Valvular heart diseases remain a major global health burden. In industrialized countries, they are primarily of degenerative origin, whereas rheumatic heart disease continues to be the predominant cause in developing nations. Among these conditions, mitral valve disease is moderately prevalent [1]. Epidemiological studies estimate that mitral regurgitation affects approximately 2–3% of the general population, with its incidence increasing substantially with age. Mitral stenosis, although less common in high-income regions due to declining rheumatic fever rates, still represents a significant concern in low- and middle-income countries [2, 3].

Conventional mitral valve repair is most commonly performed through a median sternotomy, which offers direct and extensive exposure of the cardiac structures and facilitates a wide range of myocardial protection strategies. However, the invasive nature of this approach has been associated with prolonged recovery times, particularly in terms of return to baseline physical function, and a higher incidence of postoperative complications compared to less invasive alternatives [4, 5].

Since its initial description by Carpentier in 1996, minimally invasive mitral valve surgery (MIMVS) has undergone significant evolution and refinement. Contemporary evidence, including propensity-matched analyses and meta-analyses, demonstrates that MIMVS offers equivalent valve repair rates and durable long-term outcomes when compared to conventional sternotomy [6]. Moreover, it is consistently associated with enhanced postoperative recovery and reduced rehabilitation times [7, 8]. This approach is particularly appealing to patients due to perceived benefits such as faster physical recovery, reduced postoperative discomfort, and superior cosmetic outcomes [9-11]. Advocates within the surgical community emphasize additional advantages, including shorter hospital stays, reduced incidence of perioperative complications, and potentially lower overall healthcare costs [12, 13]. Given that a substantial proportion of individuals undergoing mitral valve interventions are elderly and frequently present with multiple cardiovascular risk factors and comorbidities, the adoption of less invasive strategies holds particular clinical significance [14].

The adoption of minithoracotomy varies significantly across regions, with relatively low utilization in the United States and the United Kingdom, and considerably higher uptake in Germany. In Germany, approximately 55% of patients undergoing mitral valve surgery receive treatment via a minimally invasive surgical (MIS) approach [11]. This variation largely reflects the current lack of robust evidence from randomized clinical trials (RCTs) confirming outcomes equivalent to or better than those of traditional sternotomy. Additionally, concerns persist regarding the technical complexity of minithoracotomy, particularly its potential impact on the successful repair of complex valve lesions and the risk of perioperative complications, such as vascular injury and stroke [15, 16]. Recent guidelines and expert consensus statements have emphasized the need for a high-quality RCT to resolve these uncertainties [17]. Nevertheless, it remains important to collect data from high-experience centers, as these can provide valuable insights regarding the different types of surgical procedures.

The purpose of our investigation, therefore, was to compare the different surgical techniques (conventional and minimally invasive) mitral valve replacement (MVR) and mitral valve repair (MVr) with single administration of histidine-tryptophan-ketoglutarate solution (Custodiol®) in relation to myocardial protection. The primary endpoint of the study is the peak creatine kinase-MB (CK-MB) level, measured 24 hours postoperatively, as an indicator of myocardial injury and, consequently, the efficacy of Custodiol®. Secondary endpoints include operative time, aortic cross-clamp duration, cardiopulmonary bypass time, length of stay in the intensive care unit (ICU) and hospital, incidence of postoperative cardiac arrhythmias, postoperative bleeding, requirement for blood transfusions, rethoracotomy rates, and in-hospital mortality.

