Impossible? Reducing Risk of Revision for Failed Surgery

The Problem

Much of what we do in orthopaedic surgery has a high apparent success rate (85–90%). While encouraging, studies of shoulder arthroplasty, rotator cuff repair, and other interventions show that the proportion of patients achieving clinically meaningful improvements in comfort and function has not changed significantly for more than a decade.

This stagnation has persisted despite multiple innovations in technique, technology, implants, and biologics—which have been introduced often without clear evidence that they address the true causes of clinical failure.

Many of our colleagues believe that it is impossible to ferret out the causes of a failure requiring surgical revision because we were not there at the time of the original procedure. 

On the other hand, my collegue Chris Ahmad, points out that 

So, because of the importance of improving the quality of our patient care by reducing the revision rate, we go forward using the approach described in the recent post Surgical failures: what causes them and how can we do better for our patients. 

---

Step 1: Individual Case Analysis

Progress will depend on systematically examining the one in ten cases that require revision. The guiding question is:“Given what is known about this patient and procedure, what might have been done differently to reduce the risk of revision?”

Because the revising surgeon often did not perform the index surgery, she or he must reconstruct the case—much as aviation investigators reconstruct a crash. This requires integrating expertise and available evidence to specify plausible alternatives, knowing that many of the possible contributing factors cannot be accurately assessed (even if we have the "black box" or the operative note and preoperative images). 

Key elements we should pursue include:

• Patient and shoulder characteristics before the index surgery

• Details of the index procedure

• Surgical alternatives used in similar cases

• Findings at revision

• Recognized causal links between surgical choices and failure modes (e.g. inadequate fixation of the glenoid component).

---

Menu of Causes

To support structured causal reasoning, revising surgeons are presented with a menu of recognized causes of failure.

For anatomic shoulder arthroplasty, examples include:

• Inappropriate implant sizing

• Inadequate fixation

• Poor component positioning

• Insufficient balancing of the humeral head on the glenoid

• Other to be specified by surgeon

The revising surgeon selects the most plausible cause(s) or adds others, then assigns a rating to each:

• 3 = Very likely contributed

• 2 = Possibly contributed

• 1 = Unlikely

• 0 = No opinion / insufficient information

---

Embracing Uncertainty

This method explicitly acknowledges uncertainty. Surgeons are not asked for certainty, but for their best judgment, made transparent:

“I cannot know with certainty what would have happened with an alternative procedure, but given the case information and the published evidence, the most plausible cause of revision of this surgery is inadequate glenoid component seating and fixation.”

By making explicit the judgments that surgeons already form implicitly, this framework brings rigor and transparency to causal reasoning.

---

Step 2: Progressively Building a Case Library

By indexing and pooling these structured case analyses across many revisions, we can identify patterns that no single surgeon’s memory or experience could reveal.

Aggregated data will show:

• Causes consistently endorsed (e.g., “flawed fixation and seating = very likely”)

• Causes inconsistently endorsed

• Causes rarely implicated

This approach transforms anecdote into evidence.

---

Step 3: Sharing the Knowledge

As the indexed library grows, it will provide a searchable knowledge base—providing case-based evidence to inform surgeon decisions for future patients with the diagnosis in question.

The larger the library, the more refined the collective insights become. By presenting and publishing results, surgeons’ attention will increasingly focus on actions that demonstrably reduce the risk of revision.

We need to accept the mission.

Just because it's difficult is not a reason for not doing it.

Common Yellowthroat

Union Bay Natural Area

Spring 2022

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/

Here are some videos that are of shoulder interest
Shoulder arthritis - what you need to know (see this link).
How to x-ray the shoulder (see this link).
The ream and run procedure (see this link).
The total shoulder arthroplasty (see this link).
The cuff tear arthropathy arthroplasty (see this link).
The reverse total shoulder arthroplasty (see this link).
The smooth and move procedure for irreparable rotator cuff tears (see this link)
Shoulder rehabilitation exercises (see this link).

