Combined treatment with valsartan and fluvastatin to delay disease progression in nonpermanent atrial fibrillation with hypertension: A clinical trial

Abstract Background Atrial fibrillation (AF) is a complex cardiac arrhythmia in clinical practice with increasing incidence. However, the effects of statins on patients with AF are not quite clear. Hypothesis To investigate the protective effect of calcium channel blocker (CCB) and valsartan combined fluvastatin on hypertension (HTN) patients with nonpermanent AF. Methods In three and a half years, 189 cases of patients diagnosed as HTN combining nonpermanent AF by eight medical centers, were recruited and randomly assigned to four groups with varied treatments: CCB group; CCB + statin group; valsartan group; and valsartan + statin group. The four groups were followed up for 24 months. The 7‐day Holter ultrasound echocardiography (UCG) and biochemical indexes were completed at preset time nodes respectively. Results After 24 months of follow‐up, 178 patients completed the study. Compared with CCB group, the blood lipid level, inflammatory index, ultrasonic index and electrocardiographic measurement results of CCB + statin group, valsartan group and valsartan + statin group were improved in different degrees and had statistical significance (P < .05 or P < .01). Furthermore, the improvement trend of CCB + statin group and valsartan + statin group was more obvious. Conclusions The results indicated that valsartan can reduce AF load and recurrence rate, and delay the progression of nonpermanent AF to permanent AF in multiple ways, and the effect of combination of valsartan and fluvastatin is more significant. These results provide a new direction for the integrated upstream control strategy of AF.

pressure (DBP) ≥ 90 mmHg continuously measured twice in a s resting state, or the patient who has blood pressure controlled as less than 180/110 mmHg by using antihypertensive drugs since his/her clear history of HTN; (2) at least two symptomatic AF attacks recorded by electrocardiogram (ECG) within 6 months before enrollment, and the duration of AF was less than 1 year. In addition, at the time of enrollment, abnormal AF rhythm could be converted to normal sinus rhythm independently or by medical intervention. However, drug and/or electrical cardioversion must be performed between 6 months and 8 weeks before enrollment, and no cardioversion was undertaken in the last 8 weeks; (3) did not take angiotensin converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs) and/or statins for more than 2 weeks before enrollment, or taking the above drugs, but can withstand the washout period of more than 2 weeks after discontinuation; (4) aged between 25 to 79 years; and more than 80 U/L, aspartate aminotransferase (AST) more than 80 U/L or creatinine (Cr) more than 1.5 mg/dL; (10) severe hyperlipidemia needed essential treatment with statins/bates defined as total cholesterol (TC) more than 6 mmol/L or low density lipoprotein cholesterol (LDL-C) more than 3.36 mmol/L; (11) Other ARB, ACEI and/or statins have been taken and cannot be washed for more than 2 weeks; (12) contraindications for statins and/or ARBs; (13) pregnant/planned pregnant women or lactating women; (14) the possibility of poor compliance or noncooperation during holistic follow-up; and (15) inappropriate age.

| Group design and therapy schedule
Participants were randomly assigned to be treated either with dihydropyridine CCB (CCB group, n = 45), CCB combined with fluvastatin (CCB + statin group, n = 48), valsartan (valsartan group, n = 46), or valsartan combined with fluvastatin (valsartan + statin group, n = 50). Patients' medical history, medication history (including drug type, daily dose, starting time and ending time of medication) and results of physical examination were recorded before drug intervention. Furthermore, resting ECG, 7-day electrocardiogram (7-day Holter), UCG, related blood routine, whole blood biochemistry and other laboratory examinations were also tested. In this study, valsartan (Diovon, Beijing Novartis Pharmaceutical Co., Ltd.) was recommended to take in the morning every day as an initial dose of 80 mg; fluvastatin (Lescol, Beijing Novartis Pharmaceutical Co., Ltd.) had an initial dose of 40 mg, once a day, before going to bed; and dihydropyridine CCB was only allowed to use amlodipine, nifedipine or felodipine. After patients entering the group, the dosage of each group could be adjusted according to the clinical situation. However, all the patients must meet the standard of blood pressure control as 140/90 mmHg. The recommended dosage of valsartan and fluvastatin were ≥ 80 mg and ≥ 40 mg, respectively.

