JACC Vol. 19 . No. b May 19a2:1J6a-711
Reply The comments are well taken . However . the main purpose of the report was to focus attention on the effect of antiarrhythmic medication on atrial fibrillatory activity where dramatic us aaes ware observed with each of the antiarrhythmic medications . Clearly, propifenone as administered had a good eleclrophysioloeic effect even with the regimen used. We did not intend to rigorously compare procainamide and propafenone but merely used each drug as an example in (heir class . GEORGE 5 . KLEIN . MD . FRCPICI . FACC Arr(nthnua Senior Castor inve,fiii" t Ti ii Unit r,nito Macpear P.O. Box 5339. Penal Srnrion A Londe.,, Ontario. Condo MA 5A5
Risk Stratification for Arrhythmic Events in Postinfaretion Patients Based on Heart Rate Variability Fslrell et at. (I) described the value of heart rate variability from 24-h ambulatory ECG recordings taken I week after infarction, This heart rate variability combined with late potentials or repetitive ventricular forms from the same Holler monitor seems to have a high prognostic importance in identifying approximately 1!114 of these early surviving postinfarction patients who are at a much higher risk of developing a life-threatening arrhythmia event . 11 will be hard to disregard these results in a prospective trial evaluating any antiarrhythmic dmg in this patient population. Because this measure of heart rate variability, calculated as the base width of the main triangle that constitutes the RR interval frequency distribution curve (Appendix of 1111. is so specific and sensitive in defining this small population at high risk, the units that are used to express this base width variation of RR intervals should be correct. As described in their original report describing the technique (2), this triangular interpolation method in 20 patients of a similar population that did have an arrhythmic event averaged 258 ms, in agreement with a recent abstract 131 in 385 postinfarction patients describing the sensitivity and specificity of this heart rate variability method for values between 156 and 305 ms . In contrast, a criterion of <20 ms is given by Farrell et al . (1) for similar patients with impaired heart rate variability . The incorrect values of base width heart rate variability in (I I arc easy to explain . As the authors themselves pointed out in the Appendix, the RR intervals are measured on a discrete scale in steps of 11128 s . This 11128 s is specific for their particular computer-based analysis system and is determined by screen resolution and (limedependent) QRS complex enlargement . The original base width calculations are therefore in the first instance obtained in units of 7.815 ms and it seems to be exactly this factor that was forgotten to he taken into account . To obtain the correct heart rate variability base width, all results of this variable in 11) should be multiplied by 7.8125. This means that the summary and conclusions should be that impaired heart rate variability is defined as having a base width variability of < (56 ms. Allematively, it is also possible that the bean rate variability in (1) was actually determined by the "triangular index" method . which is describ .,! in their earlier paper (4) and is identical to "method C" in (2). The best prognostic variable for serious arrhyth-
mic ascots in 141 was this hears rate vanalnlrty index of <25 . This trion^-ular index, which was not given any units, eon in the same way easily be transformed into base width by multiplication by 2 x i vi?i those a Iriorgle arearheight = 2 x triangular index in units of 1,I'-8 U . Not surprisingly . this transformed triangular index base width. v:hcn recalculated from C) . agrees excellently with the unanguhv in crpolation base width values . except for I of the 40 described patlevts : P19. The listed value of 125 ms as base width (method El is clearly an underestimation of [me heart rate variability as visual!v can he seen from Figure I in (2). The transformed meth., C oa:r width for P19 is 304 me and points out that there is a polenti ;d problem with the tdangodar interpolation medwd whenever the frequency distribution curve happens to be bimodal . This himodality actually happens quite often as it is merely the consequence of the different day-versus nighttime basic rhythms . Because it has been shown that heart rate variability is highest at night, except possibly for those patients with vastly impaired heart rate variability (5 .6). The nighttime variability might be of an even higher prognostic value, and at the same time it would standardize the different activity levels among patients loll sleeping) and probably prevznr the oecuactice of bimodal frequency distribution v s . I wonder whether the authors have ever looked at this prognostic value of nighttime heart rate variability! L . JULES VERSCHUEREN .PHD enrdi .ru r, ulnr Clerirnl RenelnlIi beinroi the" sgnihb fn I. gulp . Terlmlpn~rrremrce 154 1170 151uv-1a.150pfvm
References I . F rvil
ne emhi, Y . Cone' T. It at Risk .nw,tcarmn to, xot thmre e roainfarnmn p ate s hosed on heal vIC,anchtity, ambutatun, ekorwo,diota phie amblm evltbevgncLevereged ekcvorerdinpvm .l AmCel)Cardid "na .isrBl~7. 1. lfaN SL Fmell S ('ripps T. Camm Al. Hears rare,aochtiry in ,,ions u propnmis Aer dial id r