Ethical Statement Given the retrospective nature and pseudonymization of the data, ethical review and approval were waived for this study. Patient Population This retrospective study analyzed 527 adult patients who underwent isolated mitral valve repair (Mr) or mitral valve replacement (MR) at the Heidelberg University Hospital between January 2014, and December 2019. Patients were stratified based on the type of surgical procedure and the surgical approach into four principal groups: conventional mitral valve replacement (MRV_C; n = 176), minimally invasive mitral valve replacement (MRV_MI; n = 63), conventional mitral valve repair (MVr_c; n = 64), and minimally invasive mitral valve repair (MVr_MI; n = 212). Additionally, two smaller subgroups were identified and analyzed separately due to their limited sample size: the conversion group for mitral valve replacement (n = 3) and the conversion group for mitral valve repair (n = 9). These conversion groups included patients initially scheduled for minimally invasive surgery who required intraoperative conversion to a conventional median sternotomy due to unforeseen complications. A further subgroup analysis was conducted among patients undergoing mitral valve repair, subdividing them according to the reconstructive technique performed: ring annuloplasty alone (n = 23), leaflet repair (n = 101), chordae repair (n = 114), and combined chordae and leaflet repair (n = 47). In all cases, annuloplasty with the implantation of a prosthetic annuloplasty ring was performed. In the ring annuloplasty group, no additional reparative measures were undertaken, whereas in the remaining three groups, supplementary reconstructive procedures were carried out alongside annuloplasty. Data Collection All data were obtained from the hospital information system i.s.h.med® (SAP) of Heidelberg University Hospital following institutional approval. Parameters included demographic data (age, sex), preoperative clinical status (primary diagnosis, comorbidities, NYHA and ASA classification), intraoperative data (cross-clamp time, cardiopulmonary bypass time, duration of the operation, Custodiol® cardioplegia volume), and postoperative outcomes (CK-MB levels, length of hospital and ICU stay, rethoracotomy, bleeding, and in-hospital mortality). Additional comparisons were made regarding AV blocks, atrial fibrillation, and transfusion requirements. Statistical Analysis Data were analyzed using IBM® SPSS® Statistics and Microsoft® Excel®. Categorical variables were evaluated using Pearson’s Chi-square test, with Bonferroni-corrected post hoc testing for subgroup comparisons. Continuous variables were assessed using one-way ANOVA with Fisher's LSD post hoc testing. A significance level of α = 0.05 was applied throughout the analyses.

Demographic Data

In total, 579 patients were analyzed, divided into four principal groups: conventional mitral valve replacement (MVR_C, n=176), minimally invasive mitral valve replacement (MVR_MI, n=63), conventional mitral valve repair (MVr_C, n=64), and minimally invasive mitral valve repair (MVr_MI, n=212). The patient age ranged widely (18–89 years) and was similar across groups. The largest age category was 61–80 years (about 52–60% of patients in each group), followed by 41–60 years (~30–43%)​. Very few patients were under 40 or over 80 years old (each <7%). The mean age at surgery did not differ significantly between groups (e.g. ~61–65 years on average, p = 0.06) (Fig. 1.)​. In terms of sex, all groups had a male majority (52.8% to 77.8% male)​. The conventional surgery groups had ~40–47% female patients, whereas the minimally invasive repair group had only ~22% females. This difference in sex distribution was statistically significant (p = 0.001)​ (Fig.1.). Overall, the minimally invasive repair cohort contained a disproportionately higher number of male patients compared to the other groups, but otherwise the cohorts were comparable in basic demographics. The demographic data are summarized in Table 1.

Fig.1. Age on the day of the operation. MVR_C: conventional mitral valve replacement; MVR_MI: minimally invasive mitral valve replacement; MVr_C: conventional mitral valve repair; MVr_MI: minimally invasive mitral valve repair.

Table 1. Patient characteristics – Demographics. MVR_C: conventional mitral valve replacement; MVR_MI: minimally invasive mitral valve replacement; MVr_C: conventional mitral valve repair; MVr_MI: minimally invasive mitral valve repair. Bolded p-values indicate statistical significance of the performed tests. Values marked with an asterisk (*) in the table are responsible for the statistical significance of the respective comparisons.

Preoperative Data

The vast majority of patients were operated for mitral regurgitation (MR). In the repair groups (MVr_C and MVr_MI), over 95% had degenerative MR as the primary diagnosis. In the replacement groups, MR was also the most common diagnosis (42% of MVR_C and 69.8% of MVR_MI), followed by infective endocarditis in a substantial subset (36.4% and 23.8%, respectively)​.

Mitral stenosis (often combined with regurgitation) was only seen in the replacement groups (15.3% of MVR_C and 3.2% of MVR_MI), and a small number of patients had other etiologies. The distributionof primary pathologies differed significantly among the four groups (p = 0.001) (Table 2.)​, reflecting that patients with endocarditis or mitral stenosis more often underwent valve replacement rather than repair.

Across all patients, most (64.2%–99.1%) did not have active endocarditis at the time of surgery​.  However, endocarditis cases clustered in the conventional replacement group: 35.8% of MVR_C patients had endocarditis, compared to 20.6% in MVR_MI, only 4.7% in MVr_C, and 0.9% in MVr_MI. This difference was statistically significant (p = 0.001) (Table 2.). Essentially, almost all patients in the repair groups were non-endocarditic, whereas a large minority of replacement patients (especially via sternotomy) had endocarditis, explaining the uneven distribution.