| Data collection and follow-up
Subjects filled in their case report form at the beginning of the trial, after that their gender, age, body mass index (BMI, = weight/ height2, kg/m2), history of diabetes, stroke, smoking or drinking, UCG, resting ECG, 7-day Holter as well as blood routine, blood lipid, myocardial enzyme, liver and renal function and other blood biochemical laboratory test results were recorded in detail. The combined medication and adverse events were observed within 24 months of follow-up. During this period, all the selected patients were followed up once every 3 months to review blood pressure, ECG, and cardiac function classification (NYHA). In addition, patient logs, disease changes, new combined drugs, adverse events and doctor's recommended prescriptions were also recorded. At the 24th month, UCG, related blood routine test, whole blood biochemistry and other laboratory tests were reexamined for comparison. Holter was reexamined at the 6th, 12th and 24th months. For  SPSS 20.0 was utilized to analyze the data. The measuring data of near normal distribution were described by mean standard ± deviation and the counting data was expressed by percentage. The comparison of changes in before and after treatments in each group was made by Student's t test and multiple comparisons of groups were made by one-way ANOVA followed by Bonferroni correction. P values below .05 were considered significant.

| RESULT
Eleven participants quit the study during the observation and treatment. In CCB group, there were six patients withdrew from the trial because of intolerable edema after taking CCB drugs (n = 2), prolonged QT interval more than 0.5 s (n = 2) or severe sinus bradycardia (n = 2). Only one patient withdrew since liver dysfunction in CCB + statin group. In valsartan group, two cases quit due to poor blood pressure control that needed to use other antihypertensive drugs except ARB (n = 1) or abnormal renal function (n = 1). In valsartan + statin group, one patient quit because of AMI while the other patient withdrew since severe sinus bradycardia. After the above patients were excluded, 41, 45, 44, and 48 patients in CCB group, CCB + statin group, valsartan group and valsartan + statin group were available for analysis. According to the rules of Statistics, the proportion of exfoliated cases was acceptable so that it would not make a significant negative impact on the research results.

| General clinical data
Finally, a total of 178 HTN patients with nonpermanent AF were followed up, including 94 males and 84 females, aged 44 to 77 years (mean 67.06 ± 8.28). There were 28 patients with diabetes, 23 patients with stroke, 74 patients with smoking history, and 51 patients with drinking history. There was no significant difference in age, gender, BMI, smoking or drinking history, diabetes, stroke and lung disease history among the four groups (P > .05). No significant difference was found in mean SBP and DBP among the four groups (P > .05). In addition, at the end of follow-up, the blood pressure of all the four groups was lower than that before treatment. However, the difference was not statistically significant, and there was no significant difference between the groups (P > .05).

| Drug applications
After adjustment of medication by follow-up doctors, no significant difference was observed in the use of antiplatelet drugs, anticoagulant drugs and nitrates drugs in each group during the follow-up period (P > .05, Table 1).

