UW Shoulder and Elbow Academy

Saturday, April 26, 2025

Objective ignorance - a problem in predicting outcomes in climbing and in orthopaedic surgery

Mount Rainier is a popular climbing destination. Approximately 10,000 registered climbers attempt to reach its 14,000 foot summit each year. From the graph below we can see that the base rate for fatalities among these climbers averages about 3 per ten thousand (0.3 per thousand). However, the per year rate varies substantially. We are unable to predict the death rate for the coming year because we cannot anticipate what factors will contribute to the death rate in 2026. Daniel Kaheman refers to this as "objective ignorance".

A striking example of our inability to predict the occurrence of climbing deaths was in 1981when an avalanche killed 11 Rainier climbers. This event stands as the deadliest mountaineering accident in U.S. history. On that day, a group of 29 climbers, including guides and clients from Rainier Mountaineering Inc., were ascending the Ingraham Glacier route. At approximately 5:45 a.m., a massive serac—an unstable block of glacial ice—broke loose from the upper Ingraham Glacier, triggering an avalanche of ice and snow. The avalanche swept through the climbers' resting area near Disappointment Cleaver, burying 11 individuals under tons of ice and snow. Despite extensive search efforts, the victims' bodies were never recovered and remain entombed within the glacier.

As an aside, my wife and I (and many others) safely summited Rainier by the Ingraham Glacier route both before and after that accident.

Climbers refer to avalanches as objective hazards - dangers that exist independently of a person's actions, skill, or decisions - they are external, uncontrollable risks inherent to the environment. In this case the National Park Service Board determined "that the accident was a random event that could not have been predicted."

In orthopaedic surgery, objective ignorance keeps us from accurately predicting outcomes for our patients. As pointed out in a prior post, two year ASES scores for patients with cuff intact arthritis having reverse total shoulder arthroplasty have a base rate averaging 83 [SD 12.6]. Yet these numbers do not enable surgeons to accurately predict the result in an individual case because we are ignorant of the "random events" that could have profound effects on the outcome realized by the patient. An unexpected optimization of their social or rehabilitation support may lead an exceptionally good recovery. Conversely, the outcome is likely to be subpar after an acromial/spine stress fracture or the onset of Parkinson's Disease leading to multiple falls and dislocations. In addition, the new implants a surgeon has started to use may be found to have consistently better outcomes, or they may have a design flaw that only becomes evident months after the procedure.

The point here is that while we may know the average base rates for the outcomes of surgical procedures performed in the past, we need to be cautious about using this information to make predictions about future results for an individual patient because we are ignorant of the objective factors that may affect that person's outcome.

You can support cutting edge shoulder research that is leading to better care for patients with shoulder problems, click on this link

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).

Thursday, April 24, 2025

How to select a research project - my thoughts

Many of us want to do shoulder research - it is engaging, holds promise for improving our understanding and care of our patients, and helps in career advancement.

Because good research requires substantial effort and funding, it is important to make good choices.

In selecting a project with high probability of success, three elements need consideration:

1. Can the question be stated clearly in quantitative terms?

2. Why is this question important?

3. Is the study doable?

Let's consider an example regarding the clinical value of 3D planning for reverse total shoulder arthroplasty for patients with primary osteoarthritis

1. Can the question be stated clearly in quantitative terms?

In a trial controlling for important confounders, does the use of preoperative CT-based planning for patients having primary reverse total shoulder arthroplasty for arthritis improve the two-year postoperative American Shoulder and Elbow Surgeons (ASES) scores by the clinically significant amount of 20.9 in comparison to comparable patients having reverse total shoulder arthroplasty without 3D CT planning.

2. Why is this question important?

Because obtaining advanced imaging and using planning software involves increased time, expense and additional radiation exposure, surgeons and patients should be able to consider the benefit of these technologies to the patient in comparison to their costs.