Patients’ preoperative status varied. Most were in NYHA functional class II or III heart failure (NYHA II 18.7%–48.6%; NYHA III 41%–64.2%), with relatively few in class I (4%–9.9%) and only small fractions in class IV (NYHA IVa 0%–8%; IVb 0%–5.1%). There was a significant difference in NYHA class distribution between groups (p = 0.001) (Table 2)​. Similarly, general health status by ASA classification showed most patients were ASA 3 (64.8%–85.9% of each group) and the remainder mostly ASA 4 (9.5%–31.2%)​. This too differed significantly among groups (p = 0.001) (Table 2.)​, suggesting that patients selected for replacement (especially conventional) tended to have higher ASA risk profiles than those for repair. In terms of family history, 70%–76% of patients had no known familial predisposition to cardiac disease, ~11%–16% did have a family history, and ~8%–19% had unknown history. There was no significant difference betweengroups in family history (p = 0.471)​ (Table 2.).

Table 2. Patient characteristics – Risk factors and anamnestic data (Part 1). MVR_C: conventional mitral valve replacement; MVR_MI: minimally invasive mitral valve replacement; MVr_C: conventional mitral valve repair; MVr_MI: minimally invasive mitral valve repair. Bolded p-values indicate statistical significance of the performed tests. Values marked with an asterisk (*) in the table are responsible for the statistical significance of the respective comparisons. MI: mitral insufficiency, MS: mitral stenosis, MV: mitral vitium, NYHA: The New York Heart Association (NYHA) classification of heart failure, ASA: The ASA Classification (American Society of Anesthesiologists Physical Status Classification System), D: diet, OAD: oral antidiabetics, I: insulin

Intraoperative Data

All patients received Bretschneider’s HTK cardioplegia (Custodiol®) for myocardial protection during the operation​. The administered cardioplegic volume differed by surgical approach. On average, minimally invasive cases required a greater volume of Custodiol solution than conventional cases. For example, the average cardioplegic solution consumption during minimally invasive replacement was 2207 ± 623 mL compared to 1874 ± 550 mL in conventional replacement; similarly, minimally invasive repair involved a consumption of approximately 2139 ± 546 mL versus 1787 ± 337 mL in conventional repair. These differences were statistically significant (p = 0.001)​. In fact, pairwise post-hoc tests showed that within each surgery type the difference by approach (minimally invasive vs sternotomy) was significant, while differences between the two replacement groups or between the two repair groups were not significant​. This indicates the minimally invasive approach necessitated a higher cardioplegia volume (likely due to longer ischemic times), whereas the cardioplegia usage was similar for replacement vs repair when the approach was the same. Notably, there were a few cases (5.1% of MVR_C, 9.5% of MVR_MI) where the exact Custodiol volume was not recorded.

Minimally invasive procedures took substantially longer than conventional open procedures. The total operation time was significantly different across all groups (p = 0.001)​. On average, a minimally invasive valve replacement lasted about 301 ± 83 minutes, versus 233 ± 71 minutes for a conventional replacement. Similarly, a minimally invasive repair took about 277 ± 52 minutes vs 208 ± 44 minutes for a conventional repair. In other words, the minimally invasive approach added roughly 30–45% more time to the surgery in both replacement and repair. Post-hoc analysis confirmed that all pairwise comparisons of operation time between any two groups were significant​, reflecting both the approach-related delay and the difference between the typically more complex replacement versus the usually faster repair procedures. The aortic cross-clamp time followed a similar pattern: minimal-access cases had much longer cross-clamp durations. Both the minimal invasive replacement and repair groups had an average aortic clamp time of ~107 minutes, compared to ~77 minutes in conventional replacement and ~76 minutes in conventional repair. All minimal versus conventional differences in clamp time were significant (p = 0.001), whereas within the same approach (e.g. comparing the two minimal groups) there was no significant difference in clamp time​. The cardiopulmonary bypass (CPB) time was likewise prolonged with minimally invasive access. Mean CPB times were 191 ± 71 min (MVR_MI) vs 129 ± 46 min (MVR_C), and 173 ± 43 min (MVr_MI) vs 116 ± 29 min (MVrt_c)​. These differences were highly significant overall (p = 0.001)​. In pairwise comparisons, bypass times did not differ by approach within the same procedure (e.g. conventional vs minimal repair, p = 0.06) but did differ between replacement and repair groups (replacement surgeries had longer CPB times than repairs)​.

In summary, the minimally invasive techniques required significantly longer cross-clamp and pump times and greater cardioplegic volume, but these intraoperative disadvantages are expected given the smaller access and more complex cannulation in minimal-access surgery. Importantly, all procedures were completed successfully with the chosen myocardial protection strategy, and no intraoperative conversions to full sternotomy were noted in the dataset.

