Background and Methods
We here summarize publications selected from a systematic review process. Since the
selection will always be biased and affected by individual perception (despite this
standardized process), we have again focused on applying a mechanistic perspective
wherever possible and also on information most relevant for a proper risk–benefit
assessment. We hope the reader appreciates our efforts, which are (as in recent years[1]
[2]
[3]
[4]
[5]
[6]) strictly adherent to the common principle that everyone is entitled to his or her
own opinion, but not to his or her own facts. In other words, readers may not agree
with our interpretation of the data, but they can rest assure that the quoted data
and associations are accurate.
We used the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses)
approach for a systematic literature review. The MEDLINE database was searched using
the following search terms combined with the publication date being between January
1, 2023 and December 12, 2023 for the different chapters of this manuscript: Coronary
Artery Bypass Grafting; Aortic Valve; TAVI; Aortic Valve Disease; Mitral Valve Surgery;
MitraClip; TEER; Tricuspid Valve; Aortic Aneurysm; Aortic dissection; LVAD; Mechanical
Circulatory Support and Heart Transplantation; Clinical Trial. [Supplementary Fig. S1] (available in the online version) shows the PRISMA diagram for the literature review.
We selected publications based on their value for indications, decision-making, and
patient information, focusing mainly on reliable sources and publications in reputable
peer-reviewed journals. Manuscripts with a focus on individual technical details without
relevant information for the above-described goals were omitted.
Fig. 1 (A) Comparative outcomes of PCI and CABG in left main disease according to lesion site.
From De Filippo et al.[12] (B) Kaplan–Meier curves of the modified composite of death from any cause, myocardial
infarction, or stroke from the FAME 3 trial.[15] The insert shows the statistical result for the original primary endpoint. (C) Risk-adjusted survival probability in propensity-matched groups comparing CABG with
PCI in over 100,000 Medicare or Medicaid patients in the United States. From Mehaffey
et al.[17] CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention.
Surgical Treatment of Coronary Artery Disease
CABG versus PCI
Recent years have witnessed a heated debate specifically regarding the invasive therapy
of left main (LM) disease.[10] In 2023, a joint ESC/EACTS (European Society of Cardiology/European Association
for Cardio-Thoracic Surgery) review of the 2018 guideline recommendations on the invasive
therapy of LM CAD in patients at low surgical risk and anatomy suitable for percutaneous
coronary intervention (PCI) or coronary artery bypass graft (CABG) appeared.[11] The task force reviewed all new data since the 2018 guidelines, including updated
aggregated data from the four randomized controlled trials (RCTs), which suggested
no statistically significant difference in all-cause mortality at 5 years between
CABG and PCI, but significantly less spontaneous myocardial infarction (MI) with CABG.
The task force now suggests a Class I A recommendation for CABG and Class IIa recommendation
for PCI in LM disease, irrespective of the SYNTAX score, but concluded that both treatment
options are clinically reasonable based on available expertise and local operator
volumes. They provide a table of clinical conditions that a heart team should use
for individual decision-making.[11]
In this context, a meta-analysis assessing CABG and PCI for unprotected LM according
to lesion site (ostial/shaft vs. distal, 3 RCTs, 6 adjusted observational) appeared,
suggesting that PCI for distal LM lesions was associated with an increased risk of
major adverse cardiovascular event (MACE), death, and re-revascularization compared
with CABG[12] ([Fig. 1A]). Additional evidence in 2023 came from two observational nationwide registry studies
comparing outcomes after CABG or PCI in unprotected LM.[13]
[14] Such data may provide external validation to RCTs, which are often criticized for
their high grades of patient selection. The SWEDEHEART analysis assessed 11,137 patients
with significant LM stenosis of which 84% had undergone CABG and 16% PCI between 2005
and 2015.[13] The Canadian study assessed 1,299 PCI and 21,287 CABG patients between 2008 and
2020.[14] Both performed risk adjustments and both showed that CABG was associated with lower
mortality and fewer MACCE compared with PCI.[13] Of note, 30-day mortality was not higher with CABG compared with PCI.[13]
[14]