Mitral Commissurotonky in India 1 he article by Hickey et at, III detailing the results of surgical mitral commissurotomy is an eye-opener in this era of a progressive trend toward open mhral commissurotomy for all patients with mitral stenosis /2 .31. In this letter we describe the shun-term re~Jlts of mitral commissurotomy . as determined echocardiogmphically, in patients with mitral stenosis . Over the past 2 years 163 patients unaerwent open or closed mitral commissurotomy at this institution and were assessed in detail echocardiographically. The distribution was nonrandomized. with a definite bias toward the open procedure in patients with severe forms of subvalvular fusion . The main differentiating factor, in a poor country like ours, was financial constraint . There were 137 palicets (47 male . 90 female) with pure critical mitral stenosis : the mean age ores 26 years in the open commissurotomy group and 25 .5
LETTERS TOTHE EDITOR
years in the closed commissurotomy group (n = i05) . The preoperative features were similar in both groups with respect to valve involvement, subvalvular pathology and pulmonary artery pressures (as predicted by echocardiography). The closed commissurotomy group had a good result of valvotomy (mitral valve area, 2.3 t 0.52) and compared favorably with the open commissurotomy group (mitral valve area, 2 .52 t 0.56) (p = NS), and in all other variables the two groups were similar . Mitral regurgitation occurred with equal frequency in the groups with open and closed commissar . otomy (68% vs . 66%) although the group with tire closed procedure had more patients with significant mitral regurgitation (9 .3% vs . 11 .3%) . In both groups mitral regurgitation occurred in patients with more severe subvalvular fusion . Chordal tear, laceration and partial tear of the papillary muscle, cusp avulsion were more often seen in patients undergoing closed commissurotomy (9 .8%). All of these patients had severe subvalvular fusion . This is only a preliminary report and more detailed analysis and follow-up are necessary before a definite conclusion can be reached . However, we still believe that the closed procedure has a place today, at least in financially indigent populations . The only watch word is to choose the appropriate patient carefully . K.PRASAD S. RADHAKRISHANAN PROF. P. S . BIDWAI Department aj Cardiology e .G.P.G .J .A9s Rae Bureli Road Larknaw 226 001, India
IACC Val. 19, No. 6 May 1992 :1768-71)
none of the currently used calcium channel blockers (verapamil, nifedipine, dilliazem) has been shown to be as favorable . Furthermore, a trial usingdiltiueem in patients with non Q-wave myocardial infarction (2) (the reference cited by Bakris and Frohlich) showed no reduction in mortality. Moreover, the Multicenter Diltiazem Postin . farctioe Trial (3) showed no difference in mortality between patient groups. On the contrary, the subgroup of patients with left ventricular dysfunction did worse with dilliazem . From the currently available data it may be concluded that the prophylactic use of calcium channel blockers including dittiazem to improve survival after myocardial infarction cannot be recommended at this time . GEORGE BITTAR, MD Deparnnem of Medicine Sdwol of Medicine Hunrington, West Virginia 25755
References 1. BAds GL, rrnhlkh ED . The emlution of anliaypenensive ttempt : an vaaeiew of fiver decadesofexpedawe. I Am Call Cordial 1989:14:1595-608. 2. Glcs.n RS, souse WE,Theroux P, n al . Dihiaxm and reatarcrion in petiems with non Q-wave myocardial infarction. N Cant I Med 1985315 :423-9. 7. Muhicemer Diltiaum Posrardurion Trial Research Group The eOed ordilriaaem on momlily and reinfarmion utter myaa dial infarnion . N Eng1I Med 1988:3t9:1g5-92 .
References L Hickey 5f5J, BDCknone EH, Kirklin JW, Ikon LS . Outcome probabilities and lire hianry,dnr curgiml commi,curmamy : implieahnns foe bannan .almlmny. J Am Cau Carve„ 11991 :17 :29-42. 2. Nekano S. Kawachima Y, Hi.. H, n .I . Recomidemrium of i,wiaetiera for open mint -m)--any based on Phallic features r,stenosed milrd valve: a fonrtwn year follow up study in 347 con-live pntimts. J Thorc Cardiseasc Sur g 1987:94:336-42. 3. No Earn., M, Bausmdia H . GadjbAhch 1 . el al, Clesed ee s open mieal ., in pure noncxlcific milml aenasi .i hemady,wmie studies bar- and -mm aftaropemlion . J Thoac WNioaasc Sun 1990 .99:639-44.
Bittar is correct that the reference we cited did not present information concerning deaths related to non-Q wave infarction in patients with hypertension . We included reference to the original paper on the Multicenter Diltinzem Postinfaretion Trial because it detailed the study design . Not included were details of a paper whose data we heard presented at a national meeting . We had planned to cite that paper on the return of proofs to us, but it was not yet published. The paper by Moss et al . (I) is now published presenting data demonstrating an overall reduction in cardiac deaths or nonfatal recurrent myocardial infarction in patients treated with diltiazem, results not observed in placebo-treated patients or patients without a history of hypertension . We are pleased to add this reference in support of our statement . It underscores the sometimes lengthy delay from presentation of new information to publication .
Diltiazem and Mortality in Myocardial Infarction
Reference Oscar D, Rutison M, cr al. Enact s of dittiaum to long,,. Mt.. aftn acme myocardial infarmian in palimls with and without a history of systemic hypertension . Ant Cardial 1991:68:429-33 .
I . Moss ar,
In their review article, Bakris and Frohlich (I) state that the calcium channel blacker dilliazem reduces mortality in patients with non Q-wave myocardial infarction. The reference cited does not support their view. Compared with the established protective effects of a betablacker in prolonging life and reducing recurrence of nonfatal myocardial infarction in survivors of acute myocardial infarction,
EDWARD D. FROHLICH, MD, FACC GEORGE L . BAKRIS, MD Alto. Oehsnee Medical Foundmi. 1516 Jeff-an Highway NetcOrlaanc,Lonioiano 70121