April 2001 Bulletin

Follow the rules for any clinical research

By E. Michael Keating, MD, and John Meding, MD

Research is defined by the unabridged Random House dictionary of the English language as the "diligent and systematic inquiry or investigation into a subject in order to discover or revise facts, theories, applications, etc." This simply means that any systematic investigation within our practice, in hospitals or offices, comes under the realm of research.

We are all familiar with research sponsored by drug and orthopaedic companies to investigate new prostheses, drugs, or techniques. However, many of us are unaware of what constitutes research and how to conduct unsponsored research. Unsponsored research is research that you or your hospital does to develop better practices or applications of existing treatment, prostheses, or methodology. This can take the form of answering the questions of which type of prosthesis is superior, what surgical approach provides the best results, what type of post-op dressing protocol is needed, or what pain control method is the most efficacious.

These types of investigations, while not sponsored by a company are still considered "research" by the Food and Drug Administration (FDA), your hospital Institutional Review Board (IRB), and your institution. For example, you and/or your hospital may decide to evaluate whether CPM machines in conjunction with physical therapy provide faster or better rehabilitation than physical therapy alone after total knee replacement. If this study is performed prospectively or systematically, then it definitely constitutes research. FDA and IRB guidelines must be followed. The study must be approved by the IRB and have full patient disclosure and informed consent.

Unfortunately, many people not conducting research on an every day basis don’t realize this important fact. If you switched prostheses that you implanted five years ago and now want to compare the results of the new prosthesis to the old prosthesis, this systematic comparison also falls into the realm of research and should be approved by the IRB and have patient informed consent.

The rules on this type of research will be discussed later. They are not as stringent from either the FDA or the IRB standpoint as research that is sponsored. In this case the IRB approval and patient informed consent is more general and can be obtained by gaining IRB approval and asking all the patients with the prostheses if the results of their hip or knee surgery can be evaluated to see which one works the best. However, this must be done with patient informed consent.

It may be possible to do this without informed consent if only clinic records are reviewed, all data is anonymous, and no patients are contacted. If patients are contacted, as they almost always are, then informed consent must be obtained. The Journal of Bone and Joint Surgery requires informed consent and IRB approval for any article published.

When do you go to the IRB and obtain patient informed consent? It is whenever you do any type of systematic evaluation on a portion of your practice involving hospital care.

How does the IRB evaluate these studies? The first and foremost job of the IRB is patient safety. Research can be broken down into two categories of study: those with significant risk (SR) and those with non-significant risk (NSR). Many studies involve a medical device that is to be evaluated for safety and effectiveness. In these studies, most if not all are sponsored by a company or investigator, and thus the investigation falls into the category of an Investigational Device Exemption (IDE). These are usually Class II or Class III devices.

An investigational device is a medical device that is the subject of clinical studies designed to evaluate the effectiveness and/or safety of the device. These investigations are undertaken to determine the safety and effectiveness of a device and must be conducted according to the requirements of the IDE regulations (21 CFR part 812) (FDA information sheet guidelines for IRB and clinical investigators).

Some clinical investigations of devices are exempt from IDE regulations. These include lawfully marketed approved devices, (for example, review of CPM vs. physical therapy, two types of approved prostheses, or two different surgical approaches.) These would fall outside of IDE regulations and would not have to go to be evaluated by the FDA, but would have to be reviewed by the IRB as a NSR study. The IRB makes a determination of SR vs. NSR. This determination has significant consequences for the study sponsor (if there is one), FDA, and prospective research subjects. SR studies must be conducted in accordance with the full IDE requirements. SR studies must be to be approved by the FDA and the IRB and requires reporting requirements that are determined by the FDA and are very stringent.

NSR studies do not require submission of an application to the FDA. Instead, the IRB can approve the study with "abbreviated requirements" of the IDE regulations. These, among other things, require IRB approval, informed consent, record keeping, and study monitoring. The fact that the IRB determines that the study is NSR and doesn’t have to be reported to the FDA doesn’t mean that the FDA can’t disagree later and require filing of the full IDE paperwork with the FDA. If there is a question about classification, the sponsor or IRB should voluntarily seek advice from and inform the FDA about the decision to proceed without filing an IDE application with the FDA. Information about device classification can be obtained by contacting the FDA or by searching the FDA homepage.

Whether the device is classified SR or NSR, the IRB and FDA have strong concerns about patient disclosure and safety. It is unacceptable to perform any type of research without full patient disclosure about the type of study being performed. An informed consent must be obtained in almost all cases. It is also imperative that patient safety and anonymity be utmost in the investigation. (continued on p.52)

Another area the FDA has been active in is full disclosure of any possible financial conflict of interest. This could be in the form of financial incentives, cash payments, or stock, which you own in the sponsoring company, be it large or small. Any conflict of interest, whether real or perceived, must be reported to the FDA by the sponsor company or individual if the study is not sponsored. This will allow the FDA to evaluate the study with any potential conflict of interest. Failure to do this will cause the FDA not to be able to accept the marketing application or reclassification petition for a device and possibly suspend research.

Greg Koski, MD, PhD, is the new head of the Office for Human Research Protections and has stated that non-compliance with regulations is unacceptable. In a recent address to IRB boards, he affirmed that institutions or individuals that fail to accept this responsibility will not be allowed to engage in research. In doing clinical research in our hospitals and clinics, even if we don’t plan to publish the results, we must follow the guidelines of the FDA, and requirements of the IRB. Thus, any research utilizing Class II or Class III devices must comply with IDE regulations and have full IRB approval. Any clinical research that we might consider a NSR must still get abbreviated IRB approval and should have informed consent before we undertake it, if we are going to adhere to the principals of the Declaration of Helsinki, as published in the JBJS. As orthopaedic surgeons, if we intend to do research of any type, we should know these requirements and follow them to the letter.


  1. Fuson, R.L.; Sherman, M.; Van Fleet, J.; Wendt, T.; The Conduct ofr Orthopaedic Clinical Trials — JBJS 79A: 1089 — 1098. 1997
  2. FDA Home page guidelines for IRB’s and Clinical Investigations http:// www.fda.gov/cdrh/d891.html

E. Michael Keating, MD, is a member of the Biomedical Engineering Committee. Dr. Keating and John Meding, MD, are partners in Mooresville, Ind.

Home Previous Page