The FAME 3 trial (RCT involving patients with 3-vessel coronary artery disease [CAD]
not involving the LM in 48 centers worldwide) published its 3-year results.[15] Patients were randomly assigned to receive FFR-guided PCI using zotarolimus drug-eluting
stents or CABG. PCI did not meet the noninferiority criteria in the primary outcome
analysis, despite its huge noninferiority margin of 1.65.[2]
[16] For the 3-year results, the authors now present a new combined endpoint of death,
stroke, and MI (excluding the need for re-revascularization). For this endpoint, the
authors describe no difference and report an incidence of 9.2% for CABG versus 12%
for PCI. The hazard ratio (HR) was 1.3 [95% confidence interval [CI]: 0.98–1.83] with
p = 0.07 ([Fig. 1B]). The original primary endpoint remained in favor of CABG (incidence of 12.5% for
CABG vs. 18.6% for PCI; HR: 1.5 [95% CI: 1.2–2.0] with p = 0.002).[15]
A comparison of CABG versus PCI in patients with acute coronary syndrome (ACS) (considered
higher surgical risk) undergoing isolated CABG or multi-vessel PCI was published from
the U.S. Medicare and Medicaid Services databases.[17] The authors analyzed 104,127 patients undergoing CABG (n = 51,389) or multi-vessel PCI (n = 52,738).[17] After adjusting for confounders, they showed that CABG (compared with PCI) was associated
with lower in-hospital mortality, a significant reduction in risk-adjusted re-admission
for MI or heart failure, fewer coronary re-interventions, and improved 3-year survival
([Fig. 1C]).
Two other publications assessed the outcomes of CABG versus PCI in patients considered
at higher surgical risk, but in chronic coronary syndrome. We conducted a meta-analysis
of reconstructed time-to-event data with studies comparing CABG and PCI in octogenarians
with LM or multi-vessel disease.[18] CABG was associated with lower late mortality and lower risk of MI and repeat revascularization
compared with PCI.[18] A landmark analyses confirmed the survival advantage of CABG over PCI after 21.5
months of follow-up but in this situation it indeed suggested an advantage of PCI
over CABG in the first 30-days (and comparable survival from 1 to 21.5 months, [Fig. 2A]).[18] Another study assessed a cohort of more than 1 million female patients[19] and showed that women experienced higher operative mortality and morbidity compared
with men, which had not changed in the past decade. Hannan et al separately compared
CABG to PCI with everolimus-eluting stents in women with multi-vessel CAD from the
New York State Cardiac Registry (2012–2018).[20] The propensity-score-matched analysis included 15,589 women (60% PCI, 40% CABG).
After matching, PCI was associated with a higher 6-year risk of mortality ([Fig. 2B]), more MI, and repeat revascularization, at similar stroke rates.[20]
Fig. 2 PCI and CABG comparisons in specific risk groups. (A) Pooled Kaplan–Meier curves of all-cause mortality following CABG and PCI in patients
80 years of age and older (I), with additional 30-day survival analysis (II) and landmark
analyses from 1 to 21.5 months (III) and thereafter (IV). From Kirov et al. (licensed
under the terms of the CC-BY license).[18] (B) Kaplan–Meier comparisons of 6-year mortality in women with multivessel CAD undergoing
PCI versus CABG. From Hannan et al.[20] CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention.
Galli et al published a network meta-analysis comparing CABG, PCI, and medical therapy
in chronic coronary syndrome. The authors included 18 RCTs in patients without LM
disease or reduced left ventricular ejection fraction.[21] They report that physiology-guided PCI and CABG are associated with better outcomes
than angiography-guided PCI. Interestingly, CABG was the only treatment associated
with a significant reduction in mortality, which was associated with a reduction in
MIs.[21] A recent post-hoc analysis of the ISCHEMIA trial also reported the ability of invasive
CAD treatment to reduce spontaneous MIs. While the study did not demonstrate a mortality
reduction, it found the lowest rates of spontaneous MIs in CABG patients.[22]
Considering the many other comparisons of CABG and PCI in 2023 and those before (which
cannot all be mentioned), the same pattern always emerges. All articles illustrating
survival advantages or relevant clinical improvements with CABG always show reductions
in MI[23] by CABG. We had addressed this observation before with our surgical collateralization
hypothesis[24] and later suggested that all prognostic effects associated with invasive treatment
of chronic CAD appear to be linked to the prevention of MI.[25] This conclusion finds again support in 2023 and also from the interventional arena.