Postoperative Data

Myocardial injury was assessed via the peak creatine kinase-MB measured within 24 hours post-op. When considering all patients, the highest CK-MB peaks were observed in the minimally invasive replacement group. The mean peak CK-MB in MVR_MI was 132 ± 154 U/L, compared to 83 ± 53 U/L (MVR_C), 81 ± 53 U/L (MVr_C), and 76 ± 56 U/L (MVr_MI)​. This difference among groups was significant (p = 0.001)​. Post-hoc analysis revealed that the minimal-invasive replacement group had a significantly higher peak CK-MB than any other group (p = 0.001 for all such comparisons). In contrast, none of the differences between the other three groups reached significance. To ensure this was not simply due to the disproportionate number of endocarditis cases in the conventional replacement group (which might confound enzyme release), a secondary analysis excluding all endocarditis patients was performed. After exclusion of endocarditis, the peak CK-MB values remained highest in MVR_MI (127 ± 157 U/L) versus MVR_C (89 ± 57), MVr_C (82 ± 54), and MVr_MI (76 ± 56). The overall group difference remained significant (p = 0.001)​. In this endocarditis-excluded analysis, the minimally invasive replacement still showed significantly higher CK-MB release than each of the other groups (p = 0.001–0.003)​. No other group differences were significant​. Thus, minimally invasive mitral replacement was associated with a higher postoperative CK-MB peak, suggesting greater myocardial stress/injury in that subgroup, whereas mitral repair (whether minimal or conventional) showed the lowest CK-MB levels. Notably, this enzyme release did not correlate with worse clinical outcomes.

Hospital and intensive care unit (ICU) stays differed primarily by procedure type (replacement vs repair). Patients who underwent mitral valve repair had shorter stays on average than those who had valve replacements. The average total hospital length of stay was ~14 ± 6–9 days for repair patients versus ~17 ± 11 to 18 ± 15 days for replacement patients​. Statistically, the four-group comparison was significant (p = 0.001)​. Post-hoc tests showed that conventional replacement patients stayed significantly longer than conventional repair patients (mean 18 vs 14 days, p = 0.004) and also longer than minimally invasive repair patients (p = 0.001)​. By contrast, there was no significant difference between the two replacement groups (18 vs 17 days, p = 0.399) or between the two repair groups (both ~14 days, p = 0.68)​. This indicates that the necessity of a valve replacement (often reflecting more advanced disease) prolongs recovery time relative to a repair, whereas the surgical approach (sternotomy vs minithoracotomy) did not significantly impact hospital discharge timing in this cohort. A similar pattern was seen for ICU stay: replacement patients required longer intensive care (approximately 3–4 ± 5–7 days) than repair patients (≈2 ± 3–4 days)​. ICU length was significantly different overall (p = 0.001)​; post-hoc analysis found conventional replacements had longer ICU stays than conventional repairs (p = 0.023) and than minimally invasive repairs (p = 0.001)​. No significant ICU stay differences were noted between minimal vs conventional approaches when the procedure type was the same (e.g. 4 ± 7 vs 3 ± 5 days for replacement, p = 0.322)​.

Postoperative complications were infrequent. Major bleeding requiring reoperation (re-thoracotomy for hemorrhage) was rare. Clinical rebleeding (requiring intervention) occurred in only about 1.6%–6.7% of patients (lowest in MVr_C, highest in MVr_MI), and this rate did not differ significantly between groups (p = 0.521)​. Correspondingly, re-thoracotomy for any cause was uncommon: 4% of patients in MVR_C and 0.9% in MVr_MI underwent a repeat thoracotomy, whereas none of the patients in MVR_MI or MVr_C needed re-thoracotomy. The difference among groups reached statistical significance (p = 0.043)​, driven by the small number of reoperations in conventional replacements. In other words, only the conventional replacement group saw a few reoperations for bleeding, while the other groups had virtually none. In-hospital mortality was low across all cohorts. There were no deaths in three of the four groups (0% mortality in MVR_MI, MVr_C, and MVr_MI). In the conventional replacement group (MVR_C), 4.5% of patients died during the index hospitalization. This yielded a significant difference in raw mortality between groups (p = 0.001 by chi-square)​. It should be noted that the MVR_C group contained the highest-risk patients (older, more endocarditis cases, etc.), which likely contributed to this mortality, rather than the surgical approach itself. No deaths occurred in the minimally invasive surgery groups, underscoring that a less invasive approach can be conducted safely in properly selected patients.

