Hospital Volume Versus Good Clinical Practice

Abstract

This letter to the editor comments on a recently published article that assessed the treatment of patients with soft tissue sarcoma in high‐volume versus low‐volume centers.

In the recent article by Martin‐Broto et al., the authors conclude that treatment of patients with soft tissue sarcoma (STS) in high‐volume reference centers provides significantly improved outcomes compared with that in low‐volume local centers [1]. The study is based on a database of 1,470 patients with sarcoma, 1,266 patients in the study period (2004–2011), of whom 622 patients with STS and complete essential data were included in the study. There are some points throughout the manuscript that have captured our attention and merit further comments.

First, although described as prospective, the study shows certain results reminiscent of those from a retrospective study: It is published 14 years after the study was initiated, 7 years after the end of data collection, and 358 cases (24%) of the main registry (1,470 patients) lacked essential data.

Second, 70 or more patients with STS evaluated each year was required by the authors as a criterion for a center to be considered a reference center in this study. Only two centers qualified. However, one of them is not considered a reference center in Spain. Unexpectedly, large‐volume prestigious centers recently appointed as reference centers by the Spanish Health System did not meet this criterion and were included in the study as low‐volume local hospitals. The 70 threshold is not justified, and it is not clear whether this was a predefined or a post hoc criterion. In addition, there is an inconsistency in the numbers. The two reference centers must have appointments with 70 new patients with STS per year from 2004 to 2011, totalling 1,120 patients with STS. But these two centers included only 285 patients (25% of their patients) in the current study whereas the 29 local centers included 337 patients in the study (1.45 patient/year). However, 1,120 + 337 = 1,457, which exceeds the 1,266 patients with sarcomas of any type reported in the general database in the study period.

Third, median overall survival of metastatic patients was significantly shorter in local hospitals as compared with the two reference centers (18 months vs. 30 months, respectively; p = .036). Authors must provide data on metastasectomies of oligometastatic patients and should explain if survival of operated patients were accounted for reference centers when patients were referred from local centers. If there is not a metastasectomy bias, the 30‐months median overall survival achieved in the two reference centers deserves further analysis. Indeed, it doubles the reported 14 months achieved when these patients were treated with anthracycline‐based chemotherapy or with an anthracycline‐ifosfamide combination [2], which were the standard treatments at that time (2004–2011).

Finally, these results must be nuanced with the findings of a recent report from Mayo Clinic investigators [3], which shows that professionals who follow clinical guidelines achieve the same results irrespective of hospital volume when treating extra‐abdominal sarcomas with curative intention.

Disclosures

The author indicated no financial relationships.