The Scientific Method

Nothing is closer to the core of surgery than the principles of the scientific method by which we discover, evaluate, validate, and implement a new technology. Until the turn of the 20th century, surgery was guided by tradition. It was Nicholas Senn's seminal article in 1908, which pointed out that rather than tradition, a surgeon should rely upon experience [1]. No longer was it acceptable to continue the practices of old simply because it had become the custom; rather, Senn declared that surgeons should look at the experience and results of previous treatments and be guided by logical judgment in surgical practice. From this modest beginning, surgery evolved into the scientific method as we know it today: hypothesis, research, conclusion, and implementation. Laboratory research began to ascend and along with it came clinical trials. Studies were carefully designed and crafted, and then rigorously conducted to gather the evidence necessary to prove or disprove the hypothesis, and culminated in publication of the scientific evidence, which resulted in the acceptance by the surgical community at large. While this method has brought clarity and understanding out of chaos, the rigorous nature of the investigation has resulted in an extremely long time from discovery to validation to implementation. Often new technologies were invoked before the evidence was confirmed, much to the detriment of the patients (laparoscopic cholecystectomy with initial increased incidence of bile duct injuries, or various chemotherapeutic agents with either unintended side effects or lack of efficacy). The converse was also true; prolonged evaluation resulted in many patients not receiving life-saving therapy while awaiting the results of trials, or new surgical procedures not being implemented for decades until the completion of trials (such as laparoscopic colectomy). While it is not the intent to suggest that the current rigorous process is neither valid nor necessary, there is a method that has been implemented by the scientific community that has not been considered by the medical profession, modeling and simulation. There is some early implementation of simulation technologies being explored for rapid rational drug design and for understanding gene-based therapies. Sophisticated computer programs are being used to simulate the effects of literally millions of pos

* The opinions or assertions contained herein are the private views of the authors and are not to be construed as official, or as reflecting the views of the Department of the Army, Department of the Navy, the Advanced Research Projects Agency, or the Department of Defense.

sible compounds, looking for the desired combinations and possible mechanisms of action. These simulations include the composition, structure, folding, bonding, etc., iterated over thousands of potential combinations to discover the most likely candidates for production and study. Thus a nearly infinite number of potential biochemical molecules are reduced to specific drugs or genetic sequences that are targeted and used in clinical trails. There is a primordial effort to take the next step, to test these candidate therapies through simulation on a "virtual cell" (in silico, or computational biology), before implementing them in clinical trials on patients. Following this example to its ultimate conclusion, it is anticipated that it will be possible to simulate an entire organ, or even a single patient or population of individuals, to test and evaluate drug or genetic therapy before implementing on patients. Perhaps all therapies— drugs, procedures, energy-directed therapy—will be simulated until validated before using on patients: in essence, a virtual clinical trial on millions of computer-simulated patients over 50 years completed in 1 week of computation on a computer. This will be a "predictive process of simulation", the ultimate clinical trial. Although it will take decades to improve the methodology, first principles already valid in engineering and other scientific disciplines demonstrate the significance of this methodology, especially in rapidly assessing a new technology. The result is a new way of technology application: discovery, laboratory investigation (scientific method), predictive simulation, clinical trial. In the near term, the use of the predictive simulation will be able to dramatically reduce the length and number of subjects required to demonstrate efficacy in clinical trials (as extrapolated by the use of simulation in industry). The ultimate goal is the removal of patients from clinical trials, just as there is now a transition of using simulation in surgical skills to decrease or eliminate the need of animals in training and assessment of surgeon competency.

With this new methodology the clinical surgeon must adapt to the changing basis of providing evidence. Clearly it is no longer acceptable to base treatment upon tradition without supporting evidence (evidence-based surgery). It will be prudent to watch the emerging evidence on the predictability of simulations, for only with carefully designed computer programming will the simulation actually match the predictability of clinical trials. What the new simulation technology will be able provide that clinical trials cannot is predictability in compressed time: days instead of decades. Thus, in reading manuscripts for the latest new technology, it is critical to look at the evidence for validity. While there are well-known statistical methods that are used as the benchmarks for validation today, the practicing surgeon may soon need to learn new benchmarks that prove the validity of a simulated clinical trial.

0 0

Post a comment