The Need for Randomized Controlled Trials in Preoperative Perforator Imaging Techniques

Abstract

Sir: We read with great interest the recent article by Dr. Gacto-Sánchez et al. in which the role of computed tomographic angiography with three-dimensional software versus Doppler ultrasonography for preoperative perforator mapping was assessed.1 The authors reported reduced surgical time and a reduced number of complications because of computed tomographic angiography for perforator imaging. However, from an evidence-based medicine point of view, we believe we have to comment on these conclusions. The authors included 35 consecutive patients from January of 2008 to February of 2009 for computed tomographic angiography before deep inferior epigastric artery perforator (DIEP) flap breast reconstruction. A comparative study, as the authors stated, was performed retrospectively with the same surgeons in charge between May of 2004 and December of 2007 using Doppler ultrasonography only. As the authors explain elsewhere, DIEP flap surgery started in 2003 with 35 to 40 cases per year. Unfortunately, the authors do not elaborate on details of preoperative Doppler ultrasonography, time of acquisition, Doppler ultrasonography performed by the surgeons, or Doppler or duplex ultrasonography. We would greatly appreciate if the authors could be more specific on this issue. Second, as the numbers indicate, the authors started with DIEP flap surgery in 2003 and included this initial learning curve within their “control” group of Doppler ultrasonography. It is obvious that the learning curve for performing the harvesting of a perforator flap well accounts for a significant amount of time in their control group, where the flap harvesting time, the ischemia time, and the total operative time were significantly longer than in their intervention group using computed tomographic angiography. Nevertheless, length of intravenous analgesia, transfusions, and reexploration rate were higher in the control group. It is debatable how long a DIEP surgical learning curve is; however, there is one.2 Given the fact that no randomized controlled trial—or even a prospective, nonrandomized, controlled trial—was performed, we can hardly accept the assumptions made by the authors as they state in the discussion, that “computed tomographic angiography-guided VirSSPA […] have been directly translated in improved outcomes.” Notably, to date, no randomized controlled study of adequate power has demonstrated the superiority of multidetector computed tomography, for example, in contrast to Doppler ultrasonography. In evidence-based medicine, we seek to provide good evidence, even in surgical trials, to change practice. Although the current publication wave focuses mainly on multidetector computed tomographic angiography for preoperative perforator imaging, we believe this has to be taken into account despite the documented accuracy of perforator imaging in multiple trials.3,4 Last, we would like to highlight another yet underreported effect of any preoperative perforator imaging technique. Given the surgical learning curve in perforator dissection, it appears, at least to us, that especially novice perforator surgeons might benefit from any preoperative information regarding the potential course of the perforators. Rozen et al. reported reduced stress levels of the surgeon in a retrospective assessment when using preoperative computed tomographic angiography.5 Therefore, it would be worthwhile, when performing a randomized controlled trial comparing, for example, preoperative Doppler ultrasonography with computed tomographic angiography, to evaluate the stress levels of the surgeons in both arms simultaneously in addition to harvesting time, complication rate, and length of hospital stay. Karsten Knobloch, M.D. Joern Redeker, M.D. Peter M. Vogt, M.D., Ph.D. Plastic, Hand, and Reconstructive Surgery Hannover Medical School Hannover, Germany