Catheter-Based Computed Tomography Angiography in Anterolateral Thigh Perforator Mapping of Chinese Patients: Statistical Estimates

Catheter-Based Computed Tomography Angiography in Anterolateral Thigh Perforator Mapping of Chinese Patients

With interest, we read the article, “Catheter-Based Computed Tomography Angiography in Anterolateral Thigh Perforator Mapping of Chinese Patients,” by Wang et al, published in Journal of Reconstructive Microsurgery in October 2018.[1] Free anterolateral thigh (ALT) flap is a common reconstructive surgical modality. However, due to the unpredictability of the perforator vessels, ALT flap has been bothersome even for the experienced microsurgeons. The purpose of their study was to evaluate the predictive value of catheter-based computed tomography angiography (C-CTA) for preoperative mapping of perforators, compared with the traditional CTA (T-CTA). They prospectively enrolled 34 Chinese patients for ALT flap. For T-CTA, the overall sensitivity was 27.6%, specificity 66.7%, and accuracy 36.7%, with a false-negative rate (FNR) of 72.3% and false-positive rate (FPR) of 26.5%. For C-CTA, the overall sensitivity was 90.0%, specificity 94.7%, and accuracy 91.1%, with a false-negative rate (FNR) of 10.0% and false-positive rate (FPR) of 1.7%. They concluded that, compared with T-CTA, C-CTA had greater efficiency in preoperative evaluation for perforators.

However, there is a methodological issue. When evaluating the validity of a test, compared to the gold standard, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), likelihood ratio positive (LR+), likelihood ratio negative (LR−), and diagnostic odds ratio are the most important estimates, instead of the FPR and FNR used in their study, which stand for types I and II errors in medicine. Moreover, when we report diagnostic added values in clinical occasions, receiver operating characteristic (ROC) should be considered. Since all these above-mentioned validity estimates could be acceptable while the diagnostic added values might be clinically negligible.[2] [3] In summary, their conclusion would be more convincing if supported by the above statistical methodology.

Xiang Fang and Weili Kong contributed equally to this work. They are co-first authors.

• References

• 1 Wang Z, Yi X, He J. , et al. Catheter-based computed tomography angiography in anterolateral thigh perforator mapping of Chinese patients. J Reconstr Microsurg 2019; 35 (03) 221-228
• 2 Szklo M, Nieto FJ. Epidemiology Beyond the Basics. 2nd ed. Sudbury, MA: Jones & Bartlett Learning; 2007
• 3 Rothman KJ, Greenland S, Lash TL. , eds. Cohort studies. In: Modern Epidemiology. Philadelphia, PA: Lippincott Williams & Wilkins; 2008: 100-110