Infective endocarditis has a global incidence of 3-10 cases per 100,000 people and an approximate mortality rate of 25%. In a European cohort, the incidence drops to 33.8 cases per million inhabitants with a similar hospital mortality rate of 22.7%. Over the past decade, the incidence of this disease has doubled, and current trends suggest it will continue to rise. In 2019, the Journal of Thoracic and Cardiovascular Surgery published a consensus document to optimize the care of patients with valvular disease. This document defined two types of centers: specialized valve centers (SVCs) or level I centers, and primary valve centers (PVCs) or level II centers. It recommended treating endocarditis patients at an SVC due to their experience in managing complex valvular cases. Should endocarditis patients be referred to an SVC?

Today’s article aims to provide evidence for this recommendation. It is a retrospective, multicenter study evaluating, from 2014 to 2020, 513 patients operated on for infective endocarditis in 8 U.S. hospitals; 2 classified as SVCs and 6 as PVCs. The study compares the outcomes between these two types of centers after adjusting for propensity scores, taking into account patient characteristics, valve type, and endocarditis status (active or inactive). A multivariable logistic regression was used to identify risk factors for surgical mortality. After propensity score matching, two comparable groups were generated, with similar mean STS/Gaca endocarditis risk scores for each type of center. In cases of aortic root abscess, SVCs showed a higher likelihood of performing the Bentall procedure (60.4% vs 21.7%; p < .01). Similarly, SVCs exhibited higher mitral repair rates in cases involving the mitral valve (50.4% vs 26.3%; p < .01). Hospital mortality was significantly lower at SVCs (6.2% vs 13.0%; p = .04), and multivariable logistic regression analysis suggested that surgery at an SVC was a protective factor, with an OR = 0.39 (p = .02). These findings were consistent even when only considering patients with active endocarditis.

The authors conclude that surgery at PVCs is associated with higher mortality compared to SVCs. Therefore, transferring these patients to level I units should be considered.

COMMENTARY:

Consensus documents are valuable resources, based on current scientific evidence, that aim to standardize and optimize the care provided to patients. This was the objective of the 2019 recommendations published by the Journal of Thoracic and Cardiovascular Surgery for the treatment of valvular diseases. However, there is no formal process currently to designate a center as either SVC or PVC, nor has such designation been validated. In this study, SVCs were defined as hospitals capable of performing aortic valve-sparing surgeries (no aortic valves were spared in the endocarditis cases), aortic aneurysm procedures, mitral repair, multivalve procedures, reoperations, with a dedicated imaging team and a 24/7 ICU with intensivist coverage. Among the two SVCs, they handle an average of 32 endocarditis cases per year, compared to just 3.2 cases per year at the PVCs. In fact, one SVC reported 6,000 valvular surgeries during the study period, while the other SVC reported 1,300 cases in the same timeframe. Would both hospitals have similar outcomes in endocarditis surgery? We do not know.

We also lack data on the referral network for these hospitals. All eight hospitals belong to the same health care group, and we do not know if any barriers prevented patient transfers to more specialized centers, if patients refused transfers to stay closer to family, or if coverage issues impacted these decisions. We also do not know if specialized centers accepted referrals based on risk scores, possibly selecting lower-risk patients. Note that the Gaca score has only one variable related to the presence of endocarditis, while the remaining items evaluate prognosis in any cardiac surgery. This results in a generic score, more simplified than the STS score, which is less applicable to actual endocarditis scenarios as it does not evaluate critical factors like native versus prosthetic valve infections, presence of vegetations, or positive blood cultures.

One-year mortality rates were similar between the two types of centers. Despite better 30-day outcomes at SVCs, this advantage was lost after discharge. Therefore, the true benefit of treatment at an SVC versus a PVC remains uncertain.

There are undoubtedly differences between these two types of centers. More human and material resources, along with accumulated experience and dedicated units, improve short-term outcomes for endocarditis surgeries. Does this justify transferring all patients to these centers? Probably not; the key lesson is that extracorporeal circulation should not be performed in centers without 24/7 intensivist availability. Fortunately, such situations are rare in Spain.

Today’s article does not clarify if it is better to transfer endocarditis patients to a level I unit, as it does not specify how the consensus document’s recommendations were applied to the patients studied and does not evaluate comparable centers. This article is a prime example of comparing apples to oranges.

REFERENCE:

Squiers JJ, DiMaio JM, Banwait JK, Mack MJ, Ryan WH; Baylor Scott & White Surgery for Endocarditis Working Group. Surgical treatment of infective endocarditis at comprehensive versus primary valve centers. J Thorac Cardiovasc Surg. 2023 Aug;166(2):442-452.e6. doi: 10.1016/j.jtcvs.2021.09.023.