“The Surgeon is the Method” by Rick Matsen

Dear Colleagues: 

"The Surgeon is the Method" is a post on LinkedIn by Rick Matsen. His Shoulder Blog is a wonderful resource. While I don’t always agree with what he says, his points are very thoughtful. Please see my comments below in areas of his post, as I don’t exactly agree with what he is saying.

Kind Regards,

Jon “JP” Warner MD

Founder, CSS

+++

Posted on Shoulder Arthritis Blogspot:

Saturday, August 23, 2025

Complications after total shoulder arthroplasty – The Surgeon is the Method

The quest for ways to make shoulder arthroplasty safer for future patients continues through Shoulder Arthroplasty Research. Here are some things we know: 

(1) most shoulder arthroplasties turn out well for the patient, thus our greatest opportunities to learn come from studying failures

I agree with this statement and so do many thoughtful surgeons and non-surgeons:

The above image is from From LinkedIn: Visuals for Growth

(2) it is insufficient to focus on the type of failure (e.g. glenoid component loosening, rotator cuff failure); rather we need to hone in on what could have been done differently at the primary arthroplasty to lower the risk of component or cuff failure - this can be thought of as actionable intelligence.

(3) we want to avoid the assumption that technologies such as 3D CT based planning, patient specific instrumentation, robotics, virtual reality, augmented reality will lower failure risk until their effectiveness in vivo has been rigorously demonstrated

There is no avoiding the future where AI provides “guardrails to errors in planning and execution of surgery and Robotics/Navigation will likely reduce errors as well. The problem, currently, is cost. That said, all technology that scales becomes less expensive and more automatic. In his book on making technology simple (The Laws of Simplicity), John Maeda from MIT Business School presents many such examples of how technology which becomes more automatic and simpler reduces cost and scales impact outside of surgery and healthcare. It is already here and growing for healthcare.

(4) our attention falls on the surgeon and the elements of care that are under her/his control

A recent article, The effect of surgeon volume on complications after total shoulder arthroplasty: a nation-wide assessment, provides some actionable intelligence. The authors retrospectively queried the Pearl Diver Mariner database for the years 2010 to 2022. Their analysis included 155,560 patients having primary anatomic total shoulder arthroplasty, excluding those younger than 40 years, those who underwent revision arthroplasty, cases of bilateral arthroplasty, and cases with a history of fracture, infection, or malignancy.

They included cases performed by surgeons with a minimum of 10 cases. 

The 90th percentile for surgeon volume was determined to be 112 cases during the study period. Surgeons above the 90th percentile (n 340) operated on 68,531 patients, whereas surgeons below the 90th percentile (n 3038) operated on 87,029 patients. Surgeons in the high-volume group were significantly more likely to have completed a Shoulder and Elbow fellowship and less likely to have no fellowship training or fellowship training outside of Shoulder and Elbow or Sports Medicine. 

Low-volume surgeons operated on patients with higher baseline comorbidities. Here's my summary of their data.

After adjusting for age, gender, CCI, obesity, and tobacco use, high-volume surgeons experienced lower rates of medical complications including renal failure, anemia, and urinary tract infection. All-cause readmission, reoperation at 90 days, and reoperation at 1 year were significantly lower among high-volume surgeons. Cases performed by high-volume surgeons exhibited lower rates of all complications including prosthetic joint infection and periprosthetic fracture. Here's my summary of their data. NB an odds ratio <1 means that cases operated by high volume surgeons had a lower rate of the complication than low volume surgeons.

Finally, the authors found that the proportion of shoulder arthroplasties performed by high volume surgeons has been decreasing with time.

Comment: This study appropriately puts the focus on the surgeon - the individual that decides which treatment is best suited for each patient, carries out the surgery, and manages the aftercare. In other words, the surgeon controls the modifiable variables for each patient. The surgeon is the method.

The authors characterize the surgeon in three dimensions: (1) case volume, (2) fellowship, and (3) the comorbidities of the patients the surgeon selects to have total shoulder arthroplasty. They then go on to compare complications for surgeons performing ≥ 112 arthroplasties to those performing < 112. 