Three trials—BIOVASC,[26] FIRE,[27] and MULTISTARS[28]—report improved clinical outcomes with complete “revascularization” in patients
with multi-vessel CAD presenting with ACS and/or ST-elevation myocardial infarction
(STEMI; compared with culprit lesion PCI only). While FIRE and MULTISTARS excluded
CABG candidates, all demonstrate a significant reduction in MI rates as part of their
positive primary endpoint outcome. Finally, a Medicare analysis of patients undergoing
PCI in over 1 million patients reports less risk-adjusted mortality and MACE if intravascular
imaging was used.[29] Again the outcome improvement correlated with a reduction in MI. Considering the
superior ability of CABG over PCI to prevent future MI if the plaque burden in CAD
patients is high (with or without STEMI),[24]
[25] the need for heart-team-driven individual decision-making in CAD patients becomes
clear again. This need is further illustrated by an analysis of the New York State's
cardiac registries which showed that ad hoc PCIs occur frequently even among stable
patients with multivessel and/or LM disease.[30] Rates of ad hoc PCI are high even among patients with diabetes and low ejection
fraction and higher in hospitals without surgery on-site.[30] This finding has initiated a discussion challenging our current practice of shared
decision making.[31]
Technical Aspects of CABG
CABG is technically demanding, which was emphasized again in 2023 by several publications.
A STS (Society of Thoracic Surgeons) database analysis of almost 1.4 million CABG
patients (93,985 re-dos) identified the risk of re-do at the surgeon and not at the
center level.[32] Naito et al also demonstrate a volume–outcome relationship for off-pump CABG.[33]
To benefit from CABG, patients need patent bypass grafts. Graft selection is one of
the crucial factors for optimizing graft patency and patient outcomes. An expert review
endorsed by EACTS and STS has addressed this issue in 2023.[34] Despite optimal graft selection, graft failure remains the main problem for optimizing
CABG results, which was additionally addressed by Gaudino et al.[35] The latter authors pooled individual patient data from randomized clinical trials
with systematic CABG graft imaging to assess the incidence of graft failure and its
association with clinical risk factors.[35] Graft failure was defined at patient level as ≥1 failed grafts. The median time
to imaging was approximately 1 year.[35] Graft failure affected 16.6% of grafts in one-third of the patients[35] (9.7% of left internal thoracic arteries, 13.8% of radial arteries, 19.7% of saphenous
veins, and 23.0% of right internal thoracic artery were affected). The manuscript
illustrates different risk factors for graft failure (e.g., age, female sex, smoking).[35] From a mechanistic perspective, graft failure was associated with an increased risk
of MI or repeat revascularization and higher rates of death.
Medical Treatment of CABG Patients
Postoperative statin use was associated with less graft failure.[35] However, statin loading before CABG failed to show any benefits in an investigator-initiated,
double-blind, and placebo-controlled RCT.[36] Sandner et al additionally illustrated in a review that adding ticagrelor to aspirin
for 1 year after CABG is associated with a reduction in the risk of vein graft failure,
at the expense of an increased risk of clinically important bleeding.[37]
In conclusion, the evidence published in 2023 supports the following conclusions:
-
CABG keeps showing a prognostic advantage over PCI or medical therapy when anatomic
CAD complexity is high (specifically in LM and multi-vessel CAD).
-
This prognostic effect may even exist in patients where CABG is perceived to be at
higher risk (e.g., female, STEMI/non-STEMI or elderly patients, above 80 years).
-
Prognostic treatment effects of CABG and even PCI, if detected, were always related
to reductions in future MIs.
-
Graft patency remains crucial for a CABG treatment effect and graft failure is associated
with increased rates of MI. They may be optimized by graft selection, surgical skill,
and optimal medical therapy.
