Baseline and changes in prognostic nutritional index associate with heart failure hospitalization and all-cause death in patients with cardiac pacemaker

Abstract

Object

The study evaluated prognostic nutritional index (PNI) levels and changes as predictors of heart failure hospitalization (HFH) and mortality in cardiac pacemaker patients.

Methods

PNI was calculated on admission and at the end of the 1-month follow-up, and their net changes (Δ) were calculated by PNI at follow-up minus the corresponding PNI on admission. The optimal cutoff value of baseline PNI was determined by the ROC curve. Based on this optimal cutoff value or the change in PNI during follow-up, patients were divided into high-PNI and low-PNI groups, or into improved PNI group (ΔPNI > 0) and deteriorated group (ΔPNI ≤ 0). The crude and adjusted cox proportional hazard models were used to analyze the associations between adverse events and PNI or ΔPNI. Restricted cubic splines were used to model the association of the PNI-endpoint events.

Results

A total of 927 patients were enrolled in the study. The risk of HFH and all-cause death remained significantly higher in patients with low PNI, even after adjusting for other risk factors (hazard ratio [HR]: 1.997, 95% confidence interval [CI]: 1.220 - 3.203, P = 0.006; HR: 2.501, 95% CI: 1.392 - 4.494, P = 0.002, respectively). Patients with ΔPNI ≤ 0 faced higher risks of HFH and mortality compared to those with ΔPNI > 0(HR: 3.146, 95% CI: 2.024 - 4.892, P < 0.001, HR: 2.082, 95% CI: 1.223 - 3.544; P = 0.007, respectively)

Conclusion

Low baseline PNI and ΔPNI ≤ 0 during follow-up effectively predicted HFH and mortality in cardiac pacemaker patients.

Clinical trial number

Not applicable.

Introduction

Bradyarrhythmia is caused by a variety of etiologies. Its main manifestation is an excessively slow ventricular rate, which leads to hemodynamic instability and a decrease in cardiac output [1]. As a result, patients may experience symptoms such as reduced exercise tolerance, syncope, and even sudden cardiac death. Currently, the implantation of a cardiac pacemaker is the most effective treatment for bradyarrhythmia. Despite the increasing maturity of pacemaker technology in recent years, approximately 30% of patients still experience adverse clinical events after PM implantation, such as heart failure [2]. In order to improve the prognosis of patients, it is urgent to find a factor that can predict the prognosis of patients with cardiac pacemaker.

Malnutrition had been reported that was associated with an increased length of hospital stay [3, 4], infections [5], and complications [6, 7]. The prognostic nutritional index (PNI) was calculated by multiplying the serum albumin concentration by the total counts of peripheral lymphocytes [8]. PNI is an index to evaluate the nutritional status and systemic immune function of patients. Compared with other scoring measures of nutritional status, such as Patient-Generated Subjective Global Assessment (PG-SGA) [9] and controlling Nutritional Status (CONUT) [10], the PNI appears to be more objective and accessible. Since Buzby et.al first proposed that PNI can effectively predict the risk of gastrointestinal surgery [11], the ability of PNI to predict adverse clinical events has been recognized in a variety of diseases, such as cancer [12,13,14], HF [15, 16], coronary artery disease (CAD) [17,18,19], and stroke [20, 21]. In recent years, studies have suggested that the decrease of PNI predicts poor prognosis [22,23,24]. Although previous studies hadconfirmed that low baseline PNI and decrease of PNI during the follow-up were associated with a higher risk of major adverse cardiovascular events (MACEs) [17, 24, 25], the prognostic ability of PNI for patients with cardiac pacemaker remains unclear. Therefore, the purpose of this study was to investigate the predictive ability of the prognostic nutritional index (PNI) and the change in PNI (ΔPNI) in the clinical prognosis of patients who underwent pacemaker implantation.

