CARDIOVASCULAR JOURNAL OF AFRICA • Volume 33, No 2, March/April 2022 AFRICA 63 periods beyond 12 months (Fig. 5). The same trend was noted for moderate to severe pulmonary hypertension. The percentage of patients who reverted to atrial fibrillation and/or flutter at any other period after the initial success (sinus rhythm) at three months, was analysed. Four patients had no record of their rhythm status at any other period other than the initial sinus rhythm at three months. Therefore, the rate of relapse could only be calculated on 25 instead of 29 patients. Of the 25 patients who were in sinus rhythm at zero to three months, only four (16%) reverted to atrial fibrillation. Of note, the long-term rhythm of the patients was determined by their three- to six-month rhythm. Therefore, 100% of patients in sinus rhythm between three and six months remained in sinus rhythm in every other future rhythm recording. Similarly, there was a 92.9% probability that patients with atrial fibrillation during the same period remained in atrial fibrillation thereafter. Only two patients (3.9%) out of the 51 available patients at zero to three months had atrial flutter. Discussion Success rates of 56.9 and 64.9% at zero to three months and six to nine months in this study fall in the reported ranges by other researchers. Blackstone et al. found success rates of about 59 and 63% for the same periods.8 Moreover, Chavez et al. found initial sinus rhythm in 63.1% of patients with bi-atrial ablation, suggesting that bi-atrial ablation has the same initial outcome as unilateral ablation.9 Such similarity in outcomes can be explained by similar patient pre-operative characteristics. The patients in these studies had rheumatic valvular heart pathology. Brazil particularly has patient characteristics comparable to those in Africa. Other researchers however had higher success rates than those in our study. Raman et al. in Australia reported an 84% success rate at three months, which is comparable to the success rate reported by Cox et al. in 2000.10 Similarly, Ad et al. reported 82, 87 and 79% success rates at six, 12 and 24 months, respectively, after left atrial ablation.11 Favourable pre-operative factors may be a major contributor to such good success rates. Firstly, patients in these two studies had small left atrial sizes compared to our study population, with all patients in the study by Ad et al. having less than 4.5 cm atrial size. In comparison, about 60% of our patients had left atrial sizes of 5 cm and above, with the largest being 8.7 cm. This suggests that our patients had longstanding valvular disease compared to those in the above two studies, hence the higher risk of failure. The general trend of initial improvement in the success rate from the first three months to about 12 months, followed by a decline in the success rate beyond 24 months, was also noted by other researchers. Ad et al. showed an initial improvement in success from 82 to 87% from six to 12 months, which declined to about 79% at 24 months after ablation.11 Our success rates were about 63, 75 and 71%, respectively, for the same periods. This may confirm what Cox et al. suggested, that the long-term success rate of an incomplete (unilateral) ablation tends to fall significantly after two years, as opposed to that of a bilateral ablation.4 Although not statistically significant, we found patients with poor ejection fraction (less than 50%) to have poor success rates compared to those with good ejection fraction. However, among patients with immediate ablation success (sinus rhythm at three months), there were significantly more patients with good pre-operative ejection fraction of 50% and above (72.4%, p = 0.016). This suggests that patients with good outcomes are likely to be those with good ejection fractions. This can be explained by the chronicity of the disease before the operation and the amount of structural and electrical remodelling of the atrium. Whereas most studies in the literature tended to have patients with good ejection fraction, our study included about 37% of patients with poor ejection fraction. The controversial use of anti-arrhythmic medication after cardiac ablation was also evaluated in our study, and it was found not to have any effect on the outcome. Patients’ rhythm status was not related to the use or not of such medication. The majority of patients were on oral amiodarone, and a few were on beta-blockers and digoxin upon discharge. The literature findings on this matter remain inconclusive, with Henrica and colleagues not finding any benefit from the use of amiodarone after bi-atrial ablation, while the results of Raman et al. show beneficial effects of such medication.10,12 The lack of effect of anti-arrhythmic medication in our study and others can be explained scientifically as follows. Once standard cardiac ablation is done by making appropriate transmural ablation lines, the flow of electrical current in the atrium is directed by physical barriers, which do not allow the current tomake continuous circles between the newcompartments (re-entry phenomena). Therefore, a pharmacological drug is not expected to allow the electrical current to cross such physical barriers, hence no effect of medication on the success rate. The effect of immediate post-operative (first 24 hours) use of amiodarone infusion however was not evaluated in our study. Large left atrial size is one traditionally poor prognosticator, according to some literature. However, there was no influence of left atrial size on the success rate of ablation in our study. A cut-off value of 60 mm was used in this study, as opposed to the 4.5 cm of Pecha et al. and Ad et al.11,13 Pecha et al. found a higher success rate with sizes less than 4.5 cm, as opposed to our Months (%) 0–3 3–6 6–12 Percentage sinus rhythm per group 80 60 40 20 0 0/mild Moderate/severe Fig. 5. Trends of sinus rhythm (success rate) over time according to tricuspid regurgitation severity. Although not statistically different up to 12 months of follow up, the diverging nature of the curves suggest a possible wider separation of the curves with a longer follow up.
RkJQdWJsaXNoZXIy NDIzNzc=