We thank our colleagues for their interest regarding our recent work.1In response to their inquiries, recurrence  in our study was defined as any (local or metastatic) detection of colon cancer after primary resection. In the commonwealth of Virginia, treating physicians are required by law to report the cancer status of all patients. The University of Virginia Cancer Center, Charlottesville, tracks this data. Therefore, we are fortunate to have access to long-term follow-up cancer recurrence data on a large number of patients. However, we fully acknowledge that any retrospective study, including ours, is limited by (1) the accuracy of the available medical records, which may include missing data, and (2) difficulty controlling bias and confounding factors that could influence cancer recurrence (e.g ., α and β blockers, statins, nonsteroidal antiinflammatory drugs, cyclooxygenase inhibitors).

We agree with Dr. Tiouririne that intraoperative use of epidural analgesia (i.e ., to supplement general anesthetics) may have different effects on cancer recurrence than epidural analgesia used only postoperatively. As Christopherson et al.  2note, a variety of factors influence cancer recurrence. For example, cancer stage and grade are almost always the best predictors of recurrence. Although our analysis corrected for major factors, our statistical modeling was, of course, restricted to the available data.

Both letters assert that our findings contradict those of Christopherson et al.  2However, this interpretation of our results is inaccurate; neither we nor Christopherson et al.  2found an overall (primary hypothesis) benefit of epidural analgesia. Unplanned post hoc  subgroup analyses—including our observation that cancer recurrence was reduced in older patients who received epidural analgesia—are notoriously unreliable. Indeed, such analyses, when statistically significant at 0.05, have only a 57% chance of being replicated in an identical clinical trial.3 

Although the idea that regional analgesia may reduce the incidence of cancer recurrence is exciting, it remains a hypothesis at this time—a question that can be answered only with prospective randomized clinical trials. Fortunately, several such studies are already in progress.

* University of Virginia Health System, Charlottesville, Virginia. en3x@virginia.edu

1.
Gottschalk A, Ford JG, Regelin CC, You J, Mascha EJ, Sessler DI, Durieux ME, Nemergut EC: Association between epidural analgesia and cancer recurrence after colorectal cancer surgery. Anesthesiology 2010; 113:27–34
2.
Christopherson R, James KE, Tableman M, Marshall P, Johnson FE: Long-term survival after colon cancer surgery: A variation associated with choice of anesthesia. Anesth Analg 2008; 107:325–32
3.
O'Neill RT: Secondary endpoints cannot be validly analyzed if the primary endpoint does not demonstrate clear statistical significance. Control Clin Trials 1997; 18:550–6