3. Is the study doable?

The recently published two year ASES scores for patients with cuff intact arthritis having reverse total shoulder arthroplasty averaged 83 [SD 12.6]. In that the maximum ASES score is 100, there are only 17 points of possible improvement before the ceiling is encountered. Since the minimal clinically important difference is 20.9, it would not be possible to show a clinically significant benefit to the patient of 3D CT based planning using the ASES score as the outcome of interest.

Thus while the question can be stated in quantitative terms and is clinically important, the proposed study is unlikely to provide the desired information.

Students, residents, fellows and faculty come up with ideas for research projects. We've found that tasking them with answering these three questions helps them make good choices for a productive research experience.

House finch making good choices

Sunday, April 20, 2025

Severe B2 glenoid in an active 51 year old man

A 51 year old general contractor, competitive bow hunter and swimmer presented with persistent and limiting left shoulder pain and grinding that has been refractory to arthroscopic "debridement" and intraarticular steroid injections. On examination he had limited motion with 80 degrees of glenohumeral flexion, 0 degrees of external rotation, and internal rotation to the gluteal area. His x-rays at presentation show osteoarthritis with inferior and posterior decentering and severe B2 pathoanatomy.

After discusscion of the alternatives of non-operative management, an anatomic total shoulder and reverse total shoulder, he elected to proceed with a ream and run arthroplasty to avoid the risks and limitations of a plastic glenoid component.

CT scanning and the use of planning software were avoided. The procedure was performed under general anesthesia without a nerve block. A subscapularis peel was performed, preserving the long head tendon of the biceps. A standard free-hand anatomic neck cut was made in 30 degrees of retroversion and at a 135 degree angle with the humeral shaft.

His humeral head showed the "Friar Tuck" pattern of central cartilage loss.

Conservative glenoid reaming was performed without attempt to modify glenoid version. Patient-specific instrumentation and augmented reality were not used. Intraoperative trialing indicated that a 54 20 anteriorly eccentric humeral head provided the optimal balance of mobility and stability. Neither a plastic glenoid component or bone cement was used. The postoperative x-rays are shown below.

Five months after surgery he reported " Shoulder doing good saw great improvements this month. Strength is improving and have resumed construction work with some limitations. Haven’t started shooting a bow yet but soon I think. Thanks for a great shoulder" and provided the video shown below of his active motion.

House finch in cherry blossoms

Saturday, April 19, 2025

Surgical thinking: fast and slow - amended per Dr Collin

As a departure for the "usual and customary" blog post, I'm sticking my neck out to call reader's attention to a book suggested to me by Philippe Collin. You may or may not be aware of it.

Thinking, Fast and Slow, written by Nobel Laureate Daniel Kahneman, can help shoulder surgeons gain important insight into how they make decisions. It explores the interaction between fast, intuitive, instinctive, emotional thinking (“System 1”) and slow, deliberate, logical thinking (“System 2”).

System 1 operates almost automatically, takes mental shortcuts (heuristics) and is prone to biases and errors. It overestimates the surgeon's knowledge and predictive ability, ignoring data/statistics in favor of recent experience, stereotypes and incomplete information (limiting consideration to "what you see is all there is - WYSIATI", failing to consider other relevant information). It is driven by an optimism bias, underestimating time, costs, and risks. When faced with a difficult question (e.g. “what are the odds of a good treatment outcome for this patient in my hands?”), System 1 substitutes an easier question (“what recent impressive outcomes come to my mind?” - known as the "availability heuristic"), overestimating positive results and underestimating complication rates. It is subject to "anchoring", e.g. a surgeon may focus on a patient's initial MRI and discount new data (e.g., improvement with physical therapy).

