Situational Ethics of Study Designs for Orthopaedic Surgery Interventions in Children With Cerebral Palsy

Tamer A. El-Sobky, MD; Mohamed El-Sayed, MD


Curr Orthop Pract. 2019;30(2):178-180. 

In This Article

Alternative Study Designs

Typically, RCTs involve blindfolding the study participants, investigators, and assessors alike. Nonetheless, in RCTs of major orthopaedic surgery or interventions this blindfolding is impractical and represents another shortcoming of RCTs in orthopaedic surgery practice. Actually, randomization aims at creating two homogenous or identical patient groups and thus intends to eliminate all actual and possible confounding study variables. In that regard any reported differences in the study outcomes between the intervention group and the control group would solely be attributable to effect of the intervention. Yet, extreme uncertainty exists over the achievability of these ideal research prerequisites. This is because random allocation of patients often results in appreciable discrepancies between the intervention and the control group due to variations in patient and disease characteristics.[3,12,13] Minimization aims to minimize the misproportion between the number of patients in each group (treatment vs. control) over a number of variables. Alternatively, future study designs could use two types of surgical treatments, i.e. two intervention groups, provided to children instead of using an intervention group and a control group with no intervention. Hip reconstruction versus hip salvage in children with CP constitutes a practical example. The use of minimization at the expense of randomization to create two homogenous patient groups can greatly enhance the level of evidence extracted, controlling bias originating from confounding variables. Such studies that experiment with the type of intervention provided also are more ethically appealing if compared to randomized case-control studies in which the control group does not receive any surgical intervention. A state of clinical equipoise between these two interventions is an important prerequisite.

Although the applicability and workability of RCTs in pediatric CP surgery may be limited due to ethical considerations, the applicability of the lower methodological studies, such as prospective observational study designs, are not as limited and are far from properly exhausted. Even more, the notion that RCTs are the gold standard of clinical evidence is being challenged on two fronts.[3,12–14] First, a considerable percentage of RCTs performed in pediatric orthopaedic practice have been associated with serious flaws in study design.[14] Second, we believe that well-executed prospective observational studies with optimal internal designs, including appropriated statistical measures and long-term follow-up, can produce credible evidence to guide clinical decision making. A large systematic review of surgical treatment of hip migration in children with CP has shown a massive tendency toward the use of "retrospective" studies.[11] Retrospective studies are inherently known for poor bias control, both selection and performance bias. This is especially true for research on children with CP because of the wide array of confounding variables, multilevel nature of orthopaedic affection, and the diversity of outcome measures implemented.[11]

The research community should start looking at alternative or existing study designs to provide a practical compromise between the ethical rights of the patient and the need to safeguard stronger evidence and guarantee valid study conclusions. Generally, well-designed and well-executed prospective studies that take into account the shortcomings of these ill-designed or retrospective studies can maximize the validity of the extracted evidence. For example, prospective observational study designs allow for accurate patient selection and allocation, adequate reporting of all actually and potentially relevant data, appropriate control for performance and assessment bias, and the choice of validated scoring systems. Ensuring long-term follow-up of these prospective observational studies can further strengthen the evidence. Therefore, if these so called "low methodological studies" are optimally exploited they can certainly provide more uniform and clear relationships between surgical procedures, patient characteristics, and awaited outcomes. Such prospective observational studies may parallel or even exceed the RCTs in terms of credibility of evidence extracted. Likewise, well-conducted systematic reviews and metaanalysis as a form of secondary research can isolate the critical shortcomings of primary research and provide useful insights for future research priorities in pediatric orthopaedic interventions in CP children.[10,11] All of the above academic remedies can aid in producing more predictability of the pediatric orthopaedic surgeons' decision making. The following recommendations can be used to bring further improvements in level of evidence and validity of studies and marginalize the risks of ethical violations simultaneously.