Methods

Study population

Patients with successful pacemaker implantation in the First Affiliated Hospital of Wenzhou Medical University from January 2012 to December 2018 were reviewed through the electronic medical record system. Adults who were discharged alive were enrolled in this study. Patients without blood routine examination and serum albumin indicators at the beginning day of operation and the end of the 1-month follow-up would be excluded. Pregnant women were rejected for inclusion in the study. Before pacemaker implantation, blood collection, 12-lead electrocardiogram (ECG), and echocardiogram are used to assess the clinical status. All patients meet the indications for pacemaker implantation, specifically as follows: atrioventricular block (AVB), sick sinus syndrome (SSS), atrial fibrillation (AF) with bradycardia, left bundle branch block (LBBB) with heart failure (HF), atrial fibrillation with atrioventricular node ablation (ANVA), etc.

The PNI was calculated by multiplying the serum albumin concentration by the total counts of peripheral lymphocytes. PNI was calculated as 10 × serum albumin (g/dL) +0.005 × total lymphocyte count (per mm3) [8]. Patients were divided into the high PNI group and the low PNI group based on the optimum cut-off value of baseline PNI according to the receiver operating characteristic (ROC) curve. According to the PNI change during the follow-up, patients were classified as either improved (PNI during the follow-up > baseline PNI) or deteriorated (PNI during the follow-up ≤ baseline PNI). Regarding the improvement and deterioration of PNI, we calculated the difference between the two PNI values obtained at admission and at the end of the 1 - month follow - up after PM implantation. Specifically, ΔPNI = PNI (after PM implantation) - PNI (before PM implantation). We defined the group with ΔPNI > 0 as the PNI improvement group and the group with ΔPNI ≤ 0 as the PNI deterioration group. For all groups, we conducted a power analysis to ensure that the minimum sample size for collection was met (Figure S1, S2).

Informed consent was signed by all patients. The present study complied with the Declaration of Helsinki and was approved by the ethical committee of the hospital (YS2022-306).

Data collection

Baseline characteristics of patients including gender, age, body mass index (BMI), lifestyle risk factors (including smoking, drinking), and concomitant disease were recorded. Laboratory indicators including routine blood tests, liver and kidney function indicators, and blood lipid routine were also recorded. Echocardiographic parameters including left ventricular ejection fraction (LVEF), left atrial diameter (LAD), left ventricular end-diastolic diameter (LVEDd), mitral regurgitation (MR), and tricuspid regurgitation (TR) were obtained from echocardiography. Echocardiograms were performed by two sonographers who were blinded to the present study. Venous blood samples were collected at the beginning day of cardiac pacemaker operation and the end of the 1-month follow-up to obtain the total count of peripheral lymphocytes and serum albumin concentration. ΔPNI was were calculated by PNI at follow-up minus the corresponding PNI on admission.

Study endpoints

The primary endpoint and secondary endpoint were hospitalization for heart failure (HFH) and all-cause mortality, respectively. These two endpoints were first analyzed as a composite endpoint and then analyzed as individual endpoints. HFH was diagnosed as new onset or worsening HF that could not be controlled with medications and required intravenous diuretics. All-cause death was considered death from all causes including death from cardiovascular disease as well as died in non-cardiovascular causes. If the patient was admitted more than once for HF, only the first admission was recorded. We tracked patients for end-point events through hospital electronic medical record systems and telephone follow-up until August 31, 2022.