-
The high rates of ad hoc PCI reported in the reviewed evidence led to the suggestion
of low guideline adherence in patients with multi-vessel and LM disease.[30]
Surgical Treatment of Valve Disease
Aortic Valve
Aortic Valve Replacement versus Transcatheter Aortic Valve Implantation
The year 2023 saw a growing number of transcatheter aortic valve implantation (TAVI)
versus surgical aortic valve replacement (SAVR) comparisons addressing longer follow-ups,
specifically in the newer low-risk trials. The Evolut Low Risk trial presented 4-year
outcomes. There was no significant difference in mortality (SAVR vs. TAVI: 12.1% vs.
9%, p = 0.07) or stroke rates (2.9% vs. 3.8%, p = 0.32).[38] This is the first trial where an initial advantage in the first 12 months (although
not significant for mortality or stroke) was not lost during follow up. The trial
was criticized for its high amount of concomitant procedures in the SAVR group (26.3%,
with half of them CABG, but also root enlargement, mitral repairs, etc.) and the fact
that no information on the type of biological prostheses was provided for the surgical
group. In addition, the number of new permanent pacemaker implantation was significantly
higher with TAVI (24.6% vs. 9.9%, p < 0.001), but mean gradients were lower with TAVI (9.8 ± 5.5 mm Hg for TAVI vs. 12.1 ± 5.4 mm
Hg for SAVR, p < 0.001).
The other low risk trial, PARTNER 3, provided 5-year data of the Sapien 3 valve which
was compared with SAVR using a mix of different biological prostheses (63% Magna or
Magna Ease, 16% Trifecta, 6% Mosaic, 7% rapid deployment valves, etc.).[39] The trial had randomized 1,000 low-risk patients with severe, symptomatic aortic
stenosis. The primary endpoints (combinations of death, stroke, rehospitalization)
showed no difference at 5 years. However, survival curves crossed and mortality was
numerically lower with SAVR at 5 years (10% vs. 8.2%, HR: 1.23). A landmark analysis
of PARTNER 3 ([Fig. 3A]) also reconfirmed observations from PARTNER II[40] and from a previous individual patient data meta-analysis of six randomized trials
at high and intermediate risk.[41] They all demonstrate a short-term advantage for TAVI in the first 12 months, but
a significant disadvantage at the 5-year mark.
Fig. 3 Main outcomes of invasive treatment of the aortic valve in 2023. (A) Landmark analysis of the 5-year outcomes of the PARTNER 3 trial comparing TAVI (with
Sapien 3) versus SAVR. From Mack et al.[39] (B) Risk-of-bias assessment for TAVI versus SAVR trials suggesting an uneven deviation
from assigned treatment resulting in biases primarily in favor of TAVI. From Barili
et al. (licensed under the terms of the CC-BY license)[43] (C) Spline curves illustrating the time-varying hazard ratios for mortality from a meta-analysis
of randomized and matched registry studies in low-risk TAVI versus SAVR comparisons.
From Sá et al. (licensed under the terms of the CC-BY license)[44] (D) Cumulative incidence of valve thrombosis or valve-related thromboembolism in a trial
comparing apixaban with warfarin for anticoagulation after mechanical aortic valve
replacement. From Wang et al.[61] SAVR, surgical aortic valve replacement; TAVI, transcatheter aortic valve implantation.
A meta-analysis of RCTs in “lower” (combining low and intermediate risk) and high-risk
patients found an early mortality reduction with TAVR only in “lower” risk patients,
which was no longer there during further follow-up.[42] Interestingly, Barili et al meta-analytically assessed the risk of bias in TAVI
versus SAVR RCTs ([Fig. 3B]). The authors found that there were substantial deviations from the randomly assigned
treatment, patients lost to follow-up, concomitant procedures, and systematic selective
imbalance in all trials. They point out that all these factors were always in favor
of the TAVI outcomes.[43]
Sá et al[44] performed a meta-analysis of reconstructed time-to-event data from both randomized
trials and propensity-score matched (PSM) studies evaluating midterm outcomes in low-risk
patients. They identified eight studies (3 RCTs, 5 PSM studies) and included 5,444
patients (2,639 underwent TAVI and 2,805 SAVR). TAVI was noninferior to SAVR up to
2 years (HR: 1.08 [95% CI: 0.89–1.31], p = 0.448) but was associated with a significantly higher risk of all-cause mortality
at 8 years of follow-up (HR: 1.22 [95% CI: 1.03–1.43], p = 0.018) ([Fig. 3C]).