System 2 strives to avoid these errors, but requires substantial effort, deliberation, concentration, logic and data; it should be brought into play for complex impactful decisions and when the lazy System 1 is at risk for failure. It asks questions like “For the different treatment options, what are the published base rates for good and adverse outcomes; improvement in comfort and function; and time to recovery for similar patients with similar diagnoses?” and "How do we reconcile the data indicating that (a) less than 5% of patients with rotator cuff tears have surgery with (b) rotator cuff surgery is the commonest of shoulder surgeries?" It also asks questions like “what accounts for the variability among surgeons (e.g.surgeon training, practice environment, economics, age)”? as well as “what factors may bias decisions among treatment options (e.g. a surgeon with a financial interest in an outpatient surgical center may be biased toward patients and treatments that can be accommodated there)? System 2 should be informed by a surgeon’s personal documented outcomes and complications for similar patients with similar diagnoses that she or he has treated. System 2 should consider contradictory evidence against a considered surgery before proceeding: what are the "cons". System 2 invites a colleague's opinion when the decision-making is difficult. System 2 requires concentration, which may be compromised by distracting conversation or music in the operating room.

Finally, the book points out that people, including surgeons and patients, are only human. The decisions individuals make are influenced by "priming" (what happened to them just before making the decision; are they happy, rested, hungry, just having a bad day?) and by "framing", i.e. how is information presented: if you have $50 would you rather keep $20 or lose $30? (most folks would choose to “keep $20” even though the two choices are equivalent!). Similarly, is a procedure with a 7% revision rate better or worse than a procedure with a 93% revision free record?

With apologies for the amateur nature of this "book report" (my first since I graduated from the University of Texas in 1964), I hope I've stimulated your interest in Kahneman and his thoughtful consideration of how we make decisions.

Philippe COLLIN • 1st

Orthopedic Shoulder Surgeon at American Hospital of Paris & Elect President of European Society for Surgery of the Shoulder and the Elbow

Fascinating to see how Daniel Kahneman’s work on the two systems of thinking resonates with daily practice in shoulder surgery.
Two examples come to mind:
🔹 For years, we linked acromion shape with rotator cuff tears. The idea was simple, logical, and compelling… yet clinical reality and recent data suggest the causal relationship is much less clear than we once believed.
🔹 Another example: the early success of the so-called “Grammont design” reverse shoulder prosthesis was largely attributed to medialization and distalization. However, it now seems likely that the real key was the robust glenoid fixation—with a central peg and four screws.
These examples are powerful reminders of the importance of stepping back and being cautious with theories that feel too coherent. A good story is not a substitute for solid evidence.
ShoulderSurgery Kahneman CognitiveBias Orthopedics ClinicalThinking

Red-shafted flicker

Friday, April 18, 2025

Surgical decision making in shoulder arthroplasty

The surgeon makes critical decisions for each patient, striving for the optimal shoulder comfort and function. These decisions include those involving prosthesis selection and positioning.

As pointed out in preoperative planning for anatomic total shoulder and ream and run arthroplasty, aspects of planning can be accomplished using standardized plain films and the available PACs tools. However, image-based preoperative planning (whether based on plain films or on 3D CT scans) does not predict important characteristics of the shoulder that are only evident intraoperatively after surgical exposure and osteophyte resection.

Let's consider two examples from this week's surgical schedule.

Case 1- a 66 year old active man with a successful ream and run for right shoulder. arthritis who presented for a left sided ream and run because of refractory functional limitations. The shoulder was not especially stiff preoperatively (140 degrees of active elevation).

Preoperative x-rays showed osteoarthritis with mild-moderate posterior decentering seen on the axillary truth view.

Preoperative image-based planning showed the ideal head cut and suggested a concentric humeral head implant with 56 mm diameter and a thickness of 24 mm.

At surgery, however, after glenoid reaming the suggested head size overstuffed his soft tissues such that range of motion was excessively limited. Trialing with a 50 mm diameter 20 mm thick concentric head component provided the desired mobility, but this implant was posteriorly unstable when the arm was lifted into flexion. An anteriorly eccentric 50 20 humeral trial provided both the desired mobility and stability. His postoperative x-rays are shown below.

The learning point here is that preoperative imaging could not have predicted the correct implant - that decision had to rest on intraoperative trialing.