Statistical analysis

The normality of continuous variables was analyzed using the skewness kurtosis normality test. All normally distributed continuous variables were presented as mean with standard deviation, whereas non-normally distributed continuous variables were expressed as median with interquartile range. Categorical variables were expressed as numbers and percentages. When the data meet the normality assumption, the independent samples t-test is used to compare the differences of continuous variables; if the data are not normally distributed, the Mann-Whitney U test is employed. Due to the large sample size, we use the chi-square test or Fisher's exact test to compare the differences between categorical variables. The optimal cut-off value of the baseline PNI was determined by the Youden index. And the Pearson correlation coefficient was used to verify the correlation between PNI or ΔPNI and other clinical parameters. The Kaplan Meier (KM) method was used to calculate the cumulative survival rate and HFH-free survival rate, statistical significance was tested using the log-rank test. The crude and adjusted Cox proportional hazards models were used for the analysis. The hazard ratios (HRs) and 95% confidence intervals (CIs) of the association between PNI or ΔPNI and adverse events were analyzed under the conditions of multiple covariates such as gender, age and BMI. Restricted cubic splines (RCS) were used to assess the nonlinear relationship between baseline PNI and HFH and all-cause mortality. Setting the reference point to where the risk of adverse events was equal to 1. The spline curve was defined using four nodes at the 5th, 35th, 65th, and 95th percentiles. A two-sided P < 0.05 was considered to be statistically significant. The project of R (version 3.5.1) and the IBM SPSS Statistics 25.0 were performed for the present study statistics.

Result

Baseline characteristics of the enrolled patients

We employed ROC curve analysis to assess the accuracy of baseline PNI in predicting heart failure hospitalization rate and all-cause mortality. When the AUC of baseline PNI was 40.45, the AUC was 0.621 (95% CI = 0.550–0.691, P < 0.001), with a sensitivity of 85.4% and a specificity of 37.8%. Therefore, we used a baseline PNI of 40.45 as the threshold to divide patients into high-PNI and low-PNI groups (Figure S3). Table 1 showed the basic clinical characteristics, biochemical and echocardiographic parameters on admission of the patients who were divided into two groups according to the optimal cut-off value of the baseline PNI and ΔPNI. The process of screening patients was illustrated in Figure 1. A total of 33 patients were excluded due to loss of follow-up or incomplete data. Finally, a total of 927 patients with cardiac pacemaker were included in this study, there were 570 males, accounting for 61.5% of the total population. The average age of the final population was 69.62 ± 10.67 years, and the mean baseline PNI of the total population was 46.10 ± 6.10. The mean PNI of the high and low PNI groups were 47.87 ± 4.86 and 36.99 ± 2.81, respectively (P = 0.011), respectively. Compared with the high PNI group, the patients in the low PNI group were older, and had poorer basic health status, while there was no difference in the prevalence of hypertension between the two groups. Patients in the low PNI group had significantly lower ejection fractions than those in the high PNI group. (50.48 ± 16.26 vs 55.79 ± 1 6.27, P < 0.001). Patients with increased PNI (ΔPNI > 0) during the follow-up were older and had lower baseline PNI compared to patients with deteriorated PNI (ΔPNI ≤ 0).

Table 1 Characteristics of the study population

Fig. 1

Schematic summary of study and patient flow. Abbreviation: PNI= prognostic nutritional index

Clinical Outcomes

During a mean follow-up of 46.58 ± 28.39 months, a total of 122 patients (13.2%) experienced HFH or all-cause death (Table 2). The incidence of HFH or all-cause death was significantly higher in the low PNI group compared to the high PNI group (28.5% vs 10.2%, P<0.001). Additionally, we found that patients with an increase in PNI during the follow-up period had a significantly lower likelihood of experiencing HFH or all-cause death compared to those with a decrease in PNI (9.9% vs 20.4%, P<0.001). Furthermore, HFH and all-cause death were also analyzed separately as independent endpoints, a total of 82 (8.8%) patients had HFH, and the incidence of HFH in the low PNI group was significantly higher than that in the high PNI group (19.9% vs 6.7%, P < 0.001). Interestingly, we found that the baseline PNI had a greater impact on the incidence of HFH in men (26.3% vs 6.4%, P < 0.001), while the incidence of HFH in women was similar in both the high and low PNI groups (7.7% vs 7.2%, P = 0.902) (Table 3). We also found that patients with increased PNI during the follow -up were significantly less likely to develop HFH than patients with deteriorated PNI (6.1% vs 14.9%,P < 0.001). The same results were found for the effect of baseline PNI on all-cause mortality (14.6% vs 4.8%, P < 0.001), however, the effect of ΔPNI on all-cause mortality was not statistically significant (5.3% vs 8.7%, P = 0.055). We recognized that patients who experienced HFH or all-cause death had a mean baseline PNI of 43.56±6.52, which was significantly lower than that of patients without adverse events (P<0.001). The mean baseline PNI of patients who experienced HFH and all-cause death were 43.80±6.94 and 42.82±5.84, respectively, both of which were also significantly lower than that of patients without adverse events (P=0.002 and P<0.001, respectively). In addition, patients who experienced adverse events had a decrease in PNI or only a slight increase during the follow-up period (Table 4).