The need for blood transfusion was analyzed intraoperatively and postoperatively. Intraoperatively, 18%–68.8% of patients (depending on group) received donor blood, while 31.2%–82% received no transfusion in the operating room (OR). The highest intraoperative transfusion rate was in the conventional replacement group (68.8% of MVR_C patients needed blood)​. In contrast, in each of the other groups a majority of patients (>58%) did not require blood during surgery. This difference was significant (p = 0.001). The volume of blood used intraoperatively also differed: conventional replacements transfused the most (on average 1020 ± 550 mL), followed by minimally invasive replacements (942 ± 566 mL). Repair patients required much less blood (about 596 ± 301 mL in conventional repair and 684 ± 376 mL in minimally invasive repair). These volume differences mirror the fact that valve replacements (especially via sternotomy) are larger surgeries with greater blood loss. Statistically, intraoperative blood loss/transfusion volume differed by group (p = 0.001); pairwise tests showed no significant difference by approach within the replacements or within the repairs (e.g. MVR_C vs MVR_MI, p = 0.475) but significant differences when comparing replacements vs repairs. Postoperatively (in the ICU and ward), 20.3%–54% of patients required blood transfusion, while 46%–79.7% needed none. Again, the conventional replacement group had the highest proportion of patients needing post-op blood (54%), whereas in the other groups the majority (>64%) had no transfusion post-op. This difference among groups was significant (p = 0.001)​. However, the total volume of blood given postoperatively did not differ significantly (p = 0.568)​, with large standard deviations in all groups (mean ~1.0–1.5 L packed cells for those who needed transfusion). In sum, patients undergoing mitral valve replacement were more likely to require blood transfusions (intra- and postoperatively) than those undergoing repairs. The surgical approach (mini vs sternotomy) had a much smaller impact on transfusion needs than the type of procedure.

Preoperatively, only a small minority of patients had any degree of atrioventricular block (AV-block). 90%–98% had no AV-block, ~1%–5% had first-degree AV-block, and <5% had third-degree AV-block; second-degree block was almost nonexistent (≤0.6%) in our cohort. There was no statistically significant difference between groups in preoperative AV-block incidence (p = 0.055). Postoperatively, new high-grade AV-blocks were rare as well. 94%–97% of patients in each group had no AV-block after surgery, and the rates of first-, second-, or third-degree AV-block post-op were all very low (each ranging 0%–4.8%)​. The distribution of AV-block severity post-op did not differ significantly between groups (p = 0.12).

Regarding atrial fibrillation (AF), many patients had a history of atrial fibrillation pre-surgery. Preoperatively, 54%–68% of patients were in normal sinus rhythm (meaning 32%–46% had some form of AF: intermittent, persistent, or chronic)​. Postoperatively, the proportion of patients in sinus rhythm actually increased in all groups – by discharge, 81.8%–91.5% of patients had no atrial fibrillation​. Correspondingly, the overall prevalence of AF (any type) decreased after surgery (only ~8%–18% of patients had AF episodes post-op, mostly transient). This improvement reflects that some patients with pre-op AF reverted to sinus rhythm after correction of the mitral pathology, although new-onset AF in the early postoperative period may also have occurred in others. The net change was an increase in sinus rhythm patients. When comparing the four groups, there were no significant differences in the rates of postoperative AF or sinus rhythm restoration (p = 0.077). That is to say, the minimally invasive vs conventional approach did not significantly affect the likelihood of postoperative atrial fibrillation; any atrial arrhythmias occurred at similar rates across groups.

In this study, minimally invasive mitral valve surgery was evaluated against conventional sternotomy approaches, and the findings were placed in context of international literature. Overall, the results demonstrate that both minimally invasive and conventional mitral operations are safe and effective, with each approach offering distinct advantages. The choice of technique influences operative parameters but not the ultimate success in terms of patient recovery and outcomes.

Prior large studies have established that minimally invasive mitral surgery (typically via right minithoracotomy) is a safe alternative to median sternotomy, even in older patients. Our findings concur with this. We observed no excess mortality or major morbidity in the minimally invasive groups – in fact, there were zero in-hospital deaths in those cohorts, underlining the safety of the approach. This aligns with the work of Holzhey and colleagues [18], who reported that a minimally invasive approach in patients over 70 had outcomes equivalent to sternotomy. Minimally invasive techniques offer well-known patient benefits: a smaller incision leading to less trauma, improved cosmetic results, and a lower risk of sternal wound infection. Gammie et al. similarly found fewer wound complications with minimally invasive mitral surgery [19]. In our study, although we did not specifically track wound infections, the absence of any sternal incision in the minithoracotomy groups would inherently eliminate sternal wound issues. Patients in the minimally invasive groups also had no instances of groin complications related to femoral cannulation in our series – a complication noted in earlier literature (e.g. Mohr et al. 1998) as a rare but serious risk (aortic dissection from retrograde perfusion) [10]. Careful cannulation techniques likely mitigated that risk in our cohort.

