Data-Driven Surgery for Orthopedics by Prof Med Heiko Graichen

March 20, 2023

World-renowned orthopedic surgeon and recognized pioneer in computer navigation arthroplasty,[i] Heiko Graichen, MD, PhD, shares his experience as a data-driven orthopedic surgeon exploring ways to improve surgical training and patient outcomes for total knee arthroplasty (TKA). Approximately 40 percent of the global population over the age of 55 experiences chronic knee pain. Of those, 50.8 million suffer from disabling pain, and as a result, that leads to 2.6 million who turn to knee replacement surgery each year.[ii]

TKA rates have increased substantially in recent decades worldwide, with Germany being one of the leading countries in the prevalence of TKA,[iii] and as a result, Dr. Graichen performs approximately 750 knee surgeries each year.

In this blog, we share how Caresyntax is working in a clinical collaborative with Dr. Graichen to identify key phases, leverage data, and variability assessment for TKA and hip procedures. We also examine the impact of data-driven surgery on surgical education, patient satisfaction, workflow, and team experience, including hospital leadership. Dr. Graichen also shares a few predictions for data-driven surgery.

Working to develop standard metrics for TKA: goals, technology, implant & workflow

Dr. Graichen has been working with Caresyntax as an orthopedic consultant in a global TKA collaborative created to develop a video-based assessment and education tool for TKA surgery. The goal is to develop TKA-specific standard metrics for video-based assessments and surgical education​. Including all different technologies, workflows, implants, and surgical techniques/goals in the analysis of TKA surgery makes the task complex.

The results of this collaboration will help experienced surgeons to reduce variability and resident surgeons quickly develop essential core competencies.

Identifying and optimizing key phases

Dr. Graichen and other surgeons collect videos from multiple surgical centers and review the videos, observing different parts of the procedures independently in order to develop a rating system for the entire procedure. This includes technical parts such as osteophyte removal and approach but also quantitative decision such as implant placement according to various alignment philosophies.

The data transfer process from the clinic to Caresyntax entails data acquisition from recording TKA cases, using GoPro cameras with SD cards. The recorded cases are transferred to Caresyntax from the hospital via laptop with a secure upload to the server in Berlin. The data goes through a process of de-identification before uploading to the cloud. Surgeon Reviewers then evaluate the workflow based on the developed assessment matrix. Surgeons then share their feedback with their surgical teams and with Caresyntax.

The collaboration is working to determine whether the surgery was technically and decision wise well done, appropriate, or somehow needs improvement. The goal is to uncover the reasons why the surgeon may have failed to perform the procedure properly. The team decided to limit the focus of this project to optimizing the valuable time spent in the OR. After developing the quantitative matrix for intra-OP quality, we can analyze how other parameters (e.g., patient factors) might influence this intra-OP quality.

“To understand what determines the quality of TKA surgery, we have to look more into quantitative data, measurable data, and then big data. There are a lot of parameters … and that is something you can only do with big data.”

One next step will be, for example, integrating patient-related outcome measurement scores (PROMS) pre- or post-operatively.

Meanwhile, various digital systems have entered the OR. The main driver is the CAS/robotic system. It has changed surgery from an experience-based qualitative procedure to a data-based quantitative procedure. Besides this system other systems–such as digital imaging and templating systems–have also become an integral part of OR theatres. Further systems exist helping to make workflows more efficient. In Dr. Graichen’s hospital, the entire procedure of TKA/THA is defined by a digital workflow. This allows breaking the complex surgery into multiple small steps. The entire workflow and each step are defined by the surgical team to optimize efficiency. The defined workflow is then displayed on the monitor step by step including all instruments needed for the next specific step. This allows quantification of each step– as well as a number of exceptions–but it also allows comparison of different workflows with regards to efficiency and quality.

Dr. Graichen is using CAS for 100 percent of his TKA procedure for the last 10+ years. This helped to create a huge database. It helps to analyze what classical surgical techniques such as adjusted mechanical alignment can achieve in knees. Simultaneously, it helped also to understand what such a workflow is also changing anatomically in the following papers: Independent of the preoperative coronal deformity, adjusted mechanical alignment leads in a high percentage to non-anatomical tibial and femoral bone cuts.[iv] and Navigated, gap-balanced, adjusted mechanical alignment achieves alignment and balancing goals in a very high percentage but with partially non-anatomical resections.[v]

Although the precision and technical options of CAS/robotics are well accepted, the problem of adapting to the new technology still exist. The reason for that can be that early users have problems with understanding all parameters and their interconnectivity completely. This leads to stress in the OR and additional time loss. Only by fully integrating the digital workflows in the daily practice the overall time can become shorter.

Leveraging data pre- and post-operatively

The research began with quantifying what happens intraoperatively. While CAS/robotic data is already a quantitative data source this part can be interpreted easily, more complex is the quantitative analysis of video data. In the Caresyntax project, an expert panel will implement a quality scale for several parts of the surgery, such as approach, osteophyte removal, etc. to analyze that. Each of these parameters will be assessed and scored. By summing all points for each step and decision, finally each surgery can be assessed independently. Based on this, in future analysis surgical workflows and alignment techniques can be compared quantitatively with regard to their overall intra-OP quality.

Variability assessment based on video and data

TKA video-based assessment is rated on technical quality and workflow. Technical quality includes the incision, treatment of soft tissue, and how osteophytes were removed. It is important to open the knee properly and not damage the soft tissue; that is very important for the quality and success of the surgery.

To assess the workflow, Dr. Graichen’s team placed a camera focusing on the OR situs. Another camera focuses on the interaction between the scrub nurse and the surgeon to see how effective they are working together. They wanted to determine how effective the nurse was in preparing the desk or table and the different instruments.