---

Updated: As many as 25,000 acromial / scapular spine fractures may occur per year after reverse total shoulders. What are we doing about it??

Two recent articles Acromial stress fractures and reactions after reverse total shoulder arthroplasty: a case-control study and Risk Factors for Acromial and Scapular Fractures Following Reverse Shoulder Arthroplasty: A Meta-analysis of Over 100,000 Shoulders  confirm that these fractures are among the most common, most serious and most difficult to treat complications of reverse total shoulder arthroplasty. We see that the rate of acromial / spine fractures is not decreasing and that with the increasing use of reverse total shoulder arthroplasty the number of patients experiencing these fractures each year will continue to rise rapidly, perhaps to as many as 25,000 per year globally.

and 7,000 per year in the US.

Although the factors associated with these fractures are well known: osteoporosis, inflammatory arthritis, female sex, older age and lower BMI, corticosteroid use, rotator cuff deficiency, prior shoulder surgery (especially cuff repair), none of these is modifiable by the surgeon.

While some surgeon-controlled risk factors have been identified (screw placement and coracoacromial ligament preservation), other possible factors (humeral and glenoid component distalization and lateralization, acromio-tuberosity contact in abduction, change in acromio-humeral distance, and the timelineness and vigor of post operative rehabilitation) have not been consistently associated with fracture risk.

In that most rTSAs have successful outcomes, by what means can we learn how to reduce the rising number of acromial-spine fractures?

The Shoulder Arthroplasty Failture Research initiative seeks to learn safety lessons - not by statistical analysis of large case series or registry data  - but rather by considering in each individual fracture case what might have been done differently to avoid the patient experiencing the complication = causal modeling.

NASA has had 178 crewed space flights and two fatal flights. The two failures taught valuable safety lessons that could not have been learned from statistical analysis of the 178 "cases".  

In each of two individual fatal accidents, NASA had to model many possible causes of the tragedy. The results of causal modeling are shown below.

Challenger (STS-51L, 1986) Cause: Failure of an O-ring seal in the right solid rocket booster. All seven astronauts died. Cold weather on launch day made the rubber O-rings brittle. Engineers had raised concerns about launching in freezing weather, but management overrode them under schedule pressure. Richard Feynman placed a piece of the O-ring material into a glass of ice water and showed that the rubber lost its elasticity at low temperatures, failing to spring back quickly. Counterfactual: had a cold-tested O-ring been used, the lives of the seven astronauts may have been spared.

---

Columbia (STS-107, 2003) Cause: A piece of foam insulation from the external tank broke off during launch, striking the left wing's leading edge, damaging its reinforced carbon-carbon panels. The crew module was destroyed on re-entry. All seven astronauts died. NASA had a history of foam shedding from the external tank before Columbia, but it was consistently downplayed. Counterfactual: had NASA addressed the prior foam insulation failures, the lives of the seven astronauts may have been spared.

---

Consider these two cases of acromial/spine fractures after reverse total shoulder arthroplasty

In the case on the right, most of us would suggest the counterfactual that "if the screw had not been placed in the scapular spine, the fracture would likely have been avoided".

However, in the case on the left, is it likely that the fracture would have been avoided if the surgeon had achieved a lower position of the baseplate, more inferior tilt of the baseplate, more or less humeral distalization, more or less humeral lateralization, by assuring lack of tuberosity-acromion contact, or...? Expert surgeons may have different opinions, but we will only learn by pushing ourselves to answer the causation question, "would the outcome have been different if..?"

So.... 

A. for each acromion/spine fracture case, we should consider

(1) a pre-defined set of causal variables that could have been changed by the surgeon

Screw position: trajectories, lengths, whether any screw is outside-in; distance from superior screw tip to scapular spine; posterior screw proximity to suprascapular notch

Scapular ring status: coracoacromial ligament  (CAL) intact vs transected; any deltotrapezial fascia compromise. 

