The effect of autoantibody against M2-muscarinic acetylcholine receptor in peripartum cardiomyopathy patients on digoxin additional to standard treatment

Background: To evaluate the effects of autoantibodies against the M2-muscarinic receptor (anti-M2-R) on digoxin additional to standard treatment in peripartum cardiomyopathy (PPCM) patients. Methods: 86 PPCM patients were separated into anti-M2-R negative or positive group according to the anti-M2-R reactivity. All the patients received digoxin additional to standard treatment regimen. Echocardiography was performed at baseline and after 5 years treatment. Serum digoxin concentration (SDC) were performed every 3 to 6 months. All-cause mortality, cardiovascular mortality and re-hospitalization for heart failure were compared after 5 years of follow-up. Results: There were 82 patients completed the nal data analysis, including 38 in the anti-M2-R (+) group and 44 in the anti-M2-R (-) group. The heart rate of the positive group was higher than that of the negative group at baseline (102.3 ± 6.3 vs. 95.9 ± 6.8, p < 0.001). The initial SDC of patients in the positive group was higher than that of patients in the negative group with the same dose of digoxin (1.21 ± 0.41 vs. 0.73 ± 0.16 ng/mL, p < 0.001). Patients in the anti-M2-R (-) group had better tolerance to metoprolol and digoxin (p < 0.05). All the PPCM patients showed prominent improvement in cardiac function, especially in the anti-M2-R (-) group. Re-hospitalization for heart failure was decreased in the negative group, but not of all-cause or cardiovascular mortality. Conclusions: Patients negative for anti-M2-R showed better tolerance to metoprolol and digoxin. Anti-M2-R maybe a predictor for vagus nerve overactivation and is associated with poor response to digoxin treatment in PPCM patients.

R maybe a predictor for vagus nerve overactivation and is associated with poor response to digoxin treatment in PPCM patients.

Background
Peripartum cardiomyopathy (PPCM) is a rare idiopathic dilated cardiomyopathy de ned by the signs and symptoms of heart failure (HF) in the last month of pregnancy through the fth month postpartum [1].
The de nition of PPCM states that there must be no previously known structural heart disease, and echocardiographic parameters must achieve one of the following: left ventricular ejection fraction (LVEF) < 45%, fractional shortening < 30%, or both, with a possible additive left ventricular end diastolic dimension > 2.7 cm/m 2 body surface area [2]. This disease is associated with a high morbidity and mortality but its aetiology remains unknown [1].
The main pathogenesis of HF are ventricular remodeling and the imbalance of endogenous neuroendocrine hormone systems. The excessive activation of the renin-angiotensin-aldosterone system (RAAS) and the sympathetic nervous system has been recognized by most doctors. Angiotensinconverting enzyme inhibitors (ACEIs), β receptor blockers, and mineralocorticoid receptor antagonists have become the golden triangle for HF. However, the importance of the vagus nervous system is relatively unfamiliar. The M2-muscarinic acetylcholine receptor is the most important receptor for the vagus nervous system to act on the heart. Autoantibodies against the M2-muscarinic receptor (anti-M2-R), have been found in HF patients of various etiologies including PPCM [3][4][5], which can interfere with radioligand binding on the target receptor, display agonist-like activities on the M2 receptors, and hence modulate cardiac function [6].
Digitalis is a positive inotropic agent, also has the function of vagus nervous system stimulation. Digoxin is the most commonly used oral digitalis in HF patients. So it could neutralize the over activation of the sympathetic system and RAAS, which is similar to the effect of anti-M2-R. Therefore, what is the clinical sense of anti-M2-R? Are there any different responses to digoxin between patients negative and positive for anti-M2-R? If so, do patients negative for anti-M2-R have better improvement in cardiac function? In this study, we investigate the presence of anti-M2-R and cardiac function in response to digoxin additional to standard treatment regimen for HF (ACEI, β receptor blocker, furosemide and spironolactone) in PPCM patients.

