Important STS National Database Research Featured as Clark Papers

Richard E. Clark Memorial Papers


8:15 a.m. – 9:00 a.m.

Grand Ballroom

Since its inception in 1989, the STS National Database has played an essential role in generating research on quality improvement and patient safety. Three of the latest practice-changing studies—one each from the Adult Cardiac Surgery Database (ACSD), the Congenital Heart Surgery Database (CHSD), and the General Thoracic Surgery Database (GTSD)—will be presented as the 2017 Richard E. Clark Memorial Papers.

Surgical Ablation for Atrial Fibrillation Concomitant to Mitral Operations Reduces Mortality

J. Scott Rankin, MD

J. Scott Rankin, MD

The Clark Paper for Adult Cardiac Surgery suggests that performing surgical ablation (SA) to treat atrial fibrillation (AFib) at the time of mitral valve repair or replacement (MVRR) may have a protective effect on mortality.

Adding SA during MVRR is known to improve late sinus rhythm. However, clinicians have been unsure of the operative mortality impact of performing the two procedures concomitantly. Early studies indicated little effect on mortality, while more recent studies suggest a reduction in operative mortality.

Researchers investigated the ACSD for MVRR patents between July 2011 and June 2014. Including tricuspid repair and coronary artery bypass grafting generated a cohort of 88,765 patients.

Risk-adjusted odds ratios for mortality were compared at the time of surgery whether or not SA was performed. Group 1, with no preop AFib and no SA, was the comparator for other groups. Group 2 had no immediate preop AFib but had SA. Group 3 had AFib but no SA. Group 4 had AFib plus SA.

Lead author J. Scott Rankin, MD, of West Virginia University in Morgantown, will present “Mortality Is Reduced When Surgical Ablation for Atrial Fibrillation Is Performed Concomitantly With Mitral Operations.”

Patients in group 3 who had AFib but who did not receive SA had an odds ratio of 1.16 for mortality, or a 16% increase in relative risk of mortality compared with group 1 patients with no preop AFib or SA. But patients in group 4 who had AFib plus SA had a mortality similar to group 1 patients.

“At the time of mitral operations, the addition of SA to treat AFib can be performed without increased risk of mortality and may even be protective,” Dr. Rankin said. “The data suggest an early mortality benefit for SA and imply that further increase in SA application may be appropriate.”

Preoperative Risk Factors in Early Shunt Failure

Nhue Do, MD

Nhue Do, MD

A new analysis of congenital heart surgery data was performed to determine the incidence of early shunt failure and identify specific patient groups that may be at increased risk of functional failure of a systemic-to-pulmonary artery shunt or a ventricle-to-pulmonary artery shunt.

Early failure of systemic-to-pulmonary shunts is a potentially catastrophic complication in infants. But evidence to identify preoperative risk factors has been lacking, with most studies focusing on prevention of failure using pharmacologic or other therapies but having small sample sizes or being underpowered.

Researchers queried the CHSD to find 9,172 infants aged 1 year or younger who underwent shunt construction as the primary source of pulmonary blood flow from 2010 to 2015 at 118 centers. The cohort included both systemic-to-pulmonary artery and right ventricle-to-pulmonary shunts. The study found in-hospital early shunt failure occurred in 674 infants (7.4%) overall and that patients with early in-hospital shunt failure had significantly higher operative mortality and major morbidity and longer postoperative lengths of stay.

Lead author Nhue Do, MD, of Johns Hopkins School of Medicine in Baltimore, will present the Clark Paper for Congenital Heart Disease, “Early Shunt Failure, Prevalence, Risk Factors, and Outcomes: An Analysis of The Society of Thoracic Surgeons Congenital Heart Surgery Database.”

Many of the identified in-hospital risk factors were intuitively expected, Dr. Do said, such as low body weight at the time of shunt creation and the presence of a preoperative hypercoagulable state. Neither use or nonuse of cardiopulmonary bypass during shunt creation was associated with an increased risk of early shunt failure. And there was no increased risk of early shunt failure in single ventricle patients, including those with hypoplastic left heart syndrome.

“The data suggest that the Norwood operation with a right ventricle-to-pulmonary artery shunt was somewhat protective against early shunt failure,” Dr. Do said. “There has been debate over whether to do a right ventricle-to-pulmonary artery shunt versus a modified Blalock-Thomas-Taussig shunt because of the need for a ventriculotomy, but we found it to be associated with decreased risk of early shunt failure.”

The study confirmed that in-hospital shunt failure is both common and associated with high mortality. Results also highlight higher-risk patients and cohorts that may benefit from increased surveillance, enhanced anti-thrombotic prophylaxis, or other strategies to minimize the risk of shunt failure.

New Quality Measure for Esophageal Cancer Surgery

Andrew C. Cheng, MD

Andrew C. Cheng, MD

Investigators have developed a composite quality measure for esophagectomy for esophageal cancer using data from the GTSD.

The researchers queried the GTSD for esophagectomies performed at 167 participating centers between 2012 and 2014 to ascertain risk-adjusted operative mortality (at discharge and at 30 days post-surgery) and risk-adjusted major complications.

Participants whose 95% Bayesian credible intervals (CrI) overlapped the STS mean composite score were considered two-star participants, while those whose 95% CrI were entirely below or above the STS mean were classified as one- or three-star sites, respectively. Discharge mortality and lengths of stay were used to benchmark GTSD participants against the National Inpatient Sample 2012 cohort.

Andrew C. Chang, MD, of the University of Michigan in Ann Arbor, will present the Clark Paper for General Thoracic Surgery, “The Society of Thoracic Surgeons Composite Score for Evaluating Program Performance in Esophagectomy for Esophageal Cancer.”

Operative mortality—a combination of discharge and 30-day mortality—was 3.1%, and the major complication rate was 33.1%. Of the 167 participants, only 70 reported an average yearly operative volume of five or more esophagectomies during the study period.

Of these 70 participants, four (5.7%) were three-star, 64 (91.4%) were two-star, and two (2.9%) were one-star. The remaining 97 (58.1%) participants did not have sufficient operative volume to score reliably.

“The pressing concern is that a significant majority of participants don’t do enough esophageal cancer operations to get valid measures,” Dr. Chang said. “We also find that there are programs that have lower volumes and very good results. This measure still needs to be validated, but it demonstrates that we can measure quality in esophagectomies.”

The next step, he added, is to consider how widely the measure can be used. It would be difficult to apply a quality measure knowing that more than half of GTSD participants would not qualify due to low volume.

Return to articles