CONGENITAL: PERIOPERATIVE MANAGEMENT: EDITORIAL

Cardiac reoperation: Should that be a marker of quality?

Bahaaldin Alsoufi, MD

From the Department of Cardiovascular and Thoracic Surgery, University of Louisville School of Medicine,
Norton Children’s Hospital, Louisville, Ky.
Disclosures: Author has nothing to disclose with regard to commercial support.
Received for publication July 8, 2018; accepted for publication July 10, 2018; available ahead of print Aug 7,

CONG
2018.
Address for reprints: Bahaaldin Alsoufi, MD, Department of Cardiovascular and Thoracic Surgery, University of
Louisville School of Medicine, Norton Children’s Hospital, 201 Abraham Flexner Way, Suite 1200, Louisville,
KY 40202 (E-mail: [email protected]).
J Thorac Cardiovasc Surg 2018;156:1959-60
0022-5223/$36.00
Copyright Ó 2018 by The American Association for Thoracic Surgery
https://doi.org/10.1016/j.jtcvs.2018.07.015
Bahaaldin Alsoufi, MD

Outcomes of surgical repair and palliation of congenital Central Message

heart disease have improved remarkably in the past 1 to Use of registry-derived cardiac reoperations as
2 decades as a result of numerous advances in cardiac imag- a quality marker is limited by lack of data gran-
ing, preoperative stabilization, surgical strategies, perfusion ularity to differentiate expected reoperations
techniques, and perioperative care. Along with those devel- from those related to complications or residual
lesions.
opments have come higher expectations from families,
referring physicians, insurance companies, and hospital
administration. In the same time frame, various multi- See Article page 1961.
institutional surgical and critical care registries were
created, with the aim of using those large databases for
risk assessment, clinical outcome research, benchmarking
and evaluation of program performance, and implementa- Surgeons–European Association for Cardiothoracic Sur-
tion of quality improvement projects.1-4 Recently, there geons Congenital Heart Surgery Mortality Categories
has been an increased emphasis on transparency and (STAT Mortality Categories).3,4 Subsequent refinements
public reporting of institutional outcomes, with the goal of the STAT Mortality Categories are underway in an
of providing families with valuable information while effort to perform more sophisticated risk adjustment that
simultaneously encouraging programs to achieve superior takes into account various clinical and demographic
outcomes. Although those objectives are uniformly factors (for example, genetic syndromes and extracardiac
desired, a considerable skepticism was legitimately anomalies).3,4
generated because of the lack of ideal risk-stratification In the current issue of the Journal, Gupta and colleagues6
models that take into consideration the numerous demo- focus on one clinical variable that can influence outcomes
graphic, clinical, and anatomic variables that intimately in- after congenital cardiac surgery, and that is cardiac reoper-
fluence outcomes. This uncertainty was accompanied by ation within the same admission for index operation. Their
concerns about negative implications of public reporting, hypothesis is that the STAT Mortality Categories risk-
such as risk aversion, program regression, and potential ef- stratification system centers on the initial (index) operation,
fects on government regulation and reimbursement and that in some cases the subsequent reoperation might be
strategies. more complex and associated with a higher STAT category;
Multiple surgical risk-stratification models have been therefore classifying patients according to the procedure of
developed; however, many of those are procedure related highest complexity, rather than the index operation, would
and are based mainly on expert opinion.5 The subsequent thus likely change the ratio of observed to predicted mortal-
creation of large congenital heart surgery databases has al- ity for that patient. To study their hypothesis, they examined
lowed analysis of 77,294 operations entered into the data from the Virtual Pediatric Systems Database on chil-
congenital heart surgery databases of the Society of dren younger than 18 years old who had undergone congen-
Thoracic Surgeons (n ¼ 43,934 patients) and the European ital cardiac surgery, with or without cardiopulmonary
Congenital Heart Surgeons Association and European As- bypass, between 2009 and 2015, and limited their patient
sociation for Cardio-Thoracic Surgery (n ¼ 33,360 pa- cohort to those who underwent at least 1 cardiac reoperation
tients). These database analyses consequently allowed the (22,393/51,047 patients). Naturally, they found that in an
development of a classification that was based on objective important number of patients (7691/22,393, 34%), the car-
rather than subjective data, the Society of Thoracic diac reoperation was associated with a higher complexity

The Journal of Thoracic and Cardiovascular Surgery c Volume 156, Number 5 1959
 Congenital: Perioperative Management: Editorial Alsoufi