Construct geometry: humeral distalization, humeral lateralization, neck-shaft angle, humeral inlay/onlay, glenosphere lateralization, glenoid  baseplate tilt, inferior  glenosphereoverhang.  Pre to post op change in acromiohumeral distance (ΔAHD)

Rehab intensity & timing: early deltoid loading milestones.

(2) factors that were non-modifiable for that operation, but critical for counterfactual simulation

Patient bone health: DEXA/T-score proxy; steroid use; rheumatoid/inflammatory arthritis. 

Rotator cuff status: Tear, cuff tear arthropathy

B. Pose explicit counterfactuals
Example queries for each fracture case, for example in a specific case ask: 

• “If the superior screw had been omitted or shortened (inferior-only fixation), would fracture probability have dropped?” 

• “If the CAL had been preserved, would modeled spine strain have stayed below fatigue thresholds?” 

• “If humeral distalization (ΔAHD) had been 3–4 mm less, would the risk have decreased?” 

Pair each patient with 2–4 closest non-fracture rTSA controls (same age/sex/diagnosis/cuff status, bone quality) and run a small within-case causal analysis (not just regression): what single change (if any) most reduces predicted risk for this patient?

C. Convert findings into micro-rules (“guardrails”).
Examples that fall straight out of current evidence:

• Avoid outside-in or long superior screws when fixation allows; favor inferior-biased screw strategy. 

• Preserve the CAL unless there’s a compelling reason to release it. 

• Limit humeral distalization; scrutinize ΔAHD and inferior overhang. 

• Create a “spine-at-risk” checklist for osteoporotic, inflammatory arthritis, steroid-using, cuff-deficient, very low-BMI patients—flagging surgeon controlled variables that appear most influential on fracture risk in this group of patients..

D. As the library of cases grows: every 10–20 fracture cases, publish short, anonymized notes summarizing the modifiable factors that appearn to have the greatest preventable impact on fracture risk? 

---

About our two example cases

• Right-hand image (with a screw traversing the spine): the counterfactual (“no superior screw / shorter screw / inferior-only fixation”) is very plausible given where many fractures localize. 

• Left-hand image (debate over baseplate height/tilt, lateralization, distalization, tuberosity–acromion contact): literature does not consistently link most of these to fractures, with the notable exception of excess distalization. Capture ΔAHD, CAL status, and screw map; then test those counterfactuals first.

This type of causal modeling is not familiar to most surgeon-scientists, but we should begin learning to put it to use for the good of our future patients.

Our colleague Jon Levy responded to this post:

"Another outstanding post.
Over the past 15 years I have integrated all of these factors into my workflow and planning efforts.
Of the modifiable actions, I strongly believe the greatest reduction in acromion fractures occurred with:
1) inlay humerus — since I philosophically believe in glenoid-based laterization to maximize motion and avoid notching induced osteolysis, implanting the humerus within the metaphysis (rather than above it) had a dramatic reduction in fracture rate
2) preop planning — focus on 3 goals — (a) maximize glenoid fixation; (b) achieve at least 70 degrees of abduction motion before greater to acromion impingement; (c) avoid final position of the humerus being more lateral than preop position.
Preop planning had the greatest impact in year to year variation in my fracture rate."

I asked Jon how he managed the rTSA when the planned reconstruction was too loose. His response was "If after reducing the shoulder, the tension is not right and soft tissue balancing is too loose, I typically will select a +4 semi constrained trial. 

 

With the system I use, constraint is added by going deeper into the poly socket. This essentially equates to a +2.5 but gains additional constraint.

 

I almost never go to a +8"  

Looking for answers

Cooper's Hawk

Matsen Backyard

2021

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/

Here are some videos that are of shoulder interest
Shoulder arthritis - what you need to know (see this link).
How to x-ray the shoulder (see this link).
The ream and run procedure (see this link).
The total shoulder arthroplasty (see this link).
The cuff tear arthropathy arthroplasty (see this link).
The reverse total shoulder arthroplasty (see this link).
The smooth and move procedure for irreparable rotator cuff tears (see this link)
Shoulder rehabilitation exercises (see this link).