Another propensity-matched analysis of the Centers for Medicare and Medicaid data
of 6450 patients having isolated SAVR (n = 3,771) or TAVI (n = 2,679) for bicuspid aortic stenosis between 2012 and 2019 showed that TAVI was
associated with a similar mortality risk compared with SAVR within the first 6 months
but a higher mortality risk between 6 months and 3 years (HR: 2.16, 95% CI: 1.22–3.83).[45]
Thus, also in 2023, the previously emerging pattern solidified. TAVI may have an early
advantage while SAVR outperforms TAVI in terms of survival in the long run. This pattern
also seems to apply to redo operations or valve in valve situations,[46] although TAVI may be a valid option in well-selected cases.[47]
In 2023, two conventional surgical registry analyses underscored the good long-term
data with SAVR. A registry analysis of all patients with primary, isolated SAVR from
the STS database (2011–2019) reported outcomes for 42,586 patients who would have
fulfilled the inclusion/exclusion criteria of the PARTNER 3 and the Evolut Low Risk
RCTs.[48] All-cause mortality at 1, 5, and 8 years was 2.6%, 7.1%, and 12.4%, respectively,[48] which is numerically lower than the results of both arms of PARTNER 3 and Evolut
Low Risk trials (PARTNER 3 at 5 years: TAVI 10%, and SAVR 8.2%[40]; Evolut Low Risk at 3 years: TAVI 6.3% and SAVR 8.3%[49]). The same study showed that in patients with predicted risk of mortality <1%, survival
after SAVR was even higher with 95% at 8 years.[48] Similarly outstanding results were reported by Johnston et al on almost 4,000 isolated
SAVRs with STS-predicted risk of mortality <4%,[50] with survival rates of 98%, 91%, and 82% at 1, 5, and 9 years.[50]
Conventional Aortic Valve Surgery
The classic surgical literature focused on aortic valve reconstruction (valve reimplantation
vs. valve remodeling) or replacement with the pulmonary autografts (Ross procedure).
Several expert groups report their experience over decades.[51]
[52]
[53] They all describe excellent long-term outcomes for valve repair/reimplantation associated
with low rates of re-intervention and valve-related complications (VRCs). The only
replacement option demonstrating similar long-term outcomes with low rates of re-intervention
and VRC appears to be the Ross procedure (in dedicated centers with high expertise).
A prominent study by El-Hamamsy et al[54] was confirmed by a recent meta-analysis[55] supporting the statements in 2023. Re-intervention appears to be required in approximately
20% of patients at 20 years for either autograft or right ventricular outflow tract
issues.[55] If root replacement has to be performed, a meta-analysis of 23 studies in over 11,000
patients suggests that mechanical conduits may outperform biological conduits.[56] Perioperative risk was similar between mechanical and biological conduits, but patients
with mechanical solutions had lower mortality after 30 days, which was associated
with lower rates of re-intervention.[56] However, previous comparisons[57] between mechanical and biological valves have reported an increased stroke rate
after mechanical aortic valve replacement.
Medical Therapy in Valve Surgery
One of the VRCs that may be addressed medically is valve thrombosis. Thrombus formation
mainly affects TAVI with up to 40% of valves.[58] Although there appears to be no direct relationship to clinical events,[58] previous evidence suggests that long-term durability may be impaired by leaflet
thrombus.[59] The Mayo Clinic reported data on temporary anticoagulation after biological SAVR
in over 10,000 patients.[60] They found that guideline-directed anticoagulation was poorly performed and that
lack of anticoagulation in the first 3 months after SAVR was associated with more
events and more deaths but less bleeding complications.