Table 2 Cumulative incidence of clinical events

Table 3 Relationship between the prognostic nutrition index and gender

Table 4 PNI and change in PNI for adverse events

Analysis of the KM curve showed that the high PNI group had significantly higher rates of adverse event-free survival, HFH-free survival, and cumulative survival compared to the low PNI group (P<0.001). And patients with deteriorating nutritional status during the follow-up period had a higher incidence of HFH and all-cause than patients with improved nutritional status, without adjusting for other risk factors (HR: 2.604, 95% CI: 1.688 - 4.018, P < 0.001, HR: 1.758, 95% CI: 1.048 - 2.948; P = 0.033, respectively) (Figure 2).

Fig. 2

Kaplan–Meier survival analysis for adverse events stratified by PNI and ΔPNI. A Kaplan–Meier curves for free-events survival stratified by PNI. B Kaplan–Meier curves for free-events survival stratified by ΔPNI. C Kaplan–Meier curves for free-HFH survival stratified by PNI. D Kaplan–Meier curves for cumulative survival stratified by PNI. E Kaplan–Meier curves for free-HFH survival stratified by ΔPNI. F Kaplan–Meier curves for cumulative survival stratified by ΔPNI. Abbreviation: PNI= prognostic nutritional index, ΔPNI=change in prognostic nutritional index, HFH= heart failure hospitalization

Analysis of Cox proportional hazard models for clinical outcomes, Model 1 included only gender, age range, and BMI. Model 2 adjusted for gender, age range, BMI, and lifestyle risk factors (current smoking, current drinking). Model 3 adjusted for gender, age range, BMI, lifestyle risk factors, and baseline health conditions (diabetes, hypertension, chronic kidney disease, dyslipidemia, hyperuricemia, liver dysfunction, anemia). Model 4 adjusted for gender, age range, BMI, lifestyle risk factors, baseline health conditions (diabetes, hypertension, chronic kidney disease, dyslipidemia, hyperuricemia, liver dysfunction, anemia), and LVEF. After adjusting for multiple factors, it was found that the baseline PNI was the main variable of interest. The results showed that the risk of HFH and all-cause death was significantly higher in the low PNI group than in the high PNI group (HR: 1.977, 95% CI: 1.220 - 3.203, P = 0.006; HR: 2.501, 95% CI: 1.392 - 4.494, P = 0.002). In addition, after adjusting for gender, age, BMI, lifestyle risk factors, and comorbidities, ΔPNI was also a key variable. Compared with patients with ΔPNI > 0, patients with ΔPNI ≤ 0 had a greater risk of developing HFH and all-cause death during the follow-up (HR: 3.146, 95% CI: 2.024 - 4.892, P < 0.001; HR: 2.082, 95% CI: 1.223 - 3.544, P = 0.007). The final results showed that after adjusting for gender, age, BMI, lifestyle risk factors, comorbidities, and LVEF, both the baseline PNI and the ΔPNI were independent predictors of HFH and all-cause mortality in patients undergoing pacemaker implantation (Table 5).