The patient kindly gave us permission to show this video of his assisted motion three days after his ream and run.

Case 2: A 76 year old man, 5 ft 3 in with cuff tear arthropathy, pseudoparalysis and osteoporosis - all placing him at increased risk for instability and acromial/spine stress fractures. His preoperative Grashey view is shown below.

Because of his pseudo paralysis, he was not a candidate for a CTA hemiarthroplasty (see Cuff tear arthropathy-current considerations); he elected to proceed with a reverse total shoulder. The surgical goals were to optimize stability without excessive lateralization or distalization and with glenosphere inferior tilt to match the orientation of the line (yellow) representing the base of the suprascapularis fossa.

Achieving this goal is a challenge: (1) the orientation of this reference line is not discernible at surgery and (3) the inferior prominence of glenoid would confound the use of a guide to orient the drill for the central screw.

In this case a surgically visible landmark was the plane of the central glenoid (black line). A line (red) perpendicular to this plane provides a reference of zero degrees of inferior tilt.

The angle between the red and yellow lines indicates the desired orientation of the drill to achieve the desired inferior tilt of the baseplate.

The superior-inferior position of the baseplate was selected so that after reaming the inferior glenoid would be covered by the glenosphere.

The final reconstruction is shown below using a 36 mm glenosphere (selected intraoperatively because the 32 mm glenosphere did not provide sufficient stability to vigorous testing), a semiconstrained humeral liner both for stability and to minimize humeral lateralization, and a long smooth impaction grafted stem to avoid varus position of the humeral component (see Dislocation of the reverse total shoulder).

In these two cases the combination of preoperative image-based planning and intraoperative trialing led to the final choice of implant components and positioning.

Recognizing that there is more than one way to skin a cat(bird).

it will be interesting to know the readers' thoughts on alternate approaches.

Thursday, April 17, 2025

A somewhat misleading article "comparing" ream and run to anatomic total shoulder.

The authors of Ream-and-run technique offers equivalent clinical outcomes as anatomical total shoulder arthroplasty but with a high rate of complications: A systematic review and meta-analysis reviewed some of the literature on the outcomes of these two types of anatomic shoulder arthroplasty. The authors reviewed included 738 patients having ream and run (R&R) and 810 having anatomic total shoulder (aTSA). These groups were not comparable with respect to important predictive factors; for example: 94% of the RnR patients were male, while only 56% of the aTSA patients were male; 61% of the RnR patients had type B or C pathoanatomy in comparison to 47% of the aTSA patients; the mean ± standard deviations for the length of followup were not provided. There was no attempt to control for these influential variables in the statistical analysis.

The authors reported that they found no clinically significant differences in patient reported outcome measures or in range or motion between the two procedures.

The revision-free survivorship was 93% for the ream and run patients and 97% for the aTSA patients.

In the RnR group 3.9 % had chronic pain and stiffness, 2.6 % had unspecified "humeral head problems" and 2% had culture-positive infection.

In the aTSA group 1.2 % had soft tissue failure, 1.2 % and chronic pain and stiffness; 0.86 % had loosening of the glenoid prosthesis.

Some articles in the Discussion section were incorrectly reported, such as "However, Schiffman et al. reported post-operative SST statistical improvement only within the aTSA group post-operatively (aTSA p = 0.01, RnR p = 0.65). " This is a misstatement; this paper referred to the Impact of previous non-arthroplasty surgery on clinical outcomes after primary anatomic shoulder arthroplasty and reported "Previous surgery is associated with inferior clinical outcomes and higher revision rates in patients undergoing index TSA but not in those undergoing the ream-and-run procedure."

Comment: The presentation of RnR and aTSA outcomes in this review indicates that each of the two procedures can yield excellent clinical outcomes for appropriately selected patients. The aTSA remains the preferred procedure for most patients having osteoarthritis with an intact rotator cuff, especially for women and for those who do not wish to participate in heavy or impact activities.