Introduction

The presence of air micro-emboli in open-heart surgery correlates with the degree of post-operative neuropsychological disorder [1, 2]. Manual de-airing techniques have proved ineffective in eliminating air micro-emboli and even meticulous techniques are associated with the risk of a large number of micro-emboli [3, 4]. The neurological outcome is difficult to evaluate, due to possible bias such as the status and the age of the patient and symptoms (e.g. changes in personality) [5, 6].

However, it is possible that these changes are not connected with the operation itself but rather with the status of the patients, their age, sex, disease severity, or genetic factors [7, 8]. The use of carbon dioxide (CO2) in minimally invasive cardiac surgery is due to its high solubility and density in blood, allowing better tolerability of air embolism [9]. The use of endo-cavitary aspirators during mitral valve surgery contributes to capture in the extracorporeal circuit the quantity of CO2 continuously insufflated in the surgical field. This aspect is represented in the blood gas analysis and in the frequent correction of hypercapnia through ventilation in the oxygenator [10].

In this context we investigated the effect of CO2 on two groups of patients undergoing minimally invasive mitral valve repair (MIMVR) through a right mini-thoracotomy with two different CO2 delivery techniques (continuous vs. one end shot) and we compared the peri-operatory micro-embolic activity, the impact of CO2 in cardiopulmonary bypass (CPB) management, the incidence of transient post-operative cognitive disorder (TPOCD), mechanical ventilation (MV) duration and intensive care unit (ICU) length of stay.

Methods

Patient and data collection

A retrospective, observational study was undertaken of prospectively collected data in one hundred consecutive patients undergoing MIMVR from January 2018 to November 2021 at our Institution Anthea Hospital, GVM Care & Research, Bari, Italy. The median (interquartile range [IQR]) age was 66 (62–76) years, one hundred patients underwent MIMVR through a right thoracotomy approach. Patient characteristics are reported in Table 1. None of the study patients reported the use of psychiatric drugs, alcohol, and carotid artery stenosis prior to the procedure.

Table 1 Characteristics of the study population
Full size table

Fifty patients were insufflated with continuous CO2 1 min before opening the left atrium and ended after its closure, and fifty patients were insufflated with one shot CO2 10 min before the start of left atrium closure, at a continuous CO2 flow rate of 3 L/min via diffuser (Table 1). The main reason for performing two different methods of CO2 delivery during MIMVR was due to the different techniques used by cardiac surgeons for minimally invasive cardiac surgery. The aim and the methodology of the study was internal discussed with the ethics committee of the hospital according to the General Data Protection Regulation. Because of the retrospective nature of this study, the local ethics committees waived the need for patient consent. The transesophageal echocardiographic (TEE) protocol for the detection of micro-emboli requires to record intraoperative TEE from cross-clamping to 20 min after end of CPB.

Post-operatively, a blinded assessor determined the maximal number of gas emboli during each consecutive minute in the left atrium, left ventricle, and ascending aorta. The primary outcome of the study was the incidence of TPOCD (in particular agitation and delirium occurring 5 h following weaning from anesthesia), MV duration and ICU length of stay.

During the two procedures, correction for hypercapnia during CPB and monitoring of VCO2 changes were recoreded.

Surgical technique

Our surgical approach for minimally invasive direct view during mitral surgery was described elsewhere [11]. Arterial perfusion was always retrograde and peripheral and aortic cross-clamping was external in all patients. Venous cannulation was peripheral with vacuum support and a double site insertion of the cannulas (jugular and femoral). The valve inspection and procedure were through the left atrium with direct vision and the reconstruction technique was standardized [11].

CO2 insufflation management and CPB de-airing

A small PVC flexible drain tube was used for CO2 insufflation as per standardized procedure [12, 13] and flow measurement was performed with a flowmeter for medical CO2. The perfusionist regulates the flow according to pCO2 and pH. PaO2 during CPB was maintained between 150 and 250 mmHg, PaCO2 was maintained through the sweep gas (air flow from gas blender) between 40 and 45 mmHg with pH stat management, and mean arterial pressure was maintained between 50 and 70 mmHg [14, 15]. In both groups, the venting flow was maintained 800 ml/min after cross-clamping. Air embolism was managed under TEE guidance; the heart sections were filled, thus obstructing the venous return from CPB and increasing the cavity diameter, and the lungs were manually expanded using an Ambu® resuscitator (Ambu A/S,Ballerup, Denmark) at a rate of 4 inflations per minute. The ventricular and aortic intracavitary aspirators were managed at 750 ml/min and 800 ml/min after cross-clamp removal, and the aortic root vent was removed at the elimination of total gaseous micro-emboli.