Study population
This was an prospective observational study, which began in January 1998 and ended in December 2019. A total of 86 consecutive newly diagnosed PPCM patients were enrolled at the HF clinic of Beijing Chaoyang Hospital. We obtained the demographic data and related information by in-person interview using the structured questionnaire. The inclusion criteria were as follows: (1) age, between 18 and 40 years old, (2) cardiac function, New York Heart Association functional classes (NYHA) II-IV, (3) symptoms of HF, happened in the last month of pregnancy or during the rst 5 months postpartum, (4) no other identi able causes for HF, and (5) LVEF < 45% by transthoracic echocardiography. Exclusion criteria were as follows: (1) clinical conditions with increased levels of autoantibody, such as rheumatoid arthritis, HIV, and evidence for sepsis, (2) moderate-severe anemia, (3) metabolic disorders such as thyroid disease, or (4) moderate-severe hepatic or renal dysfunction. The study protocol complied with the Declaration of Helsinki and was approved by the Ethics Committee of Beijing Chaoyang Hospital, Beijing, China. All the PPCM patients provided written informed consent before study entry.
Serum anti-M2-R detection About 2 mL of blood was withdrawn from the antecubital vein of each patient when enrolled into this study and after ve years treatment. Serum samples were separated by centrifugation at 3,000 rpm for 10 min and stored at -20 °C. Peptide corresponding to the sequence of the second extracellular loop of human M2-muscarinic receptor (amino acid sequence number 169-193: was synthesized by Genomed (Genomed Synthesis, Inc., San Francisco, CA, USA) with the solid-phase method of Merri eld. The purity of the peptide was 98%, on the basis of HPLC analysis on a Vydac C-18 column. Levels of serum anti-M2-R was measured with SA-ELISA and positive was de ned as a ratio of (sample A -blank A)/(negative control A -blank A) ≥ 2.1 [7]. Titers of autoantibodies was the highest when this ratio ≥ 2.1 with serum diluted from 1:20 to 1:160. The coe cient of variation of intraassay and inter-assay were less than 5%.
Beginning of the standard pharmacological regimen All the patients received digoxin additional to standard therapy regimens (perindopril or losartan, metoprolol, furosemide, and spironolactone). Perindopril was taken at an initial dose of 2 mg/day, and then uptitrated according to the blood pressure. If perindopril wasn't tolerated, losartan was used instead. Metoprolol was taken at an initial dose of 12.5 mg/day that was up-titrated over a 2-4-week period by doubling the twice-daily amount to the maximum tolerated dose or a target of 100 mg/day [2]. The maximum tolerated heart rate and blood pressure were 60-75 bpm and 120/65 ± 10/5 mmHg, respectively. The initial dosage of furosemide was 10-20 mg/day, which allowed to increase if a patient displayed signs or symptoms of HF progression. The dosage of spironolactone was 10-20 mg/day. Digoxin was taken at an initial dose of 0.125 mg/day and then adjusted according to the serum digoxin concentration (SDC). The target SDC was 0.5-0.9 ng/mL as suggested by the ACCF/AHA guideline for the management of HF [8]. If the SDC was 0.9-1.5 ng/mL, the dosage of digoxin was reduced to 0.0625mg/day. The dosage of digoxin was reduced to 0.0625mg every other day when the SDC was higher than 1.6 ng/mL. In addition, patients were advised to control their salt intake and body weight.

Follow-up examination
All patients were assigned to a xed investigator for up to 5 years of follow-up after they were included in the study. The primary endpoint events were all-cause mortality, cardiovascular mortality, and rehospitalization for HF. Patients received follow-up once a month for the rst year and every 3-6 months for up to 5 years or until the primary endpoint. Echocardiography, 6-minute walk tests, and clinical laboratory tests including SDC were performed regularly. We collected heart rate, blood pressure, body weight, cardiac function, the presence of peripheral edema, drug dosages during the examinations.
Subjects were also questioned and examined for the presence of any adverse drug reaction.

Statistical methods
Quantitative data are presented as mean ± SD, and categorical data are presented as percentage. Titers of serum anti-M2-R was reported as the geometric mean. For two groups comparison, we used Student's t test or Mann-Whitney U test for continuous variables, and the chi-square test or Fisher's exact test for categorical variables. Chi-square statistics and log-rank test were used for all-cause mortality, cardiovascular mortality, and hospitalization for HF. Data on the titration of metoprolol were t to a variable slope sigmoidal equation (Y = Initial Dose + (Maximum Dose − Initial Dose)/(1 + 10(LogEC50 − X)* Slope)), in which the independent variable (X) is the log of the time of the dosage value (Y). The LogEC50 denotes the time that corresponds to halfway between the minimum and maximum dosages. All the tests were 2-tailed. P < 0.05 was considered to be statistically signi cant. Data were analyzed using SPSS 20.0 (SPSS, Chicago, Illinois, USA).