The trade-off for a less invasive incision is longer cardiopulmonary bypass and aortic cross-clamp times, as was evident in our results. We found that minimally invasive cases took significantly longer (by ~45–60 minutes on average) and had prolonged CPB and ischemic durations compared to conventional cases. This is a consistent theme in the literature: multiple studies report longer cross-clamp and pump times with the thoracotomy approach [18]. Importantly, these prolonged operative times in minimally invasive surgery have not been shown to adversely affect patient outcomes in terms of survival or valve durability. Our study reinforces that point – the minimal-access groups, despite longer ischemic times, had outcomes comparable to the sternotomy groups. All patients in our series received the same cardioplegia (Custodiol®), which provides long myocardial protection; this likely helped buffer the effect of extended cross-clamp times. We observed no relationship between longer cross-clamp duration and in-hospital mortality. In fact, the only group with any operative mortality was the conventional replacement group, not the longest clamp-time group. This suggests that patient factors (such as the higher prevalence of endocarditis and advanced age in the conventional replacement group) rather than operative time per se influenced mortality. The literature also indicates that differences in long-term outcomes between minimally invasive and conventional approaches are minimal. Holzhey et al. noted no significant difference in late survival or freedom from reoperation between approaches [18], a finding echoed by Sündermann and colleagues [20]. Our early outcomes support the notion that minimally invasive surgery does not compromise efficacy. Long-term follow-up would be needed to confirm durability, but previous studies have been reassuring on this front [18, 20, 21].

One of the notable findings in our study was the higher peak CK-MB release in the minimally invasive replacement group. This enzyme is a marker of perioperative myocardial injury. The elevated CK-MB in that group likely reflects the more prolonged cross-clamp time and technical challenges inherent in minimally invasive valve replacement. Interestingly, our data showed that even though CK-MB was higher, it did not translate into worse clinical outcomes for those patients. We specifically found no correlation between elevated CK-MB levels and postoperative mortality. This is an important point: a moderate increase in cardiac enzymes post-surgery may not have prognostic significance if myocardial protection is adequately maintained. Our findings are in line with some reports that transient enzyme elevations after cardiac surgery (even minimally invasive) are common and usually not predictive of long-term damage, especially when troponin and CK-MB elevations remain below very high thresholds. Thus, while we documented a statistically significant difference in CK-MB between groups, clinically all groups had good outcomes. Internationally, there is some debate about the impact of minimally invasive surgery on myocardial injury. Some studies have reported no difference in enzyme release between approaches, while others, like our study, have found slightly higher markers in minimally invasive cases. The consensus, however, is that as long as the patient is well-managed, these biochemical differences have no meaningful impact on patient recovery, which our results confirm.

Our analysis also compared mitral repair versus replacement outcomes. Consistent with expectations, patients who received valve repairs tended to recover faster and had shorter hospital stays than those who underwent full valve replacement. This is logical because repair avoids the trauma of prosthesis implantation and often is performed in patients with degenerative disease who are younger and healthier. International literature supports the superiority of repair when feasible, citing lower mortality and better preservation of heart function. In our study, repair patients (regardless of approach) had about a 4-day shorter hospital stay than replacement patients. They also needed fewer blood transfusions and had slightly lower complication rates. Our findings reinforce the guideline that mitral repair should be preferred over replacement whenever possible, provided the valve pathology is suitable, as it leads to enhanced recovery. That said, our data showed that when replacement was necessary (as in cases of extensive calcification or endocarditis), performing it minimally invasively did not worsen the outcome. In fact, the minimally invasive replacement group in our series had zero hospital mortality, suggesting that with careful patient selection and technique, even complex procedures like valve replacement can be done safely through a less invasive incision. This mirrors reports by Iribarne et al. and others, who found that minimally invasive surgery can be successfully applied in high-risk and elderly patients with outcomes similar to conventional surgery [18, 22]. Our study contributes to this body of evidence by showing excellent results in the minimally invasive cohorts.

