In two papers published in this issue of the Journal [1, 2] and presented at the last EACTS Annual Meeting, the group from the National Heart Institute of Kuala Lumpur, Malaysia, discuss their long-term experience of repair for rheumatic mitral valve regurgitation, with the aim of identifying predictors of durability and of comparing them with repairs for degenerative mitral valves, in one of the manuscripts, and to analyse a particular aspect of the technique of repair—leaflet extension, in the other. Mitral valve repair for rheumatic disease is commonly believed to yield poorer results by comparison with other aetiologies, especially degenerative. This is a natural consequence of the evolutive nature of the rheumatic process, which continues to distort the valve apparatus even beyond a successful repair. These rheumatic populations from developing countries are characterized by the low mean age of 20–30 years (32 years in the current series) [3], which renders them susceptible to further bouts of the disease. Hence, antibiotic prophylaxis must be continued until a much older age, at least to 40 years, and the WHO now recommends life-long prophylaxis in patients with severe valve disease or who have had valve surgery [4]. But it is essential to recognize that valve replacement in rheumatic populations also has poorer outcomes, in some reports clearly worse to those of repair, mainly due to deficient socio-economic conditions. My own experience with a similar population in Johannesburg, South Africa, attests to the superiority of repair in these patients [5]. Others have reported similar experiences [6]. It is, however, necessary to distinguish between the rheumatic disease of these patients and that of series originating from developing countries. The latter, now a rarity, represents a different stage of the disease and the mean age of the patients is much higher ( 50 years) [7]. In this case, the histological process has stabilized and most patients only have dilatation of the annulus. Hence, in the absence of significant calcification, the feasibility of repair is higher and the results better. During a 14-year period, the group from Kuala Lumpur repaired the mitral valves of 627 patients, predominantly with regurgitation, a number similar to that done for other aetiologies, and found that the actuarial freedom from valve failure for rheumatics at 5 and 10 years was 85.6 ± 2.3 and 72.8 ± 4.6%, respectively, similar to that for degenerative valves. Although this study has the limitation, acknowledged by the authors, of being a single-centre non-randomized series, it certainly adds to the evidence that contemporary repair of the rheumatic mitral valve is possible in the majority of patients and has excellent results, hence should be the procedure of choice also for this pathology and in these populations. What has changed in recent years that so significantly improved the outcomes? In my opinion, the use of artificial chordae made of polytetrafluoroethylene, to replace ruptured chordae or to re-inforce elongated chordae, has had the most impact. The majority of patients with rheumatic regurgitation have prolapse of the anterior leaflet due to elongated chordae tendineae. In the past, this was treated by one of the techniques of chordal shortening or transfer, as originally described by Carpentier [8]. But with progression of the rheumatic process, the chordae continue to elongate or rupture, causing the repair to fail. For this reason, most surgeons, lately also Carpentier, have turned to the artificial chordae to re-level the free edges of the leaflets to promote adequate coaptation. This can be used even in the younger and smaller patients, and several studies have confirmed the superiority of the results [9]. Another technical alternative which has evolved in the past two decades is leaflet extension, to increase leaflet area and coaptation surface, which is the subject of the second paper authored by the Kuala Lumpur group. The authors treated 62 of 446 rheumatic patients by leaflet extension with glutaraldehydetreated autologous pericardium as part of their mitral repair. The authors analysed prospectively the clinical and echocardiographic data of the patients and found the estimated 5-year freedom from reoperation and valve failure to be 96.8 and 91.6%, respectively, but the follow-up extended for only a little over 3 years, which is much too short to draw definitive conclusions. Leaflet extension is not new. It was used extensively by some of us in the late 1970s and early 1980s, but the results then reported were not very encouraging, most patients developing mixed regurgitation and stenosis due to shrinkage and thickening of the pericardium. But in those days the autologous pericardium was used untreated. The glutaraldehyde-treated pericardium was introduced later as a sequel of the utilization of the similarly treated bovine pericardium, for different uses. The current study of Dillon et al.