Quality Accounting: Understanding the Impact of Multiple Surgical Complications
In this issue of Annals of Surgery, Dr. Clavien and colleagues present the results of a 1-year prospective review of the CCI's© utility. 5 They report that, of 290 patients who developed a complication, 127 (44%) developed more than 1. Those with more severe complications were also more likely to have multiple complications. When the postoperative window for assessing complications was lengthened from hospital discharge to 3 months after surgery, both the incidence of multiple complications and the CCI© increased significantly, emphasizing the value of longitudinal follow-up beyond hospital discharge or a 30-day window. These findings reinforce the conclusions of their earlier report, which showed that the CCI© provides information that is complementary to assessment of the highest-grade complication alone. 4
The CCI© could be more widely incorporated into morbidity and mortality conferences in its current form. Our tendency in that forum is often to focus on the complication most proximate to the operation (eg, the pancreatic fistula that leads to pseudoaneurysm, delayed gastric emptying, and wound infection) or the one that leads to the most significant intervention (eg, the anastomotic leak that leads to reoperation). The American College of Surgeons National Surgical Quality Improvement Program reports individual complications and, for selected procedures, the “most severe clinical outcome.” But we often lack a means of summarizing the collective impact of multiple complications. The CCI© provides such a measure that can be broadly applied to a variety of operations and can be intuitively understood across subspecialties. The authors’ survey of (primarily European) surgeons indicates that 42% of respondents use the CCI© routinely, and 27% use it occasionally. 5 In our opinion, it could be adopted more widely as a descriptive tool, especially given the convenient web-based calculator provided by the authors.
A particularly intriguing notion is that the CCI© might be useful as an endpoint in randomized controlled trials (RCTs) for surgical procedures. On the basis of reanalysis of data from 3 RCTs, the authors have previously suggested that the CCI© may better detect differences in treatment effects than endpoints such as any complication or severe complications. 6 For example, a large multicenter RCT demonstrated that external pancreatic duct stenting reduced pancreatic fistula after pancreaticoduodenectomy. 7 Although there was no difference in the incidence of “severe” complications (≥grade IIIb), reanalysis of the data using the CCI© revealed a statistically significant 12-point difference, providing a useful summary of the clinical impact of the collective impact of stent use, pancreatic fistulas, and related sequelae.