Study characteristics
All the 86 patients were diagnosed as PPCM for the rst time. 30 patients were primiparous, and 25 patients had multiple gestation. There were 30 patients who had pregnancy-induced hypertension, and 18 patients had gestational diabetes mellitus. There were 44 patients with symptoms in the postpartum period. At baseline, 33 patients were in NYHA functional class II, 38 patients were in class III, and 15 patients were in class IV.
According to the anti-M2-R reactivity, 46 patients were assigned to the negative group and the other 40 patients were assigned to the positive group. The baseline characteristics of the two groups were shown in Table 1. Four patients lost to follow-up during the rst year (2 patients in the negative group and 2 patients in the positive group), and the other 82 patients completed the nal data analysis, including 38 patients (38/40, 95.0%) in the positive group and 44 patients (44/46, 95.7%) in the negative group. Of all the parameters, only the mean resting heart rate of the negative group was higher than that of the positive group. We posit that anti-M2-R maybe in uence heart rate via the activation of the vagus nervous system.

Drug dosage
All the patients received digoxin and standard pharmacological regimen for HF, which includes perindopril, metoprolol, spironolactone, and furosemide. There were no differences between patients with anti-M2-R (-) and those with anti-M2-R (+) regarding the dosage of perindopril, spironolactone and furosemide. The dosages of metoprolol and digoxin in the negative group was higher than that in the positive group, shown in Table 2. There were no differences between patients with anti-M2-R (-) and patients with anti-M2-R (+) regarding the dosages of perindopril, spironolactone and furosemide (all p > 0.05). The mean dose of metoprolol and digoxin for anti-M2-R (-) patients was higher than that for anti-M2-R (+) patients (p < 0.001).  anti-M2-R was 9.0% (7/78) and the mean titers of anti-M2-R was 1:59, which were signi cantly decreased compared to baseline (all p < 0.001) (Fig.2).

Adverse events
There were no obvious effects on blood glucose, serum lipid, or hepatic and renal function during the 5year treatment and follow-up. Only one patient in the positive group showed symptom of digoxin intoxication such as somewhat weakness and nausea. The SDC of this patient was 2.5 ng/mL, and the dosage of digoxin was reduced to 0.0625 mg every other day, and the above symptoms disappeared subsequently.

Cardiac function and 6-minute walk test
Clinical data, NYHA functional class, echocardiographic results, and 6-minute walk distance at baseline and 5-year were determined, shown in Table 3. With long-term low dose of digoxin additional to standard pharmacological treatment for HF, the left ventricular end-diastolic diameter (LVEDD) and left ventricular end-systolic diameter (LVESD) decreased from 58.7 ± 7.8 to 47.5 ± 4.5 mm and 47.2 ± 7.3 to 33.1 ± 5.5 mm in the anti-M2-R (-) group, respectively, and from 58.9 ± 7.0 to 49.9 ± 4.7 mm and 46.9 ± 6.9 to 36.2 ± 6.5 mm in the anti-M2-R (+) group, respectively. Meanwhile, the LVEF increased from 39.4 ± 4.4 to 62.7 ± 6.8% in the anti-M2-R (-) group and from 38.9 ± 6.1 to 57.7 ± 7.4% in the anti-M2-R (+) group. It is worth noting that the improvement of the structural and functional of left ventricular including 6-min walk distance in anti-M2-R (-) patients was better than that in anti-M2-R (+) patients. The LVEF returned to normal in 91% (71/78) of patients with ve years of treatment. Laboratory data, including levels of hemoglobin, creatinine, glutamic pyruvic transaminase, and potassium were stable throughout the followup.