Navigation and guidance devices in joint replacement and spine surgeries, as well as arthroscopy cameras for joint arthroscopies generate a wealth of new data. Combined with exoscope and room camera video, imaging data is being collected and used to create a unique High-Fidelity Surgical Record™ that can be used to increase efficiency and reduce variability.

Data-driven surgery to improve education and patient satisfaction

Dr. Graichen is a strong believer in data-driven surgery: one important reason is for education. In his role as a professor at the University in Frankfurt, educating students and residents is important. “I think if you want to educate the next generation on improving the quality of the process, data is very helpful. It makes everything transparent.”

“I think if you want to educate the next generation on improving the quality of the process, data is very helpful. It makes everything transparent.”

The next most important use of data-driven surgery is to improve patient satisfaction. A study of six literature databases from 2005 to 2016 found that approximately 20 percent of patients report dissatisfaction with their TKA.[vi] Dr. Graichen believes “If we want to reduce the rate of unhappy patients after total knee arthroplasty, it cannot be done without data. We have to measure and use as much data as we can get.”

“If we want to reduce the rate of unhappy patients after total knee arthroplasty, it cannot be done without data. We have to measure and use as much data as we can get.”

If you have too much data, it can get confusing, but that is where technology and artificial intelligence might help. Many scientists are now working to help interpret the data that is being collected in a correct way. “My job is to deliver 100 percent quality for each patient,” says Dr. Graichen, “and if there is something that can support me in delivering that, I am happy to include it in my daily workflow.”

Improving team workflow efficiency with data

Dr. Graichen also learned from data-driven surgery in using SPM. As described above, it is a team approach, the team has to agree on one workflow and instrument set and after that the standardization can begin without a long learning curve. Efficiency has improved from the beginning with an overall reduction in OR time of around 20 percent and in a reduction of outliers of 20 percent. As a result, they were able to reduce the average time of surgery by 20 percent for total hip. That amounts to one extra surgery a day.

“We were able to reduce the average time of surgery by 20 percent for total hip. That amounts to one extra surgery a day.”

Improving surgical team experience

This reduction in time is accompanied by reduced stress levels in the OR, because everyone knows what the next step is because it is shown on the monitor. Besides the next step and a short description about it, also the required instruments are shown. This helps the scrub nurse to be prepared for the next step. This improves the nurse-surgeon interaction and that makes it efficient and also reduces stress. Even if an expert surgeon has performed thousands of procedures, the next surgery may be with a completely different team, and this is where standardization is extremely helpful.

Data-driven hospital leadership

Dr. Graichen is part of a large hospital chain of 150 hospitals throughout Germany. The hospital management has a clear interest in improving surgical efficiency by leveraging data. The SPI project originally came from the management side; they started a pilot project in Dr. Graichen’s hospital, and the surgeons here liked it. However, not all surgeons appreciated this step-by-step transparency, because it became easier to learn how long certain steps took for Surgeon A versus Surgeon B.

Predictions for the future of data-driven surgery

Algorithms

“In the near future,” Dr. Graichen predicts, “the individual analysis of X-ray/CT images, of the intraoperative gaps and of the anatomy of a patient, will be detected by something smarter than me.”

Decision support

The machine will offer a suggestion for optimized individual implant placement—such as recommending to cut the tibia in three degrees instead of zero degrees–but it is still my decision. I can overrule the machine if I want to, but if it’s a good machine in 99 percent of the cases, that machine will be right, and it will be faster than me. However, the surgeon will be important in collecting the data correctly, otherwise the best algorithm will not be able to deliver a perfect suggestion. It’s not so far away. I think we already have some algorithms that can help us with that.”

In the next step, all alignment philosophies that are already published will be included in the intra-OP analysis and algorithms will show the differences for each case. Based on a scaling system then the best workflow suggestion will be made. The surgeon will have the opportunity to accept that recommendation and potentially minimize mistakes. “It will raise efficiency dramatically. This will be the future. I think we will be there in the next three to five years.”

 

RESOURCES

[i] Career In Arthroplasty By Prof. Dr. Heiko Graichen YouTube

[ii] https://orthospinenews.com/2019/06/04/demand-for-knee-replacement-grows-5-percent-worldwide/

[iii] Klug, A., Gramlich, Y., Rudert, M. et al. The projected volume of primary and revision total knee arthroplasty will place an immense burden on future health care systems over the next 30 years. Knee Surg Sports Traumatol Arthrosc 29, 3287–3298 (2021). https://link.springer.com/article/10.1007/s00167-020-06154-7#citeas

[iv] Luderer V, Strauch M, Hirschmann MT, Graichen H.Knee Surg Sports Traumatol Arthrosc. 2023 Feb 18. doi: 10.1007/s00167-023-07337-8. Online ahead of print.PMID: 36807723

[v] Graichen H, Luderer V, Strauch M, Hirschmann MT, Scior W.Knee Surg Sports Traumatol Arthrosc. 2023 Mar;31(3):768-776. doi: 10.1007/s00167-022-07014-2. Epub 2022 Jun 9.PMID: 35678853

[vi] Gunaratne R, Pratt DN, Banda J, Fick DP, Khan RJK, Robertson BW. Patient Dissatisfaction Following Total Knee Arthroplasty: A Systematic Review of the Literature. J Arthroplasty. 2017 Dec;32(12):3854-3860. doi: 10.1016/j.arth.2017.07.021. Epub 2017 Jul 21. PMID: 28844632. https://pubmed.ncbi.nlm.nih.gov/28844632/