A prospective, randomized, open-label trial assessed the value of apixaban versus
warfarin in patients with On-X mechanical aortic valves.[61] The trial was stopped after 863 participants were enrolled owing to an excess of
thromboembolic events in the apixaban group[61] ([Fig. 3D]).
Finally, a Swedish study assessed medical therapy in patients after aortic valve replacement
for regurgitation.[62] Over 2,200 patients underwent the procedure between 2006 and 2017 in Sweden. The
authors identified protective effects of renin–angiotensin system (RAS) inhibitors
and statins on their combined endpoint of mortality, MI, and stroke.[62]
Mitral Valve
Although current guidelines recommend transcatheter approaches to the mitral and the
tricuspid valve only in patients who are no candidates for surgery, the number of
publications addressing the impact of transcatheter options alone continuously increases.
This was also the case in 2023.
The randomized COAPT trial reported its 5-year outcomes comparing transcatheter edge-to-edge
repair (TEER) for severe functional mitral regurgitation to guideline-directed medical
therapy alone.[63] The lower rate of hospitalizations for heart failure and lower all-cause mortality
with TEER remained, with those patients having the least degrees of mitral regurgitation
doing best.[63] Since there was no surgical arm in this trial, a direct comparison of TEER and surgery
is still missing, but results of the MATTERHORN trial (NCT02371512) are on the horizon.
For structural (degenerative) MR, two trials are currently active in comparing TEER
and surgical repair (the Abbott sponsored REPAIR-MR-trial [NCT04198870] and the CTSN-sponsored
PRIMARY-trial [NCT05051033]). In the meantime, significant fractions of structural
MR patients are already treated with TEER in the absence of randomized evidence. Makkar
et al[47] reported on 20,000 patients who received TEER for structural MR. The study concludes
that TEER is safe (mortality 2.7%, O/E ratio 0.6) with high procedural success, which
is in most TEER studies defined as moderate or less MR after the procedure. The authors
report severe MR at discharge in 11% of patients.[47] However, only half of all patients left the hospital with a surgically acceptable
result (i.e., MR ≤ 1), although chances for MR ≤1 increased over time ([Fig. 4A]). An important observation in this context is the finding that the lowest 1-year
mortality was observed in patients who had both residual MR ≤1 and mean mitral gradients
of 5 mm Hg or less (11.4%) versus those with an unsuccessful procedure (26.7%).[47]
Fig. 4 Repair success of invasive treatment of structural mitral regurgitation (MR) in 2023.
(A) Proportion of patients with MR ≤1 at discharge from the hospital after TEER in 19,088
structural MR patients in the United States between 2014 and 2022. Graph contructed
from data from Makkar et al.[47] (B) Proportion of patients with MR ≤1 at 12 weeks and 1 year after surgical repair of
structural MR performed either via mini-thoracotomy or via sternotomy in the UK-Mini-Mitral
trial. Data from Akowuah et al.[70] TEER, transcatheter edge-to-edge repair.
This relationship between repair success and 1-year mortality (and rehospitalizations)
was also found by others in 2023, for instance in the Prospective German Multicenter
Registry (MITRA-PRO).[64] Biasco et al suggested sustained mitral regurgitation reduction to be key to survival
improvement even in patients 80 to 89 years of age.[65] Watt et al finally report the same finding in surgical patients with severe functional
MR. Reductions of postoperative MR grade resulted in an incrementally lower risk of
death or reoperation during follow-up.[66]
Considering this relationship, it appears only logical to ask for a successful and
durable elimination of MR. A meta-analysis that reviewed six studies comparing PASCAL
or MitraClip TEER devices quantified the real-world outcomes after TEER in 1,581 patients.
