Nai-Hsin Chi

Background: Robotic cardiac surgery represents a transformative advancement in minimally invasive surgical techniques. However, the adoption of robotic systems involves a steep learning curve, requiring systematic standardization and technological advancements to achieve proficiency and optimal patient outcomes. Methods: This retrospective, single-center study evaluated 500 robotic cardiac surgeries performed from January 2012 to December 2024. A milestone-driven framework (technical, process, and team-based) was employed. Primary outcomes were operation time and a composite of major complications (reoperation for bleeding, stroke, renal failure requiring dialysis, reintubation, or 30-day mortality). Secondary outcomes included cross-clamp and bypass times, ICU and hospital length of stay, and New York Heart Association (NYHA) class change. We modeled learning using overall cumulative sum (CUSUM) curves for all consecutive cases with milestone timings annotated segmented regression for temporal trends. Results: The learning curve exhibited three distinct phases: initial (Cases 1–50), proficiency (Cases 51–150), and mastery (Cases >150). Operation times decreased from a median of 420 minutes in the initial phase to 280 minutes in the mastery phase. Complication rates declined from 15% to 5%. Milestone interventions, including standardized port placement, suture alignment, and the introduction of the Cor-Knot system, corresponded to inflection points in the learning curve. On the single overall CUSUM plot, inflection points temporally aligned with specific milestones. Postoperative outcomes demonstrated a low complication rate (3.5% for isolated mitral valve repair) and a 30-day mortality rate of 0.2%. Most patients (60%) achieved NYHA Class I postoperatively, indicating significant functional improvement. Conclusion: Within a structured program, temporal improvements in efficiency and safety coincided with progressive standardization and team specialization. These results demonstrate associations still provide a replicable model for the integration and optimization of robotic surgical programs, highlighting the importance of procedural standardization, team specialization, technological innovation, and iterative refinement of robotic cardiac surgery implementation.

Background: Cusp-level chordae shortening was largely abandoned after suboptimal durability in the sternotomy era. We postulated that the enhanced visualization and wristed instrumentation of totally endoscopic robotic mitral repair could enable precise shortening with durable hemodynamics and acceptable peri-operative risk. Methods: We retrospectively analyzed 523 consecutive degenerative mitral regurgitation (MR) repairs performed robotically at the National Taiwan University Hospital (January 2012–November 2024). Chordae shortening was used in 68 patients (13%; 46 primary, 22 adjunct) and compared with 455 contemporaneous repairs without shortening. Results: Baseline characteristics and operative times were similar [cross‑clamp 74±27 vs. 75±25 min; cardiopulmonary bypass (CPB) 120±30 vs. 120±25 min]. All shortening cases achieved successful repair without conversion or valve replacement. At discharge, none/trace MR was more frequent after shortening (95.6% vs. 93.4%); no patient in the shortening group had ≥ moderate MR. Early events were low: 30-day mortality 0% vs. 0.2%, stroke 0% vs. 0.2%, renal failure 1.5% vs. 1.8%, and reoperation for bleeding 1.5% vs. 0%. Reintubation (2.9% vs. 0%) and new-onset atrial fibrillation (33.8% vs. 13.6%) were more common with shortening, yet median ventilation time was comparable, and resource use favored shortening [intensive care unit (ICU) stay 19±28 vs. 27±25 h; hospital stay 7.0±2.9 vs. 9.5±2.7 days]. Follow-up echocardiography (median, 3.6 years) showed sustained competence: 92% of shortening patients remained MR-free. Ten-year freedom from ≥ moderate MR was 95.6% with shortening vs. 94.5% without (log-rank P=0.66). Conclusions: Robotic chordae shortening adds negligible operative time, shortens ICU and hospital stay, and delivers long-term valve competence equivalent to contemporary robotic repairs relying on chordal transfer or neochordae. Although new-onset atrial fibrillation was more frequent, other adverse events were rare. The robotic platform appears to overcome historical limitations of chordae shortening, restoring this tissue‑preserving option for anterior leaflet prolapse in degenerative mitral disease.