Thus, the data available are ripe for a multivariable analysis (MVA) characterizing the relationship among these variables - individually or in combination - to the occurrence of medical and surgical complications. Without such an analysis we cannot know the relative importance of each of these dimensions.

Let's look at each of these characteristics:

Surgeon case volume: One of the big questions in orthopaedics is whether more is more, i.e do we continue to get a bit better with each case, or is there a threshold above which we are "good"? In this light it might be more informative to characterize surgeon case volume as the number of cases rather an whether they exceeded a threshold for qualification as "high volume"?  This would get around the problem of having a surgeon performing 111 cases designated as "low volume" whereas if the surgeon had done one more case, he/she would suddenly become "high volume". On reading this paper, a patient might ask "should I travel four hours to have an arthroplasty by a surgeon who has done 120 cases rather than sticking with my local surgeon who has done 110?  Numbers may be better than categories. An MVA should be able to sort this out.

See the Wall Street Journal article, “A New Factor When Choosing a Surgeon”, Sept 19, 2016: “The doctor you want may be the one who specializes in just one kind of procedure, new research suggests.”

Fellowship: The additional year of specialized training afforded by fellowship exposes trainees to a greater case volume and breadth. High volume surgeons were more than twice as likely to have taken a shoulder fellowship; however, fewer than 30% of high-volume surgeons took a shoulder fellowship. As a result, we do not know from the data presented whether taking a shoulder fellowship results in a significantly greater arthroplasty practice volume or whether taking a shoulder fellowship reduces the surgeon's complication rate. An MVA should be able to sort this out. (See my comments below on effect of virtual planning on surgeon decisions)

Comorbidities: The patient population of high-volume surgeons was significantly healthier, i.e., comorbidities as reflected by the Charlson Comorbidity Index were lower in patients operated by higher volume surgeons (perhaps because experience teaches to think carefully before offering elective surgery to patients who are ill or perhaps high-volume surgeons operate in outpatient centers that exclude sick patients). The question is whether a shoulder fellowship or being a high-volume surgeon enables safer surgery on patients with comorbidities. An MVA should be able to sort this out.

Complications: This article presents data on medical and surgical complications in terms of odds ratios but does not present data on the rate of each complication. In an MVA it may be easier to characterize complications in terms of their rates.

Arthroplasty choice: The authors point out that "distinctions between anatomic and reverse shoulder arthroplasty were not made because of limitations associated with CPT coding". This is an important shortcoming of the analysis, because experienced (and perhaps fellowship-trained shoulder surgeons), may be better at deciding which patients are the best candidates for each procedure in terms of avoidance of medical and surgical complications.  

Incremental value of each case: Numbers are not the only important thing. How much the surgeon learns from each case depends in large part on whether the surgeon conducts an After-Action Report (AAR) after each case. An AAR is a structured process used to review the case to identify what happened, what went well, and what could be improved in future cases. We can assume that a 100-case surgeon who routinely conducts AARs will have better outcomes than a120 case surgeon who goes on to the next case without introspection. This is important because most shoulder arthroplasties are not operated on by high volume surgeons.

Most of our analysis is conducted in M & M rounds and patterns of failure are frequently missed. This has been my experience in my role as Vice Chair for Quality and Safety at my institution. We should remember E.A. Codman’s words, when he spoke about the “End-Result Concept” more than 100 years ago, “Every patient should be measured to determine if the surgery they had worked, and if not why not?”

References:

Bernstein, J: “Not the Last Word: Codman was Right – Spread the Word.” Clin Orthop Rel Res. 2015, June 5;473(8):2455-2459

Berwick, D.M.: “Measuring Surgical Outcomes for Improvement. Was Codman Wrong?” JAMA, 2015;313;(5):469-470

We can do a better job of helping our patients avoid problems.