Surgical failures: what causes them and how can we do better for our patients. Warning: this post is lengthly but informative!

The Book of Why   is transforming our understanding of the causation of surgical outcomes and how to optimize them for our patients. The book is both terrific and dense. Here I try to provide some "Cliffsnotes" that relate to our surgical practices. 

Chapter 1: The Ladder of Causation

Level 1: Association

Most of the publications relating to surgical outcomes are observational, reporting associations of various factors with the result ("12% of patients having an anatomic total shoulder had a surgical revision, 36% had rotator cuff failure, 60% had cementless glenoid components, older patients have a lower hazard ratio for revision  ....").  Comparative tools - such as p values, hazard ratios, and Kaplan-Meier curves - are commonly used in these reports to show associations, but not which factors determine the outcome. While these tools can identify the factors associated with surgical failure, they don't tell us how we can avoid the causes of adverse outcomes for our future patients. 

Level 2. Intervention. 

When we do surgery for an individual patient, we have to select a procedure from among the alternatives (for an irreparable cuff tear, we select debridement, partial repair, subacromial balloon, superior capsular reconstruction, bioactive graft, tendon transfer, or reverse total shoulder). If I chose one rather than the others, how would the outcome have been different for the patient?  Even the best attempt at a randomized controlled trial of these seven different procedures would not be able to guide our patient management. Is it realistic for our surgical actions to be objectively "data driven" or does our choice need to be informed by a subjective analysis as in Level 3?

Level 3. Counterfactual thinking

Learning from failure. Fortunately most orthopaedic surgeries turn out well for the patient; thus the greatest opportunity for learning comes from our failures. When a patient experiences a complication, we need to ask retrospectively, "if I had chosen a different procedure or different implant, might the outcome have been better for the patient". This question cannot be answered objectively, but asking it forces the surgeon to try to assess the root cause of each failure and to ask the subjective question, "in hindsight, how it might have been addressed". Progress will be accelerated when we treat every failure as a causal case study, not just a statistic. "For a specific patient with a loose anatomic glenoid component, is it likely that revision could have been avoided if we had used an augmented glenoid?"

Chapter 2: The Genesis of Causal Inference

Every time we adopt a new implant or technique our patients become unwitting participants in uncontrolled experiments. Retrospective case series can only provide observations (e.g. patients with osteoarthritis treated by reverse total shoulders had a complication rate of 20%) that are confounded by inter-patient variability and inter-surgeon variability, leaving us with no information on causation or prevention. 

What about RCTs?

1. Prospective randomized controlled trials (RCTs): the good and the bad.

The good: Only by deliberately assigning treatments randomly to patients from a defined population can we separate correlation from causation.

The bad: surgical RCTs are rare, often underpowered and limited by 

(a) those patients with diagnosis X who consent to a randomized study of treatment by procedure A or B may not represent the typical patient with diagnosis X ("I don't want my treatment to be decided by the flip of a coin, I want to decide my treartment in partnership with my surgeon")

(b) eithics: (is there really equipoise?)

(c) surgeon skill variation (surgeon A is really good at fixation of fractures while surgeon B is really good at endoprosthesis treatment of fractures)

(d) challenge of obtaining long term followup on a sufficiently large number of patients.

(e) only answering "does procedure A work better than procedure B on average", but not "for which patient, with which anatomy, and in which surgeon's hands?"

Chapter 3: From Evidence to Causes

and 

Chapter 4 Confounding and Deconfounding

The problem of confounding: 

A third variable can create a spurious associate between two other variables. For example, older, lower demand patients may preferentially be offered a reverse total shoulder. If they have lower revision rates than the patients receiving anatomic total shoulders, it may look like reverse total shoulder arthroplasty causes fewer failures, when age may be the primary factor influencing the revision rate. Similarly, surgeon case volume may influence both implant choice and complication rate (obscuring the relationship between implant and complication). Cuff quality may influence both implant choice and functional outcome. 