STAT category. As a result, by assigning patients to the contrast, it may be a marker of excellence in centers that
higher complexity operation category rather than the index are taking on patients at very high risk and implementing
operation, the number of patients in the lower STAT cate- thoughtful staged approaches to overcome the many clinical
gory decreased and the number of those in the higher and anatomic challenges that these patients might pose. In
STAT category increased. Not surprisingly, the ratio of 34% of cases in the study by Gupta and colleagues6 cardiac
observed to predicted mortality was higher when patients reoperation was more complex and could be assigned to a
were assigned to the lower STAT category of the index oper- higher STAT category. When cardiac reoperation is un-
CONG

ation rather than to that of the succeeding, higher planned, assigning the patient to the highest complexity
complexity operation. Naturally, this difference was highest operation rather than the intended index operation might
in patients with lower index STAT categories (I, II, and III), seem like turning a blind eye to poor performance in a ma-
and it was also true for extracorporeal membrane oxygena- jority of instances. On the other hand, when cardiac reoper-
tion use, cardiac arrest, and other major surgical ation is planned or expected, assigning the patient to the
complications. highest complexity operation rather than the index opera-
Although the results of this study are straightforward and tion might be the fair thing to do for centers taking care
expected, the question arises in this era as to how to use car- of patients whose care is complex and high risk. The ques-
diac reoperation as a marker of quality and how to incorpo- tion arises whether the current registry databases are gran-
rate that into a model that is used to assess program ular enough to differentiate those completely different
performance. Cardiac reoperation can be planned as a staged groups of cardiac reoperations to allow the use of their
strategy in the treatment of patients with complex congenital data for assessment of program performance. My guess is
cardiac anomalies (eg, pacemaker implantation for cardio- that the answer is no. Although the article by Gupta and col-
myopathy followed by heart transplantation, coarctation leagues6 may seem to be stating the obvious, it does high-
repair followed by atrioventricular septal defect repair, palli- light another limitation of the current risk-stratification
ative aortic valvuloplasty followed by Ross-Konno opera- models and call attention to another area where refinements
tion) or with clinical conditions that prohibit a single stage- of those models can be attempted, with the aim of creating
approach (eg, pulmonary artery branch banding followed an ideal system that serves the objectives of quality
by Norwood operation in patients deemed to face a high improvement while being fair and truly representative of
risk with a primary Norwood operation, coarctation repair truthful risk at the same time.
followed by arterial switch and ventricular septal defect
closure in very small neonates). On the other hand, cardiac References
reoperation may be unplanned as a result of a missed diag- 1. Jacobs JP, Mayer JE Jr, Pasquali SK, Hill KD, Overman DM, St Louis JD, et al.
The Society of Thoracic Surgeons congenital heart surgery database: 2018 update
nosis, residual lesion, or surgical complication (atrioventric- on outcomes and quality. Ann Thorac Surg. 2018;105:680-9.
ular valve replacement after partial atrioventricular septal 2. Jacobs ML, Jacobs JP, Hill KD, Hornik C, O’Brien SM, Pasquali SK, et al. The
defect repair, tricuspid valve repair after ventricular septal Society of Thoracic Surgeons congenital heart surgery database: 2017 update on
research. Ann Thorac Surg. 2017;104:731-41.
defect closure, aortic or mitral valve repair after subaortic 3. O’Brien SM, Jacobs JP, Pasquali SK, Gaynor JW, Karamlou T, Welke KF,
membrane resection, pulmonary vein stenosis repair after et al. The Society of Thoracic Surgeons congenital heart surgery database
repair or palliation of other cardiac lesions, aortic arch repair mortality risk model: part 1—statistical methodology. Ann Thorac Surg.
2015;100:1054-62.
for a previously unidentified aortic obstruction, ascending 4. Jacobs JP, O’Brien SM, Pasquali SK, Gaynor JW, Mayer JE Jr, Karamlou T, et al.
aortic replacement for iatrogenic aneurysm or obstruction, The Society of Thoracic Surgeons congenital heart surgery database mortality risk
pacemaker implantation after ventricular septal defect model: part 2—clinical application. Ann Thorac Surg. 2015;100:1063-8; discus-
sion 1068-70.
closure, pulmonary artery banding for a residual intracardiac 5. Jacobs JP, Jacobs ML, Lacour-Gayet FG, Jenkins KJ, Gauvreau K, Bacha E, et al.
shunt, aortopulmonary shunt for residual pulmonary artery Stratification of complexity improves the utility and accuracy of outcomes anal-
obstruction, and Norwood operation after pulmonary artery ysis in a multi-institutional congenital heart surgery database: application of the
risk adjustment in congenital heart surgery (RACHS-1) and Aristotle systems in
banding for single-ventricle anomalies and unrecognized the Society of Thoracic Surgeons (STS) congenital heart surgery database. Pediatr
systemic outflow tract obstruction, to name a few). Cardiol. 2009;30:1117-30.
Cardiac reoperation thus may be a marker of poor perfor- 6. Gupta P, Rettiganti M, Shinkawa T, Gossett JM, Brundage N, Jeffries HE, et al.
Reclassifying by highest complexity operation rather than first operation influ-
mance, especially when it is unplanned because of missed ences mortality after pediatric heart surgery. J Thorac and Cardiovasc Surg.
diagnosis, surgical complication, or residual lesion; in 2018;156:1961-7.

1960 The Journal of Thoracic and Cardiovascular Surgery c November 2018