| Biochemistry and UCG
After 24 months treatment, the levels of serum TC, triglyceride (TG) and LDL-C in CCB + statin group and valsartan + statin group were obviously decreased and were significantly lower than those in CCB group and valsartan group (P < .05). At the beginning of enrollment, there was no significant difference in the inflammatory index (high-sensitivity C-reactive protein, hs-CRP), liver and renal function, myocardial enzyme, N-terminal brain natriuretic peptide (NT-proBNP) or blood lipid level among the groups (P > .05). After 24 months of treatment, the levels of hs-CRP (mg/L) in CCB + statin group, valsartan group and valsartan + statin group were lower than those before treatment (3.40 ± 1.05 vs 6.85 ± 1.27, 3.39 ± 1.06 vs 6.76 ± 1.31, 3.29 ± 1.02 vs 6.94 ± 1.03, respectively; all P < .01), and were significantly lower than that in CCB group (P < .05). However, no significant difference was noted in hs-CRP between CCB + statin group, valsartan group and valsartan + statin group (P > .05). In addition, the level of NT-proBNP (ng/L) in valsartan + statin group was significantly decreased after combined treatment (315.44 ± 50.45 vs 290.35 ± 36.13; P < .05). There were no significant changes in liver or renal function and myocardial enzyme level in each group compared with those before treatment, and no significant difference was observed between the groups (P < .05, Table 2). valsartan + statin group were significantly reduced after medication (P < .05). It is worth mentioning that the values of IVST, LVEDd and LAD in valsartan + statin group were significantly decreased after 24 months combined intervention (7.69 ± 2.07 vs 11.03 ± 1.98, 39.42 ± 9.22 vs 50.01 ± 6.89, 32.26 ± 2.23 vs 37.91 ± 3.13, respectively; all P < .05). Although LVEF of each group was higher than before, no significant difference was observed (P > .05, Table 2).

| f-f interval and f wave amplitude
In all the four groups, 7-day holter illustrated that there was no significant difference in the f-f interval or f wave amplitude at the beginning of the trial (P > .05) while reexamined 7-day holter demonstrated the f-f interval was gradually shortened and f wave amplitude was gradually increased at 6, 12, and 24 months after corresponding treatment.
In detail, reexamined holter indicated that the f-f intervals of CCB + statin group, valsartan group and valsartan + statin group were significantly longer than that of CCB group (P < .05) while no significant difference between CCB + statin group and valsartan + statin group was noted (P > .05). However, there was no noticeable change of the f-f interval in CCB group after drugs intervention (P > .05). Similarly, the amplitudes of f-wave in CCB + statin group, valsartan group and valsartan + statin group were significantly higher after 24 months treatment (P < .05) while no significant difference was found in CCB group (P > .05). In addition, in CCB group, the f-f interval moderately shortened while f wave amplitude gradually decreased during the follow-up period. Nevertheless, in valsartan group and valsartan + statin group, their f-f intervals extended gradually while their amplitudes of f-wave increased slightly in the same period. Furthermore, the amplitudes of f-wave and f-f intervals in valsartan group and valsartan + statin group were significantly higher and longer than those in CCB group after 24 months of treatment (P < .05). Interestingly, holter suggested the f-f interval of the 12th month in CCB group was significantly shorter than those before treatment, 6 or 24 months after treatment, therefore the specific reasons for this abnormal phenomenon need to be further studied by expanding the sample (Table 3).

| AF load and mean heart rate
After 24 months of grouping treatment, the AF loads of the four groups were higher than those before medication, but the levels of AF load in CCB + statin group, valsartan group and valsartan + statin group were significantly lower than that in CCB group (55. 26

