Difficulties in implantation of a cardiac resynchronization therapy system: causes of failure and importance of operator's experience

This editorial refers to ‘Predictive factors of difficult implantation procedure in cardiac resynchronization therapy’ by L. Bisch et al. , on page 1141. Clinical practice and many studies demonstrate that success of cardiac stimulation with resynchronization is, to a large extent, associated with the site of left ventricular stimulation, which is important to identify correctly. 1 In the vast majority of cases, it is reached via the right-sided endovascular approach, but the need for an optimal position of the catheter gives rise to the fact that some alternatives to this traditional method have been proposed, such as the surgical direct approach (mini-thoracotomy or video thoracoscopy) or left ventricular positioning via the transseptal approach. The article by Bisch et al.2 underlines the difficulties of the approach most commonly used, the right-sided endovascular approach. In a very interesting manner, the authors sought to determine the predictive value of difficulties with implantation based on analysis of several parameters. They emphasized the importance of operator's experience and of the severity of haemodynamic conditions. 3 The difficulties encountered during implantation of a resynchronization system can have consequences that are much more serious when patients who are to undergo this method of stimulation are frail. When duration of the procedure is prolonged, and in cases where a second procedure is required, there is increased risk of infection or traumatic complications (dissection and perforation). Among many causes of failure of the method, some relate to anatomy of the cardiac venous system: failure to insert the lead into the coronary sinus, inappropriate implantation site for resynchronization, and phrenic nerve stimulation. Other causes relate to immediate complications (such as dissection of the coronary sinus wall or venous perforation). Lastly, they depend on the equipment used—either it is not suited to the patient's anatomy, anatomy sometimes being modified by heart failure, or the epicardiac veins are not conducive to positioning of a lead which is too large or which is made unstable by the structure of the blood vessel. However, the majority of failures result from operator's lack of experience: every method requires a learning curve, and stimulation with resynchronization is no exception to this rule. Furthermore, experience needs to be maintained by performing a minimum number of procedures per year. What is the estimated frequency of technical failures of the method? (We will not consider cases which do not respond to treatment; although these are certainly failures of the method, they are often associated with causes different from those which we will be considering and discussing in this article.) Studies report between 4 and 15% of technical failures, all causes combined and for first implantations. 4 , 5 It is necessary to distinguish failures of cannulation where it is impossible to penetrate the coronary sinus or to advance into it (about 5% of implantations), dissection of the coronary sinus (1–2% of implantations), and instability of the lead or its positioning in a vein which is unsuitable for resynchronization due to its location, size, or the threshold of stimulation measured in it (2–5% of implantations). The success or failure of implantation of a left lead obviously depends on the anatomy of the cardiac venous system. 6 In the first place, insertion of the lead into the coronary sinus, directly or most often via an introducer catheter, can be hindered by the Thebesian valve: it exists in 75–95% of subjects, and therefore is totally absent in only 5–25% of cases. This valve is most often a membrane which is almost always cribriform, with an upper free border which is concave upwards and which partially occludes the opening of the coronary sinus into the right atrium. This configuration is the most common type (up to 70% of cases). In 9–30% of subjects, the Thebesian valve is so developed that it is almost occlusive or at least is a major obstacle to penetration of the coronary sinus by a catheter or an introducer catheter. The use of a flexible guidewire can be very useful in such cases. Lastly, in rare instances, a well-developed Chiari network is encountered, but this generally does not pose special difficulties in crossing it. A few centimetres upstream in the coronary sinus, Vieussen's valve most often marks the ending of the great cardiac vein and the opening of Marshall's vein into the coronary sinus. This small valve exists in 75–90% of cases. Generally, it is not an obstacle for the lead to progress into the venous system. Presenting as a single cusp in almost 60% of cases and two cusps in only 25% of subjects, they are rarely sufficiently developed to impede positioning of the lead, but the operator may experience the feeling of a projection at the end of the lead evidencing their existence. Lastly, in rare cases (1%), a tricuspid structure has been described, but it has not been possible to confirm that this condition increases difficulties in implantation. Success of resynchronization, as we have seen, greatly depends on the site of stimulation. We know that it is preferable to position the lead in a posterior, marginal, or lateral vein rather than an anterior one. Most often, the adequate veins exist. 7 However, the venous network does not always offer the possibility of optimal positioning. The left marginal vein is found in 70–90% of cases and the posterior vein of the left ventricle in 60–65% of cases. Regarding the posterior interventricular vein, in 90% of cases, it empties into the coronary sinus and therefore can be approached if other veins are absent or too thin for insertion of a stimulation lead. A venogram is often very useful: some systematically perform this procedure (during the implantation procedure rather than the day before), whereas others use it only in the case of failure of cannulation of an appropriate vein. Haemodynamic conditions in heart failure patients sometimes modify anatomical relations. It is often difficult to place a lead or an introducer catheter into a very dilated right atrium and the aid of an electrophysiological steerable catheter can be useful. However, heart failure can prove to be an appreciable aid in dilating the coronary sinus and its ostium. Is it less difficult to implant the lead from the left side of the patient rather than the right side? Most often implantation is performed in the left subclavian area and the design of the equipment used gives the impression that this solution is most appropriate. However, operator's dexterity and personal experience are more decisive than theoretical conditions. An introducer catheter is most often used to facilitate insertion of the lead into the coronary sinus. Manufacturers have progressively improved this equipment to best suit patient's anatomy. Since the structure of the right atrium and the direction of the ostium of the coronary sinus vary depending on each subject's anatomy and underlying disease, various curved catheters should be available. Equipment entering into the composition of such procedures is essential: it is necessary to avoid damage to the walls of the coronary sinus which can produce dissection, and any bending of the catheter making its placement impossible. Lastly, the design of the lead is one of the most important items defining success of implantation: it is not decisive but its thin aspect, the curvature of its end, and the ‘sliding’ feature of the catheter should allow its easy insertion into narrow veins or a stable position in a large vein. All these factors largely condition the success of implantation of a left-sided lead in the setting of resynchronization. Nevertheless, the essential factor remains the operator's personal experience. 8 It is irreplaceable and justifies that practice of implantation should be determined by the yearly number of procedures. In every team, a learning curve is observed, inherent with every interventional method. Once learning has been completed, clearly it is necessary to maintain the experience acquired by performing a minimum number of implantations each year. What is the optimum number? It is difficult to define as it varies according to each operator's ability. About at least 30–50 procedures a year per operator seems to reflect reasonable practice. Conflict of interest: none declared.