Objective: To describe the process involved in using the World Wide Web to coordinate a randomized, multicenter international trial of treatment for twin-twin transfusion syndrome.
Method: A Web site was designed by members of the research team, a Web consultant, and a senior computer programmer. The original intent was to provide patient randomization only, but the Web site later was designed so that centers could download a data collection form. Data could be entered directly into the Web site and subsequently imported into a database at the coordinating center.
Experience: The Web site has been active for 3 years, with 13 participating centers and 31 patients enrolled.
Comment: Use of the World Wide Web to coordinate an international, multicenter trial is an efficient method. Although there are many benefits, the most obvious is the capability to initiate and conduct a large international trial at minimal cost.
Randomized trials represent the highest order of scientific clinical investigation. When the clinical condition of interest or the outcome variable occurs infrequently, collaboration between multiple centers is necessary to enroll a sufficient number of patients in a reasonable period. Methods of random allocation of treatments in these trials vary but usually involve assignment to treatment either locally by individual center randomization or by central randomization with the use of a toll-free telephone number. The latter is used more frequently in multicenter trials and offers several advantages such as the capability to screen patients. However, central randomization is costly because staff are needed to answer telephones and enter data. Differences in time zones of participating centers and the need to enroll patients after typical work hours at the central randomization center often make this approach impractical. Central randomization becomes more problematic in international clinical trials, in which differences in time zones are even greater.
Improved central randomization systems have been developed, such as the facsimile–optical character reader system, in which a patient information datasheet is faxed to the central randomization center.1 The data are entered automatically into a computer program that checks eligibility criteria, performs the treatment allocation, and notifies the enrolling center of the treatment. If the patient does not meet eligibility criteria, a rejection notice is faxed to the referring center. Another system incorporates telephone access and voice-recognition software linked to decision-making software.2 Both systems are complex and require additional software and hardware, thereby increasing cost and the chance of malfunction.
Investigators interested in studying certain fetal conditions with low prevalences must conduct international multicenter trials, which make it possible to enroll sufficient numbers of subjects. An example of such a condition is twin-twin transfusion syndrome, which complicates up to 35% of monochorionic, diamniotic gestations.3 Untreated twin-twin transfusion syndrome has been associated with a perinatal mortality of 60–100%.4–7 Therapies include laser ablation of anastomotic placental vessels,8 serial amnioreduction of excessive amniotic fluid,9 and amniotic septostomy.10 Despite divergent costs and rates of maternal morbidity, perinatal survival rates of 50, 49, and 83%, respectively, have been reported for these three treatments. There are, however, no published randomized trials comparing these treatments with one another or with expectant management. To compare serial amnioreduction and amniotic septostomy, we proposed a multicenter international trial at the tenth annual meeting of the International Fetal Medicine and Surgery Society. The purpose of this report is to describe the use of the World Wide Web to coordinate this international trial.
An international multicenter clinical trial of treamtent for twin-twin transfusion syndrome was conducted using the Internet for randomization.
Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina; the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Texas Branch at Galveston, Galveston, Texas; and the Department of Medical Illustrations, Baylor College of Medicine, Houston, Texas.
Address reprint requests to: Karen Dorman, RNC, MS, Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, CB #7576, 214 MacNider Building, Chapel Hill, NC 27599; E-mail: firstname.lastname@example.org
Received January 10, 2000. Received in revised form April 17, 2000. Accepted April 27, 2000.