We need to ask, "what factors could plausibly influence both the procedure and the outcome?" These confounders need to be controlled for either by randomization or stratification.

Such confounders may be difficult to identify; here are some possible approaches

   1. From the surgeon's intuition, experience and domain knowledge, does the varible: 

        (a) influence treatment choice?

        (b) independently influence outcome?

        (Note that bone quality influences both).

   2.  Are there empirical clues such as imbalances beween treatment groups (are patients receiving one procedure older, sicker, or lower demand)?

        (Note that patient age influences procedure and outcome).

Examples of confounders

    (a) Age: older patients more likely to get a reverse AND are more likely to have lower revision rate

    (b) Surgeon volume: high volume surgeons prefer certain implants AND have better outcomes

    (c) Cuff integrity: determines both implant choice AND prognosis.

    (d) Youth and male sex: determines procedure choice (e.g. ream and run vs total shoulder) AND outcome

    (e) Healthier patients with less deformity: determines implant choice (e.g. stemless humeral component) AND revision rate

Each of these must be considered in analysis to avoid misleading conclusions

     

Some important confounders are rarely measured:

    (a) frailty

    (b) social determinants of health

    (c) patient motivation.

   When planning a study, we need to explicity list potential confounders and decide how to measure and account for them. Once identified, confounders can be handled by randomization, restriction, matching, regression, stratification, and/or propensity methods. However, each of these methods carry their own risks, for example propensity matching risks loss of the cases that cannot be matched (which reduces the generalizability of the result). The same can be said for randomized controlled trails. In any event the conclusion of the study needs to acknowledge the identified confounders, the potential for other confounders and how the authors endeavored to mange confounding.

Chapter 5 Colliders

Not every variable should be controlled for. Ask "is this variable a cause or and effect?"

A confounder (e.g. surgical volume) is a cause of both the exposure (implant choice) and the outcome (revision). We should control for this variable in deciding the relationship between implant choice and revision.

A collider (e.g. revision) is an effect of unrelated factors (patient comorbidities) and (poor surgeon skill). If we control for revision, it will create a spurious connection between patient comorbidities and poor surgeon skill.

Chapter 6 Causal Paradoxes

Suppose revision rates appear higher for anatomic TSA compared with reverse TSA when looking at raw totals. But when stratified by age group, we find: (1) younger patients (who are more likely to get anatomic TSA) have higher revision rates overall but (2) within each age group, anatomic TSA actually perform  better than reverse TSA. The paradox occurs because age (a confounder) wasn’t adjusted for in the aggregate data. This is an example of Simpson's paradox.

Simpson's paradox is a statistical phenomenon where a consistent trend appears in different groups (below right), but disappears or reverses when the groups are combined (below left) for the same data. This occurs because a hidden confounding variable (in this case patient age) distorts the relationship between the main variables being studied.

We need to be aware of paradoxes when considering:

Registry data: Revision risk comparisons between aTSA and rTSA can show Simpson’s paradox if patient factors (e.g., rotator cuff status, age, bone quality) are not stratified.

Center outcomes: High-volume centers may appear to have “worse” outcomes overall because they take on more complex patients. Within complexity strata, they may actually have better results.

Implant comparisons: Stemless vs stemmed TSA may look different in revision risk until stratified by deformity or bone quality.

Chapter 7 Intervention

Why Traditional Statistics Fall Short.

When we ask: "If we do a reverse total shoulder instead of an anomic in this patient what is the likelihood of revision?" we are asking a causal question, not simply observing what factors have been reported for revision.

Observation (association): "In the registry, patients with a reverse TSA had more revisions"

Causal question (intervention): "If I do a reverse total shoulder rather than an anatomic on this 72-year old man with poor cuff integrity, how would that change the probability of a revision?"

Traditional statistical tools such as regression and stratification can balance measured variables (age, cuff status, glenoid type) but they fall short when two fundamental challenges are present:

(1) Complex Causal Structures

Surgical decisions involve networks of influences where confounders and colliders play critical roles

Example of confounder: Bone quality. Bone quality influences both implant choice and revision. Implant choice also influences revision. If we fail to account for bone quality, the apparent relationship between implant type and revision is biased.