Statistical analysis

Continuous data are expressed as median with IQR and categorical data as percentages. Cumulative survival was evaluated with the Kaplan–Meier method. All reported P-values are two-sided, and P-values of < 0.05 were considered to indicate statistical significance. All statistical analyses were performed with SPSS 22.0 (SPSS, Inc., Chicago, IL, USA).

Results

CPB duration was 78 ± 13 min and cross-clamp time was 40 ± 9 min (Table 2). The most predominant pathology was degenerative disease, followed by rheumatic mitral valve disease (Table 3). Mitral valve repair was performed in all patients with peripheral cannulation. Repair techniques included annuloplasty, leafleat resection, neochordae implantation and sliding plasty. Table 4 depicts the median number of micro-emboli during the first 15 min after release of the aortic cross-clamp in the three areas of interest taken together. All patients in both groups had micro-emboli after release of the aortic cross-clamp in all three areas of interest. The number of micro-emboli recorded with TEE was higher in the control group (Table 4) and remained constantly higher during all four time periods and in all three studied locations.

Table 2 Intraoperative data for surgical techniques and procedures
Full size table
Table 3 Mitral valve pathology for minimally invasive mitral valve repair
Full size table
Table 4 Number of microemboli on transesophageal echocardiographic evaluation of the left atrium and ventricle and the proximal ascending aorta
Full size table

In the continuous field flooding CO2 group, the median number of detectable micro-emboli after CPB fell to zero 9 ± 5 min after CPB versus 19 ± 3 min in the one-shot CO2 control group (p = 0.01). In patients of the continuous field flooding CO2 group, correction of ventilation for hypercapnia during CPB was applied, with an increase of mean sweep gas air (2.5 L) and monitoring of VCO2 changes. One patient of the continuous CO2 group vs. 9 patients of the control group reported agitation at discontinuation of anesthesia (p = 0.022). MV duration was 14 ± 3 h vs. 27 ± 4 h (p = 0.016) and ICU length of stay was 33 ± 4 h vs. 42 ± 5 h (p = 0.029) in the continuous CO2 vs. control group, respectively (Table 5). In the whole study population, no transient ischemic attack or stroke was reported at postoperative clinical evaluation (Fig. 1).

Fig. 1
figure 1

Right thoracotomy for minimally invasive mitral valve repair

Full size image
Table 5 Peri-operative and post-operative outcome
Full size table

Discussion

Previous studies [16, 17] demonstrated that the patients without CO2 use had persistent air bubbles for many minutes after the end of CPB but these studies were not performed under TEE control, as in our analysis, and no cerebro-vascular outcome was reported [18, 19].

Moreover, subsequent randomized studies showed no difference or were too small to demonstrate a difference in neurocognitive outcome between CO2 and no-CO2 use [20, 21]. In other words, our study is the first that demonstrate a clinical impact of that strategy.

However, the centrality of TEE use has been previously highlighted for bubble observation [19] but not yet for the clinical outcome effect. Other authors, on the other hand, demonstrated an impact on cardiac function due to less air bubbles in the heart [22]. It should be noted that all these studies tried to compare the use vs. non-use of CO2. We are the first that tried to demonstrate a difference in the use of the CO2 strategy trying to reduce the possible site effects of CO2 (e.g. high pCO2) with the support of the perfusionist and a strategy that focuses the use of gas only during the phase of chamber opening. An excess of micro-embolic activity could influence the patient’s awakening by giving drowsiness and transient agitation, this would seem to have an indirect impact on the lack of collaboration by prolonging the time of MV and ICU length of stay.

The main limitation of our study is the quantitative assessment of gaseous micro-embolic activity with a correlation for the primary endpoint of the incidence of TPOCD (in particular agitation and delirium upon discontinuation of anesthesia), MV duration and ICU length of stay, which should be further explored in future studies with instrumental investigations (e.g. magnetic resonance imaging), and be correlated with intraoperative bispectral index, electroencephalogram, and evaluated with cognitive tests in the short, medium and long term in relation to the patient age and gender and the impact of retrograde perfusion and atherosclerotic burden [23].

Conclusion

Continuous field flooding insufflation of CO2 in MIMVR is associated with a lower incidence of micro-emboli, possibly due longer exposure to CO2, and a lower incidence of agitation at discontinuation of anesthesia as well as improved MV duration and ICU length of stay.