Cutting the second order chordae
Tethering of the second-order chordae has been recognized as one of the most important mechanisms at the basis of the onset and of worsening of secondary MR. Moreover its persistence is recognized as the cause of failure of isolated mitral reductive annuloplasty (MRA) (fig. 2). Transection of the second-order chordae has been proposed by Messas et al9 as an adjunct to MRA to improve leaflet coaptation and to reduce MR return. However, even if introduced in 2001, there is no general agreement on the efficacy of this technique.
The curvature of the leaflets is designed to reduce the stress on the leaflets during the cardiac systole. The presence of tethered second-order chordae eliminates the curvature and increases the stress on the leaflet. Salgo et al10 demonstrated, in a finite study, that, compared with a flat leaflet, 10% curvature reduced leaflet stress by 100% from 335 to 3 MPa. However, cutting the second-order chordae from one side recovers curvature, but on the other side transfers the tension on the first-order chordae, that now have to face a tension higher than usual. The risk of rupture of first-order chordae under excess of tension and the role of the second-order chordae in maintaining the annulo-papillary stability and LV geometry, casted a shadow on the benefit that chordal cutting (CC) could provide, especially due to controversial experimental results. The positive results obtained by Messas et al9,11,12 were questioned by other experimental works13-17, that showed impairment of the LV function after CC. However, the first clinical reports, appeared in 200518,19, showed good results and no reduction in LV function after a follow-up of 6 months and 2 years, respectively. Furthermore, Fayad et al19identified an aortotomy as the best surgical approach for CC, because of the optimal access to the ventricular side of the AL. In an observational comparative study, Borger et al20 showed that, at 2-year follow-up, MR return was reduced in the CC group with improved ejection fraction (EF) by +10%. Our group, in a propensity matched study, reported, at a follow of 35 months. reduced MR in the CC group with increased EF (from 31% to 40%)21, showing the safety of the procedure. Nevertheless, CC is slowly absorbed by the surgical community, still concerned about possible harmful effects of the technique.
We postulate that the role of the second-order chordae is different in a heart with previous acute myocardial infarction and ischemic MR than in a normal heart. In a normal setting, the second-order chordae, connecting the PMs to the trigones through their insertion to the MV leaflets, absorb the tension generated by the PMs themselves. By their insertion into the MV leaflets, they contribute to reduce the peak systolic stress. In a pathologic situation, the second-order chordae, which insert on the AL, being tethered, eliminate its curvature, increasing the systolic peak stress by 100-fold10. By cutting the second-order chordae, the AL recovers its curvature, reducing the systolic peak stress by 100-fold, but now the transmission of the systolic stress falls completely on the first-order chordae. In an experimental setting, Padala et al22 found that, displacing the PMs in three different directions (apically, apico-laterally and apico-lateral-posteriorly), the tension on the marginal chordae, compared to the basal values, increased more on the right-sided (till 4.6-fold) than on the left-sided chordae (till 2.1-fold) and were generally lower when the PMs were displaced apically and higher when displaced apico-lateral-posteriorly. When the anterior second-order chordae were cut, the increase of tension, compared with basal values, was respectively till 7.3-fold for the right-sided and till 3.7-fold for the left-sided marginal chordae. Then the tension on marginal chordae increases with PMs displacement, with a further increase after CC. As the first-order chordae are stiffer but less elastic than the second-order chordae, the increased tension can bring the marginal chordae closer to the rupture point. However, in experimental settings, the measured forces supported by the first-order chordae of the AL after CC were not higher than 0.4 N22,23, significantly smaller than the failure load of 6.8 N required for rupture of the anterior marginal chordae reported by Sedransk et al24. The risk of rupture of the marginal chordae seems then to be only theoretical, and anyway overwhelmed by the benefits (recover of the AL curvature and improvement of the coaptation).