Background: Late outcomes among patients with systemic autoimmune disease (SAD) who underwent heart valve replacement are not well established. This study aimed to compare the long-term outcomes after an initial, isolated mitral valve replacement (MVR) with either a bioprosthesis or a mechanical prosthesis in patients with SAD. Materials and Methods: In this nationwide retrospective cohort study, we retrieved longitudinal data from the National Health Insurance Research Database for patients with SAD undergoing primary MVR from 1 January 2001, to 31 December 2018. Results: A total of 34 563 patients undergoing heart valve replacement for the first time between January 2001 and December 2018 were identified. After applying exclusion criteria and propensity-score matching, we included 273 bioprosthesis recipients and 273 mechanical prosthesis recipients, and the median duration of follow-up was 5.98 years and 7.44 years, respectively. There was no significant difference between the bioprosthesis group and the mechanical prosthesis group in the incidences of all-cause mortality [hazard ratio (HR): 1.26, 95% confidence interval (CI): 0.97 to 1.64] and the composite outcome of major adverse prosthesis-related events (MAPE) (HR: 0.91, 95% CI: 0.75 to 1.10). The risk of ischemic stroke was lower in the bioprosthesis group (HR: 0.60, 95% CI: 0.42 to 0.87). Subgroup analyses showed that bioprostheses were associated with a lower risk of MAPEs [10-year restricted mean survival time (RMST) difference: 1.25 years, 95% CI: 0.28 to 2.21] and major bleeding (10-year RMST difference: 1.11 years, 95% CI: 0.03 to 2.19) in male patients. Conclusion: Among patients with SAD undergoing MVR, there was no significant difference between bioprostheses and mechanical prostheses in all-cause mortality and the composite outcome of MAPEs. However, bioprostheses showed a lower risk of ischemic stroke. In male patients, bioprostheses were linked to fewer MAPEs and major bleeding.

Background: Veno‐arterial extracorporeal membrane oxygenation (VA‐ECMO) is used in cardiogenic shock, but sex‐specific outcomes remain unclear. This study investigated in‐hospital mortality differences by sex among patients receiving extracorporeal cardiopulmonary resuscitation (ECPR). Methods: We retrospectively reviewed adults with cardiogenic shock treated with VA‐ECMO at National Taiwan University Hospital between 2010 and 2021. After propensity score matching to improve comparability between groups, survival outcomes were assessed using Kaplan‐Meier estimates, and Cox proportional hazards models were used to evaluate the effect of sex on in‐hospital mortality. Results: Of the 1329 patients (average age: 57.1±15.0 years; 953 men), 670 underwent VA‐ECMO for ECPR. Women in the VA‐ECMO group exhibited a lower prevalence of out‐of‐hospital cardiac arrest (6.7% versus 10.7%, P =0.031), a lower body mass index (24.0±4.4 versus 25.0±4.3, P P =0.017) and coronary artery disease (20.9% versus 28.6%, P =0.005) after propensity score matching. No discernible sex differences were observed in the baseline characteristics of the ECPR subgroup. Kaplan‐Meier analyses showed no significant sex differences in mortality for VA‐ECMO (log‐rank P =0.1), but significant disparities were noted for ECPR (log‐rank P =0.006). In the ECPR group, female patients exhibited higher mortality rates compared with men (hazard ratio, 1.37 [95% CI, 1.09–1.72]; P =0.007), independent of Survival After Veno‐Arterial ECMO score severity. Conclusions: Women who underwent ECPR had higher in‐hospital mortality rates regardless of the severity of their Survival After Veno‐Arterial ECMO scores, despite the absence of significant sex differences in VA‐ECMO mortality. This emphasizes the necessity for sex‐based strategies in ECPR administration.

Aims: Cardiac surgery, as an intervention to repair cardiac pathology, may improve frailty, but the evidence is limited. This secondary analysis explored changes in frailty before and 6 months after cardiac surgery. Methods and results: The study included 273 adults with pre-surgery frailty phenotypes. Frailty was measured using the Fried frailty scale before and 6 months after surgery. Frailty after cardiac surgery was classified as ‘better’, ‘similar’, or ‘worse’ compared to baseline. Odds of frailty change at 6 months postoperatively (‘better or similar’ vs. ‘worse’) were calculated. After excluding 12 deaths and 49 losses to follow-up, 212 participants were analysed. At 6 months post-surgery, 92.5% either maintained or improved their frailty status, while 7.5% experienced worsening. Notably, 79.4% of the frail participants improved their status. All participants were more likely to have a ‘better or similar’ frailty status compared to a ‘worse’ status [odds: 12.25 (95% CI: 7.36–20.39)]. In a worst-case scenario analysis (n = 273), where deaths and missing frailty status were reclassified as ‘worse’ status, 71.8% showed ‘better or similar’ frailty status, whereas 28.2% had a ‘worse’ status. All participants remained more likely to have a ‘better or similar’ frailty status [odds: 2.55 (95% CI: 1.96–3.31)]. Conclusion: This study revealed that the majority of participants alive at 6 months experienced either better or similar frailty status postoperatively. The worst-case scenario analysis supported this trend. These findings suggest that cardiac surgery, when survived and tolerated, may contribute to frailty improvement.

Lateral approach in robotic aortic valve replacement: optimizing visualization from the orient and down under