Digoxin and HF
Digoxin is a traditional drug for HF treatment for more than 200 years. It enhances LVEF by inhibiting the Na + -K + -ATPase activity on the cardiomyocytes and enhancing myocardial contractility. The effects of digoxin on multiple neurohormones imply that long-term treatment with digoxin can provide bene ts to HF patients through inhibiting the Na + -K + -ATPase activity in the afferent bers of the vagus nerve, thus increasing the activity of vagus nervous system and inhibiting the sympathetic activity [9]. Digoxin can also increase the sensitivity of the carotid sinus baroreceptor to further reduce the activity of the sympathetic and RAAS [10].
It is demonstrated that digoxin was able to improve maximum exercise endurance of patients with mildmoderate stable HF and reduce acute exacerbation of HF whether the basic treatment used diuretics alone or diuretics in combination with ACEIs [11][12][13]. Subsequent studies on digoxin demonstrated that additional use of 0.25 mg/day digoxin on the basis of diuretics and ACEIs could reduce the all-cause and HF exacerbation-induced hospitalization, although it could not reduce the overall mortality of HF patients [14]. A meta-analysis reported that digoxin could reduce the hospitalization rate and relieve clinical symptoms, and it is believed that HF patients could still bene t from the use of digoxin [15].
It is important to note that the SDC was closely correlated with the prognosis of HF. Ahmed et al [16] reported that 0.5-0.9 ng/mL of SDC could reduce the mortality, all-cause hospitalization and HFassociated hospitalization, while SDC ≥ 1.0 ng/mL could only reduce the HF-associated hospitalization without impact on the mortality. Similarly, Rathore et al [17] also demonstrated that 0.5-0.8 ng/mL of SDC could reduce the HF-associated mortality. Low-dose of digoxin (≤ 0.125 mg/day) was the strongest predictive factor of the low SDC [16]. Digoxin should be added to maintain the SDC of 0.5-0.9 ng/mL in the early stage of treatment in patients with severe HF, or those whose symptoms are still present after ACEIs and β receptor blockers have begun working [8].
Xu et al [18] have recruited a total of 756 chronic HF patients with reduced LVEF. All the patients received digoxin additional to standard treatment regimen. The SDC was maintained at 0.5-0.9 ng/mL. A clinical follow-up for up to 15 years was performed and it demonstrated well safety and e cacy to long-term low dose of digoxin. Hou et al [19] found the SDC was higher in the anti-M2-R (+) group compared to anti-M2-R (-) group with similar dose of digoxin in chronic HF patients, similar to this study. In this study, all the PPCM patients reacted well to long-term low dose of digoxin treatment. Symptoms of digoxin intoxication occurred when the SDC was signi cantly elevated, almost about 2.5 ng/mL. Regularly testing of the SDC was essential when using digoxin.
Anti-M2-R was detected in the idiopathic dilated cardiomyopathy (IDCM) patients rstly [7]. It could induce ventricular enlargement and thinning of the walls, the typical changes of IDCM and PPCM in humans, by monthly immunization peptides in accordance to the sequence of the second extracellular loop of the M2 receptor in rabbits [22,23]. Gimenez et al [24] have immunized mouse by using plasmid DNA encoding entire M2 receptor proteins, which leads to cardiac remodeling and contractile dysfunction.
Ribeiro et al [25] found immunization with plasmids encoding M2 receptor epitopes impairs cardiac function in mice and induces autophagy in the myocardium, indicating novel roles for the anti-M2-R. Our previous study suggested anti-M2-R not only existed in IDCM patients, but also in HF caused by different causes, including PPCM. We posit that serum anti-M2-R may be related to cardiac structural and functional changes, which need further studies to clarify [3].
Regulation of the autonomic system has an important in uence on the progression of PPCM. Elevated activities of sympathetic system is associated with an adverse prognosis. Activation of the vagus nervous system seems to be a double-edged sword. With the extensive use of ACEIs and β rceptor blockers, activities of sympathetic system was attenuated mainly, but the effects was poorly to the standard pharmacological regimen in a number of patients. For these patients, the tension of vagus nervous system may be activated pathologically. In anti-M2-R (+) patients, the chronic interaction between anti-M2-R and the M2 receptor causes a pathological activation of cardiac vagus nervous system. Therefore, these patients showed a slower heart rate and lower maximum tolerated dose of metoprolol and digoxin compared to anti-M2-R (-) patients. In the anti-M2-R (-) group, the enlarged left ventricular chamber could return to normal more rapidly, which maybe partially referred to different dosages of metoprolol and digoxin between the two groups.
The prognosis of PPCM is better than that of other cardiomyopathies with reduced LVEF. We found the LVEF of 71 patients (91.0%) return to normal at 5 years, similar to previous studies [26,27].

Declarations
Ethics approval and consent to participate: The present study was performed in compliance with the principles outlined in the Declaration of Helsinki and was approved by the Ethics Committee and the Prescription and Therapeutic Committee of Beijing Chaoyang Hospital, Capital Medical University (Beijing, China). All the patients provided written informed consent prior to enrolment.
Consent for publication: The authors con rm that written consent from each patient has been obtained to publish the manuscript.
Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no competing interests.
Funding: This project was supported by the National Science and Technology 863 Key Project Foundation (2002BA711A07), the Natural Science Foundation of China (81370340) of Lin Zhang. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Authors' contributions: LZ and XYL conceived and designed the study and were responsible for performing data collection, analysis and interpretation and giving nal approval of the version to be published. GLM was responsible for drafting the manuscript and revising it critically for important intellectual content, and made substantial contributions to follow-up of the patients and analysis of data. YY and YDW performed the echocardiographic examination and were responsible for follow-up of the patients. WS and YSL were responsible for collection of patient and laboratory data, data interpretation and revision of the manuscript regarding content. LC and TYL were responsible for performing the study and the data analysis. FS and CS performed the ELISA for anti-M2-R. All authors read and approved the nal version of the manuscript for publication.