Fresh Grizzly Bear Footprint

Devil's Gap, Alberta

Photo by Laura Matsen, M.D

8/23/2025

Here are a few additional comments about Dr. Matsen’s arguments on Volume and value we provide our Patients:

Learning curves are real! These affect everything we do, and it is important to understand that creation of value really starts with good decision-making and surgical execution. The question is how can we bend the learning curve? Similar to the article referenced by Dr. Matsen, Best and Colleagues (Clin Orthop Relat Res. 2023 Aug 1;481(8):1572-1580.): Higher Volume is Associated with Lower Rate of Subsequent Revision Procedures after Total Shoulder Arthroplasty: A National Analysis.  Here is what they found in their analysis of CMS database for 151,000 patients undergoing TSA and RSA with a minimum of 2 years follow-up:

Results: After controlling for confounding variables including patient age, comorbidity risk score, surgeon and hospital volume, surgeon graduation year, and hospital size and teaching status, they found that an annual surgeon volume of ≥ 10 aTSAs was associated with a 27% decreased odds of revision within 2 years (95% confidence interval 13% to 39%; p < 0.001), while surgeon volume of ≥ 29 aTSAs was associated with a 33% decreased odds of revision within 2 years (95% CI 18% to 45%; p < 0.001) compared with a volume of fewer than four aTSAs per year. Annual surgeon volume of ≥ 29 rTSAs was associated with 26% decreased odds of revision within 2 years (95% CI 9% to 39%; p < 0.001).

Conclusion: Surgeons should consider modalities such as virtual planning software, templating, or enhanced surgeon training to aid lower-volume surgeons who perform aTSA and rTSA. More research is needed to assess the value of these modalities and their relationship with the rates of subsequent revision.

From Best, et al, Clin Orthop Rel Res., Aug, 2023

Weinheimer and Colleagues (Arthroscopy, Vol. 33, No. 7, July 2017: pp 1273-1281) published on “Patient Outcomes as a Function of Shoulder Surgeon Volume: A Systematic Review.” The observed that low volume surgeons (<5 arthroplasty surgeries/year) compared with high volume surgeons had more surgical complications, increased length of hospital stay, longer surgical time, and increased surgical cost.

Warner and Higgins offered an opinion article entitled “Volume and Outcome: 100 years of Perspective on Value from E.A. Codman to Michael E. Porter” (Arthroscopy, Vol. 33, No.7, July, 2017: pp 1282-1285). They suggested that most orthopedic residents have a greater exposure to total hip replacement vs total shoulder/Reverse shoulder replacement, and this is why “low volume” shoulder surgeons who have not had fellowship training vs have a higher complication rate with shoulder arthroplasty than with total hip replacement surgery.

Waiter and Colleagues (A Complication-based Learning Curve From 200 Reverse Shoulder Arthroplasties: Kempton LB, Ankerson E, Wiater JM.  CORR 2011) have shown a learning curve for complications and revision surgery when they started to perform Reverse Replacement. Their work showed a threshold of 40 surgeries for the complication rate to drop from double to single digits.

Registries, such as the Australian Registry, may suffer from blending experienced and inexperienced surgeons and thus weight outcomes to the lowest common denominator when it comes to failure.

Min and Colleagues (Bone Joint J 2020;102-B(3):365–370.) published on “Patient-Specific Planning in Shoulder Arthroplasty: Influencing the Learning Curve for Assessment the Glenoid in Surgical Planning. They observed that when surgeons used a 3-Dimensional planning tool based on CT scans, “The information provided by PSP has the greatest impact on the surgical decision-making of low volume surgeons (those who perform fewer than ten shoulder arthroplasties annually), and PSP brings all surgeons into closer agreement with the recommendations of experts for glenoid classification and surgical planning.

There are many articles on the impact of Robotics and Navigation improving the execution of surgery especially in less experienced surgeons, while experts appear to have no such need for such technology.

It appears to me that Volume is a reasonable Surrogate for Value, especially when one tried to derive revision rates from a registry where surgeon experience is not a variable which is controlled.

The above image is from From LinkedIn: Visuals for Growth