Another consideration is postoperative atrial fibrillation, a common issue after mitral surgery. Interestingly, we observed a lower rate of AF postoperatively than preoperatively in all groups. Many patients who had pre-op AF (often due to long-standing mitral disease and enlarged atria) regained sinus rhythm after their mitral valve was corrected. This led to an overall increase in sinus rhythm prevalence post-surgery. All groups had similar rates of postoperative AF, and while we did see a slight uptick in persistent AF in some minimal invasive cases, overall there was no statistically significant difference attributable to surgical approach. This is consistent with other studies noting that the primary driver of postoperative AF is patient-related factors (age, atrial size, etc.) more than the surgical technique itself. Aggressive prophylaxis and management of AF is warranted in all mitral surgery patients, whether done minimally or via sternotomy.

In light of international comparisons, our results align well with the global experience of mitral valve surgery. Minimally invasive mitral surgery offers clear benefits in terms of reduced surgical trauma and wound complications [18], with at least equivalent if not improved recovery times [18, 20, 21]. We confirmed these advantages (no sternal wound issues, rapid mobilization) qualitatively, and found no evidence of increased risk. Our only significant mortality occurred in a group of patients (conventional replacements) known to be higher risk due to pathology, not because of the approach. Large meta-analyses have similarly found that minimally invasive approaches have comparable mortality to conventional surgery, even though bypass times are longer. Some publications (e.g. by Modi et al. and a meta-analysis by Cheng et al.) have suggested minimal access might slightly reduce hospital stay and blood transfusion requirements, which may stem from less tissue trauma and pain [6, 23]. In our cohort, hospital stay was more dependent on whether a repair or replacement was done. Notably, we did not observe a reduction in length of stay with minimally invasive approach for replacement; literature on this point is mixed, but many centers do report shorter stays for minimally invasive cases due to faster ambulation and less pain. Our finding of no difference could be related to our institution’s fast-track protocols for all patients or the particular case-mix. Regardless, the hospital stay in minimally invasive groups was no longer than sternotomy, implying no penalty in recovery.

Finally, our study adds to the discussion on myocardial protection. By using a single cardioplegia strategy (Custodiol®) for all patients, we eliminated one source of variability. Custodiol’s long protection interval is well-suited for minimally invasive cases where reopening for repeated dosing is cumbersome. The uniform use of Custodiol in our series likely contributed to the overall low incidence of myocardial infarction or low-output syndrome post-op (none of which occurred). Internationally, some surgeons prefer blood cardioplegia for such cases, but our good results with HTK solution echo the experiences reported in European centers. We did note that minimal access cases required higher volumes of cardioplegia, as expected, but myocardial preservation remained effective throughout the longer ischemic periods. The lack of correlation between high CK-MB and adverse outcomes in our data suggests that Custodiol provided sufficient myocardial protection to prevent clinically significant damage, even though enzyme levels rose in the most prolonged cases. This finding is important because it supports the continued use of CK-MB (or troponin) as a comparative research metric without overinterpreting slight differences as long as clinical endpoints are acceptable.

In conclusion, this study’s findings are in line with international literature: minimally invasive mitral surgery is a safe and viable alternative to conventional open surgery, offering benefits of reduced invasiveness and cosmetic advantage without compromising surgical efficacy. Patients undergoing minimally invasive mitral repair or replacement had outcomes equivalent to those of conventional sternotomy in terms of survival and key complications. The minimally invasive approach does entail longer operative and perfusion times, but with careful myocardial protection and operative technique, these did not adversely affect patients in our series. We observed that procedure type (repair vs replacement) influences recovery more strongly than the surgical approach, reinforcing the priority of repairing the valve when possible. Both techniques resulted in excellent symptom relief and a low complication profile. These results add to the growing body of evidence that for appropriately selected patients, minimally invasive mitral valve surgery can achieve the same quality of repair/replacement as the traditional approach while leveraging the advantages of modern, less invasive cardiac surgery. Future studies with long-term follow-up will further clarify if any differences emerge in late outcomes, but current data, including ours, suggest no significant long-term trade-offs. Thus, the main takeaway is that surgeons and patients can be confident that minimally invasive mitral valve surgery yields outcomes on par with conventional surgery, with particular benefit in terms of patient satisfaction and recovery, aligning with global trends and literature in contemporary mitral valve therapy.

