We thank Dr Tafur et al for their comments on our article, which raises issues specific to the gynecologic oncology patient. Regarding the utility of the risk assessment tools, we agree that when the highest-risk group of the Caprini score is substratified (score ≥5 group divided into smaller groups), it is highly correlated with venous thromboembolism (VTE) as reported in Figure 1 of our paper.

In this setting, the Caprini score has utility to risk stratify gynecologic oncology patients. However, both the American College of Chest Physicians (ACCP) chest guidelines and the 2005 Caprini risk assessment model (RAM) define a Caprini score of ≥5 as a single highest-risk group (although, as noted by Dr Tafur, the lower-risk groups are defined differently by the ACCP and Caprini risk assessment model as seen in Tables 2 and 3 of our paper , respectively). We found that 97% of gynecologic oncology patients have a score of ≥5, meaning that risk stratification is limited if all are in a single group.

Placing all patients in a single high-risk group would be acceptable if all gynecologic oncology patients harbored a high risk of VTE and all required maximum prophylaxis. However, we believe the Caprini score highest-risk group overestimates the VTE risk for a significant proportion of gynecologic oncology patents. In modern gynecologic oncology surgery, greater than 40% of patients undergo minimally invasive surgery (MIS), and these percentages are increasing each year. Patients undergoing MIS have a risk of 30 day VTE as low as 0.57%, even with no perioperative mechanical or pharmacologic prophylaxis. The Caprini score assigns the same point values to patients undergoing open laparotomy or MIS, although the VTE risk is quite different (relative risks of 3.9 and 3.1 for VTE for open vs MIS).

The cited validation of the Caprini score in gynecologic oncology included only patients undergoing laparotomy and did not address MIS. Furthermore, modern-era prophylaxis is not one size fits all. According to ACCP guidelines, patients undergoing abdominopelvic operations for cancer with a Caprini score of ≥5 require mechanical, pharmacologic, and extended-duration prophylaxis. The benefit of extended-duration prophylaxis or even pharmacologic prophylaxis for gynecologic oncology MIS is unclear.

We agree that with substratification of the highest risk group, the Caprini score could be a useful tool to preoperatively assess gynecologic oncology patients. However, the inability to distinguish VTE risk between patients undergoing MIS and open surgery limits the Caprini score’s utility in this population.