Residual MR ≤1 at discharge was present in 73.1% and 63.8% for PASCAL and MitraClip,
respectively. These values decreased at midterm follow-up to 71.3% and 56.2%, respectively.[67] These outcomes together with data from [Fig. 4A] suggest that more than one-third of patients after current interventional therapy
sustain at least moderate MR after the procedure. Also in 2023, Badhwar et al[68] reported that surgery can generate repair rates for structural MR of over 95% of
cases at very low surgical risk (<1%) and an editorial emphasized the role of surgery
as gold standard for patients with structural MR.[69]
One of the greatest arguments mentioned against surgery is always the need for median
sternotomy, which set in motion significant activity to develop minimally invasive
access to the heart for all procedures. The mini-mitral trial was a multicenter RCT
comparing 166 mini-thoracotomy with 164 sternotomy patients undergoing isolated mitral
valve repair performed by expert mitral valve surgeons in the United Kingdom.[70] The primary outcome was physical functioning and associated return to usual activities
measured by change from baseline in the SF-36 version 2 physical functioning scale
12 weeks after the index surgery, assessed by an independent researcher blinded to
the intervention.[70] While there was an advantage for the minimally invasive approach after 6 weeks,
the primary outcome was not different at 12 weeks. Considering the above addressed
relationship of repair quality and survival, it may be worth noting that rates and
quality of valve repair in the mini-mitral trial were high at 1 year in both groups[70] ([Fig. 4B]).
While the mini-mitral trial did not further specify the technical details of the repair,
we performed a meta-analysis of reconstructed time-to-event data comparing the resect
and respect approaches for posterior prolapse mitral valve repair. We found no difference
in long-term mortality, MR recurrence, or reoperation between the two techniques after
adjusting for patient risk factors.[71]
Tricuspid Valve
The current guidelines state that tricuspid valve interventions are underused and
often initiated too late.[72] Surgical treatment (preferably repair) is indeed recommended as first-line therapy
in patients with severe tricuspid regurgitation (TR) even if symptoms are low or absent
(see guidelines for details). Interventional treatment is only recommended for inoperable
patients. Despite this, a sudden increase of TR cases during the last years has been
observed.[73] Reviewing the literature in 2023, the same impression to that for the mitral valve
emerges, as interventional results are increasingly reported with sole comparisons
to medical therapy.
For instance, the TRILUMINATE trial[74] compared medical therapy with transcatheter tricuspid valve edge-to-edge repair
in patients with isolated moderate to high surgical risk and severe TR. The trial
reported an improvement in quality of life in the interventional arm, which correlated
with a significant reduction of TR grade. However, there was no difference in any
of the clinically quantifiable parameters (outcomes of the 6-minute walk test even
decreased), and mild residual regurgitation was only achieved in half of the patients.
Dreyfus et al used their TRI-SCORE to assess outcomes from a multi-center registry
of 2,413 patients with isolated TR.[75] They found an improvement in survival associated with successful reduction in TR
in patients with mild and moderate risk. However, if repair success was absent with
the intervention or if the score reflected the highest risk, long-term mortality was
unaffected. [Fig. 5] illustrates that surgical risk at low TRI-SCORE was equal to interventional risk
(panels C and D), but 2-year outcome was better. Periprocedural mortality risk increased
most in the surgical group with increasing TRI-SCORE ([Fig. 5C]), suggesting that increased invasive risk consumes potential long-term benefits
specifically at the highest risk group. The community often considers isolated tricuspid
surgery a high-risk procedure. We addressed this potentially erroneous conclusion
by illustrating the influence of an immortal time bias on surgical outcomes.[76] If patients die after a diagnosis has been made and while waiting for surgery, this
mortality is not considered in surgical outcomes (hence, immortal time bias). This
bias may decrease the surgical effect if medical therapy is the comparator. However,
in case of tricuspid surgery (and other conditions), waiting for surgery increases
operative risk, which may offset the immortal time bias, suggesting earlier surgery
in these cases.
Fig. 5 (A–D) Survival rate according to treatment modality and TRI-SCORE category from a multi-center
registry of 2,413 patients with isolated severe tricuspid regurgitation. From Dreyfus
et al.[75]
The main findings in 2023 for classic valve surgery are:
-
The evidence for the treatment of aortic valve stenosis confirms the outstanding long-term
results for conventional surgery. Only the Evolut Low Risk trial preserved an early
TAVI advantage over time. These facts are not reflected in the guidelines.
-
The best long-term outcomes (normalization of life expectancy) can be achieved with
valve repairs or the Ross procedure, when these techniques are feasible.