| DISCUSSION
Cardiogenic thrombus caused by atrial fibrillation (AF) accounts for about one third of the causes of ischemic stroke. Stroke may even attack before the patient is diagnosed with AF. 10 The high incidence T A B L E 1 Comparison of medication between four groups (n, %) of thromboembolism in AF significantly increased the morbidity and mortality of patients and further increased the economic burden and affected the quality of life. 11 Therefore, it has important clinical significance to strengthen the control of AF risk factors and delay the pathologic progression of paroxysmal AF to persistent or permanent AF with the social structure of population aging worldwide. The conventional clinical risk factors for increased incidence rate of AF include age, valvular heart disease, HTN, HF, obesity, diabetes, alcohol consumption, etc. 12 HTN, especially excessive SBP, is an independent risk factor for AF, 13,14 while RAAS may be closely related to the occurrence and development of AF in patients with HTN. 15 In addition, our previous studies also confirmed that inhibition of RAAS is beneficial to prevent AF attack, suggesting that RAAS may play an important role in AF atrial remodeling. 16,17 To block RAAS by ARBs has a beneficial effect in reducing the incidence of AF. 18  Statins, as secondary prevention drugs for coronary heart disease, can not only significantly reduce blood lipid, but also effectively decrease the incidence and mortality of cardiovascular events by regulating serum lipid levels and stabilizing artery plaque. 23 In recent years, adding statins as the optimal treatment has been proved to be beneficial to AF. 24 Statins can significantly improve the prognosis of patients with AF and independently reduce the risk of all-cause death. 25 Hung et al. 26 observed that the statins can significantly reduce the risk of new-onset AF in elderly HTN patients those aged ≥64 years, and the patients with CHADS2 score ≥ 2 benefited more than those with a score of 1. In addition, Ma et al. 27 confirmed that statins can reduce the risk of AF in the elderly aged ≥65 years by about 19%. Furthermore, Kunt et al. 28 showed that atorvastatin can significantly reduce the incidence of AF and the mortality of cardiovascular events in those elderly patients who had underwent coronary artery bypass grafting (CABG) before. All the above researches suggested that statins can play an antiarrhythmic effect that independent from regulating blood lipids. Similarly, our results showed that compared with CCB group, the level of hs-CRP in CCB + statin group was significantly decreased after treatment with fluvastatin (P < .01); the level of hs-CRP in valsartan group after treatment with valsartan was also decreased, but the trend in valsartan + statin group with fluvastatin was more obvious ( Table 2). The antiarrhythmic function of statins is deemed to be closely related to its beneficial effect on anti-inflammatory and anti-oxidative stress response, reducing active oxygen radical production and improving atrial remodeling caused by the rise of AngII, so as to reduce the recurrence and to delay the progression of AF. 29 Our previous study has also shown that fluvastatin is beneficial for maintaining normal sinus rhythm in patients with nonpermanent AF. 30 Furthermore, Inohara et al. 31 suggested that the progression and recurrence of AF are closely related to BNP level. In this study, we showed that compared with using valsartan alone, the combination of valsartan and fluvastatin can improve the heart structure as a significant decrease in NT-proBNP (P < .05). The combined treatment also improved inflammatory response and maintained heart rate F I G U R E 1 Comparison of four groups of drug mechanism variability with normal sinus rhythm. These results indirectly confirmed the role of statins in the prevention and treatment on AF.
It is worth noting that in this study, we innovatively investigated the different effects of four groups drug interventions on the f-f interval and amplitude of f-wave. The shortening of the f-f interval that recorded by ECG can be used as an effective index to predict the prolonged duration of paroxysmal AF and to reflect the electrical remodeling of atrium while the down regulated changes in amplitudes of f-wave may be related to the recurrence of AF. It is apparent to find in valsartan + statin group, the trends of prolongation in f-f interval and increase of f wave amplitude were more obvious.
Compared with heart rate, which is a common index, electrocardiographic indexes f-f interval and amplitude of f-wave can interpret the beneficial effect of combined application of valsartan and fluvastatin on HTN patients with nonpermanent AF in a deeper level.
As mentioned above, existing interventions cannot completely cure AF, and paroxysmal AF always progresses to permanent AF. Our results were consistent with the above consensus. After 24 months of follow-up, the heart rate of the four groups decreased, but the AF load increased significantly. However, the AF load in valsartan group and valsartan + statin group was significantly lower than that in CCB group after treatment (all P < .01,  Table 2). These results suggest that the protective effect of valsartan combined with fluvastatin on AF load may be achieved by improving atrial remodeling ( Figure 1).
There are also some limitations in this study. First, although this study is a multicenter clinical study, there are few selected cases, and the follow-up time is short, which affects the persuasion of the test results. Further larger randomized clinical trial will be necessary to confirm the conclusions of our study. Secondly, this study did not assess the effect of application of fluvastatin alone on AF, and failed to directly prove whether fluvastatin can prevent AF independently.
At last, although 7-day Holter can capture most AF episodes, the potential subclinical AF may be underestimated.