1. Aluru, J.S., et al., Valvular Heart Disease Epidemiology. Med Sci (Basel), 2022. 10(2).
2. Nkomo, V.T., et al., Burden of valvular heart diseases: a population-based study. Lancet, 2006. 368(9540): p. 1005-11.
3. Otto, C.M., et al., 2020 ACC/AHA Guideline for the Management of Patients With Valvular Heart Disease: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation, 2021. 143(5): p. e35-e71.
4. Sibilitz, K.L., et al., Long-term effects of cardiac rehabilitation after heart valve surgery - results from the randomised CopenHeart(VR) trial. Scand Cardiovasc J, 2022. 56(1): p. 247-255.
5. Gammie, J.S., et al., Trends in Mitral Valve Surgery in the United States: Results From The Society of Thoracic Surgeons Adult Cardiac Database. The Annals of Thoracic Surgery, 2009. 87(5): p. 1431-1439.
6. Cheng, D.C., et al., Minimally invasive versus conventional open mitral valve surgery: a meta-analysis and systematic review. Innovations (Phila), 2011. 6(2): p. 84-103.
7. Goldstone, A.B., et al., Minimally invasive approach provides at least equivalent results for surgical correction of mitral regurgitation: a propensity-matched comparison. J Thorac Cardiovasc Surg, 2013. 145(3): p. 748-56.
8. Sündermann, S.H., et al., Mitral valve surgery: right lateral minithoracotomy or sternotomy? A systematic review and meta-analysis. J Thorac Cardiovasc Surg, 2014. 148(5): p. 1989-1995.e4.
9. Cetinkaya, A., et al., Long-term results after mitral valve surgery using minimally invasive versus sternotomy approach: a propensity matched comparison of a large single-center series. BMC Cardiovasc Disord, 2021. 21(1): p. 314.
10. Mohr, F.W., et al., Minimally invasive port-access mitral valve surgery. J Thorac Cardiovasc Surg, 1998. 115(3): p. 567-74; discussion 574-6.
11. Beckmann, A., et al., German Heart Surgery Report 2018: The Annual Updated Registry of the German Society for Thoracic and Cardiovascular Surgery. Thorac Cardiovasc Surg, 2019. 67(5): p. 331-344.
12. Seeburger, J., et al., Minimal invasive mitral valve repair for mitral regurgitation: results of 1339 consecutive patients. European Journal of Cardio-Thoracic Surgery, 2008. 34(4): p. 760-765.
13. Iribarne, A., et al., A minimally invasive approach is more cost-effective than a traditional sternotomy approach for mitral valve surgery. J Thorac Cardiovasc Surg, 2011. 142(6): p. 1507-14.
14. Iung, B., et al., A prospective survey of patients with valvular heart disease in Europe: The Euro Heart Survey on Valvular Heart Disease. Eur Heart J, 2003. 24(13): p. 1231-43.
15. Beckmann, A., et al., German Heart Surgery Report 2021: The Annual Updated Registry of the German Society for Thoracic and Cardiovascular Surgery. Thorac Cardiovasc Surg, 2022. 70(5): p. 362-376.
16. Yokoyama, Y., et al., Conventional sternotomy versus right mini-thoracotomy versus robotic approach for mitral valve replacement/repair: insights from a network meta-analysis. J Cardiovasc Surg (Torino), 2022. 63(4): p. 492-497.
17. Falk, V., et al., Minimally invasive versus open mitral valve surgery: a consensus statement of the international society of minimally invasive coronary surgery (ISMICS) 2010. Innovations (Phila), 2011. 6(2): p. 66-76.
18. Holzhey, D.M., et al., Minimally invasive versus sternotomy approach for mitral valve surgery in patients greater than 70 years old: a propensity-matched comparison. Ann Thorac Surg, 2011. 91(2): p. 401-5.
19. Gammie, J.S., et al., J. Maxwell Chamberlain Memorial Paper for adult cardiac surgery. Less-invasive mitral valve operations: trends and outcomes from the Society of Thoracic Surgeons Adult Cardiac Surgery Database. Ann Thorac Surg, 2010. 90(5): p. 1401-8, 1410.e1; discussion 1408-10.
20. Sündermann, S.H., M. Czerny, and V. Falk, Open vs. Minimally Invasive Mitral Valve Surgery: Surgical Technique, Indications and Results. Cardiovasc Eng Technol, 2015. 6(2): p. 160-6.
21. Eqbal, A.J., et al., Minimally invasive mitral valve surgery versus conventional sternotomy mitral valve surgery: A systematic review and meta-analysis of 119 studies. J Card Surg, 2022. 37(5): p. 1319-1327.
22. Iribarne, A., et al., Comparative effectiveness of minimally invasive versus traditional sternotomy mitral valve surgery in elderly patients. J Thorac Cardiovasc Surg, 2012. 143(4 Suppl): p. S86-90.

23.          Modi, P., A. Hassan, and W.R. Chitwood, Jr., Minimally invasive mitral valve surgery: a systematic review and meta-analysis. European Journal of Cardio-Thoracic Surgery, 2008. 34(5): p. 943-952.