-
When treating mitral regurgitation or TR, a durable, high-quality repair is associated
with the best long-term outcomes. This applies to both surgical and interventional
techniques.
-
The quality and durability of a repair is surgically higher than interventionally,
but surgical risk may be higher.
Surgery for Terminal Heart Failure
For the surgical treatment of heart failure, important publications in 2023 assessed
the possibilities of expanding the donor pool and the role of mechanical circulatory
support.
A randomized, noninferiority trial compared results of transplanting hearts from brain-death
donors (BDDs) with traditional cold storage with those from circulatory-death donors
(CDDs) that were resuscitated and preserved in a mobile extracorporeal perfusion and
preservation system (Organ Care System Heart, TransMedics). The primary end point
was risk-adjusted survival at 6 months. A total of 166 transplant recipients were
included in the as-treated primary analysis (80 CDDs and 86 BDDs). CDD was noninferior
to BDD (94% vs. 90% 6-month survival).[77]
After the first xenotransplantation with genetically modified pig hearts[78] in 2022, the results of two transplantations using gene-edited pig hearts transplanted
into brain-dead human recipients were published in 2023.[79] Both xenografts demonstrated excellent cardiac function after early after transplantation
and continued to function for the duration of the observation. Cardiac function in
one of the hearts declined postoperatively, attributed to a size mismatch. There was
no evidence of cellular or antibody-mediated rejection, as assessed using histology,
flow cytometry, and a cytotoxic crossmatch assay. Importantly, there was no evidence
of zoonotic transmission from the donor pigs to the human recipients. The second person
to receive a transplanted pig heart lived for 6 weeks, but beyond press releases,
there was no scientific publication in 2023.
Another approach to address the problem of growing patient numbers with terminal heart
failure and organ shortage is reducing the need for transplantation by improving the
exploitation of classic therapeutic option. The team in Bad Oeynhausen prominently
published the CASTLE HTx trial,[80] in which 194 terminal heart failure patients with symptomatic atrial fibrillation
referred for heart transplantation were randomized to either catheter ablation or
control. They reported a significant reduction in the primary endpoint of death, left
ventricular assist device (LVAD) implantation, or urgent heart transplantation after
18 months in the ablation group (HR: 0.24, CI: 0.11–0.52).
Finally, in 628 LVAD patients, a randomized study of aspirin (100 mg/d) versus placebo
in addition to vitamin K antagonists was conducted in 51 centers from 9 countries.[81] Aspirin avoidance was associated with reduced nonsurgical bleeding (RR: 0.66 [95%
CI: 0.51–0.85]; p = 0.002) without an increase in strokes or other thromboembolic events.
Surgery of the Aorta
In 2023, the size of the aorta and its relationship to aortic dissection was the main
focus of publications. An analysis of 964 unoperated thoracic aortic aneurysms, followed
over a median of 7.9 years, showed that the risk of adverse aortic events was relatively
flat until 5 cm of ascending aortic size, at which it began to increase rapidly[82] ([Fig. 6]). The mean annual growth rate was estimated to be 0.1 cm/year. The authors conclude
that an aortic size of 5 cm, rather than 5.5 cm, may be a more appropriate intervention
criterion for prophylactic surgery.[82]
Fig. 6 Relationship of the size of the ascending aorta with adverse aortic event (AAE) risk.
Note the rapid increase (p for nonlinearity <0.001), when the diameter exceeds 5 cm. From Wu et al.[82]
An analysis on 407 naturally occurring, acute, flap-type ascending or descending aortic
dissections between 1990 and 2022 suggested that a “left shift” to 5.0 cm in the criteria
for ascending aortic intervention in the American guidelines could prevent an additional
29.3% of type A dissections.[83] They also demonstrate that type B dissections generally occur at smaller aortic
diameters and would not benefit from alterations in size criteria.[83]
A publication comparing the perioperative outcomes of patients with acute type A aortic
dissection undergoing hemiarch versus extended arch repair in nine centers (923 patients)
showed that extended arch interventions pose similar perioperative mortality and neurologic
risks to hemiarch.[84]
