To the Editor:
We read with great interest the article by Gottschalk et al .1and wish to congratulate them for their great and remarkable work. They showed that, although the use of epidural analgesia for perioperative pain control during colorectal cancer surgery was not associated with a lower rate of cancer recurrence, a potential benefit was observed in older patients. Their results were in contrast with those of Christopherson et al. ,2whose study population was similar, apart from sample size.
Nevertheless, the primary endpoints of these two studies were different. Gottschalk et al. 1reported the incidence of cancer recurrence whereas Christopherson et al. 2found a difference in terms of survival.
However, the definition of cancer recurrence used by Gottschalk et al. 1would be of interest. The application of their findings may be limited as a result of this omission. Establishing a diagnosis for local recurrence can be particularly difficult in rectal cancer patients. It is for this reason that researchers3often consider reported rates of recurrence in this population to be of limited value. In this study, readers are uncertain about study methodology, potential bias, and the incidence of cancer-related death in both groups (epidural vs . nonepidural). Therefore, the results of Gottschalk et al. 1might best be considered inconclusive.
Moreover, epidural analgesia is not the only intraoperative variable that could influence cancer outcomes. In the study by Gottschalk et al. ,1the epidural group presented with more rectal cancer, higher histologic grade, more frequent radiotherapy and chemotherapy, and more blood loss, suggesting more difficult surgeries. All these characteristics are potential confounding factors that may impact the rate of cancer recurrence. Even if multivariate analysis has been well performed, these patients' characteristics increases the risk that the epidural group includes other unknown negative factors. In addition, Gottschalk et al. ,1do not make available data concerning the number of nodes removed or the quality of the mesorectum—both indicators of surgery quality.
Finally, another potential bias in this study is the intraoperative use of nonsteroidal antiinflammatory drugs. The overexpression of cyclooxygenase type 2 and the possible immunoprotective effect of these drugs may have an impact on long-term cancer recurrence.4It would be of interest to know if the groups are balanced for the use of these medications. Indeed, the older patients are often more prone to not receive nonsteroidal anti-inflammatory drugs. And the effect of the epidural analgesia might then have been unmasked for this reason. Following this hypothesis, in younger patients, these drugs may have masked the effect of epidural analgesia.
In conclusion, interpretation of the results reported by Gottschalk et al. 1must wait until their methodology is further clarified.
* Université catholique de Louvain, Steet-Luc University Hospital, Brussels, Belgium. firstname.lastname@example.org