    Example of a collider: Revision. Comorbidities and surgical technique are not directly related, but both influence revision. If we analyze only patients having a revision, we are likely to find a spurious correlation between comorbidities and surgical technique. 

    Note that adjusting for a confounder is essential, adjusting for a collider is misleading. Traditional regression models often cannot distinguish between the two.

(2) Unmeasured confounders. 

Even the best registries do not document important factors that influence surigical decisions and outcomes. Examples are:

    (a) Surgeon philosophy (preference for reverse or anatomic)

    (b) Patient motivation (adherance to rehabilitation, pain tolerance)

    (c) Subtle anatomic features (bone stock, tendon quality, preoperative stiffness)

    (d) Social determinants of health

    (e) System accessibility 

Because these variables are unmeasured, statistical adjustment cannot account for their influence, leaving residual bias in comparisons.

To translate evidence into surgical choices a surgeon needs to 

(1) Identify all confounders that could affect the outcome (patient related, surgeon related, system-related)

(2) Estimate the weight (influence) of each confounder based on surgical experience and the literature - which are the most important?

(3) Determine which of the important confounders are known for the case in question. 

(4) Integrate these insights to answer the question, "If do a reverse total shoulder rather than an anatomic on this 72-year old man with poor cuff integrity, how does that change the probability of a revision?"

Chapter 8 Counterfactuals

 

Association (observation) : "Patients with a reverse TSA had more revisions"

Intervention (choosing between options): "If we do a reverse TSA instead of an anatomic in this 72-year old man with poor cuff integrity, how does that change the risk of revision?"

Counterfactual ("what if"): "For this patient who had a reverse TSA and required revision, what would have happened if I had done an anatomic instead?"

Counterfactuals cannot be answered with traditional statistics alone. We need (1) a causal model describing how patient, surgeon, implant and systems interact (2) evidence (from registries, trials, cases studies) to estimate probabilities of different paths, (3) to "re-run" history under different choices.

Example: our 72-year old man with poor cuff integrity had a reverse TSA and required a revision. If this patient had instead received an anatomic TSA, would revision have been avoided? Answering this question requires a causal model that considers

 both measured (age, sex, glenoid type, bone quality, cuff status, comorbidities) and unmeasured confounders (motivation, pain threshold, compliance, change, social determinants of health). 

Counterfactuals are central to learning from failure: given what we could not control (characteristics of the patient, shoulder, and environment) what could we have been done differently that may have reduced the risk of failure (surgeon-controllable variables), including

glenoid implant: choice, placement, sizing, bone preparation, fixation

humeral implant: choice, placement, sizing, bone preparation, fixation

glenohumeral relationships:  lateralization, distalization, centering, compression

soft tissue managment: cuff, subscapularis, capsule 

infection prophylaxis:

rehabilitation

Such an analysis can be informed by classifying the type(s) of failure leading to revision: 

dislocation

acromial fracture

glenoid fracture

humeral component breakage

glenoid component breakage

polyethylene wear

subscapularis failure

rotator cuff failure

stiffness

infection

Each of these failure modes drives a different set of counterfactuals: would a differnent implant, surgical technique, or postoperative treament have avoided a particular type of failure?

If you've read this far, congratulations. I've tried to take concepts that are very important and make them accessible. 

Comments welcome!

Finding the meat

Western Tanager

Seattle

2022

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/

Here are some videos that are of shoulder interest
Shoulder arthritis - what you need to know (see this link).
How to x-ray the shoulder (see this link).
The ream and run procedure (see this link).
The total shoulder arthroplasty (see this link).
The cuff tear arthropathy arthroplasty (see this link).
The reverse total shoulder arthroplasty (see this link).
The smooth and move procedure for irreparable rotator cuff tears (see this link)
Shoulder rehabilitation exercises (see this link).