Table 5 Cox proportional hazard models for clinical outcomes

Analysis of nonlinear relationship of PNI and adverse events, the spline curve clearly shows that when PNI is less than 46, HFH and all-cause mortality exhibit a continuous and relatively stable upward trend as the PNI value gradually decreases, indicating a significant negative correlation between PNI and HFH as well as all-cause mortality. However, when PNI is greater than 46, the curvature of the curve gradually decreases with the increasing PNI value, and the risk of adverse events does not change significantly (Figure 3).

Fig. 3

Relationship between baseline PNI and the risk of adverse clinical events. A: Relationship between baseline PNI and the risk of HFH. B: Relationship between baseline PNI and the risk of all-cause death. Abbreviation: PNI= prognostic nutritional index, HFH= heart failure hospitalization, HR= hazard ratio

Correlation of PNI

According to correlation analysis, the baseline PNI was associated with many clinical parameters, such as age, body mass index, and creatinine, but the correlation was not obvious. However, ΔPNI was only related to a few variables, such as age, hemoglobin, and blood lipids (Table 6).

Table 6 Correlation between prognostic nutritional index and clinical parameters

Discussion

This study focused on the effect of PNI and ΔPNI on the long-term prognosis of patients undergoing pacemaker implantation. The key findings of this article were as followings: (1) The basic health status of patients in the low PNI group was worse than that in the high PNI group; (2) The cumulative survival rate and HFH-free survival rate of patients in the high PNI group were significantly higher than those in the low PNI group; (3) Compared with patients with ΔPNI > 0, patients with ΔPNI ≤ 0 had a significantly higher risk of HFH and all-cause death; (4) When the PNI < 46, the risk of adverse events increased greatly as the PNI decreased, while the risk of adverse events does not change significantly when the PNI > 46; (5) Both PNI and ΔPNI were independent predictors of HFH and all-cause mortality in patients with cardiac pacemaker.

PNI as a nutritional screening tool was first reported by Buzby et al. [11], subsequently, Onodera et al. demonstrated that PNI was associated with postoperative complications, and concluded that the relationship between nutritional status and postoperative adverse events [8]. PNI was a combination of lymphocytes and serum albumin, reflecting the body's nutritional status, immune system and inflammatory response. Therefore, many studies have proposed that a low-cost, non-invasive tool, namely the nomogram based on PNI, can provide good prediction accuracy for cancer patients with breast cancer, colorectal cancer, etc., and provide them with individualized treatment strategies [26, 27]. Evidences had previously shown that low PNI is associated with poorer outcomes in patients with cardiovascular disease [16, 28,29,30]. Chen et al demonstrated that low PNI was independently associated with both short- and long-term overall mortality, and MACEs in patients with acute heart failure (AHF) [16]. We found that low PNI was associated with decreased LVEF. This finding may support cardiointestinal interaction as a cause of cardiac cachexia [31]. In patients with low LVEF, insufficient intestinal blood perfusion caused local edema of the intestinal tract, and abnormal intestinal mucosal function, resulting in poor absorption and subsequent inflammatory response. Therefore, patients had a low PNI. Conversely, low serum albumin concentration in patients with low PNI leads to fluid retention and insufficient immune response [32]. The predictive value of PNI may be related to the association of albumin and lymphocyte counts with adverse outcomes. Our results indicated that low PNI was accompanied by low serum albumin and low lymphocyte counts (Table 5). Previous evidences suggested that hypoalbuminemia remains an independent risk factor for poor prognosis in patients with cardiovascular disease, even after adjustment for common risk factors as well as prognostic markers [33]. Uthamalingam et al. proved that hypoproteinemia independently related to increased one-year mortality in patients with acute compensated HF [34]. Recent studies had shown that inflammation is the cause of cardiac hypertrophy, fibrosis and pyroptosis [35]. Additionally, studies had confirmed that low lymphocyte count was an independent risk factor for mortality in patients with advanced HF (HR: 0.942, 95% CI: 0.928 - 0.956, P < 0.001) [36].

We also found that patients with ΔPNI > 0 had significantly lower HFH and all-cause mortality after pacemaker implantation, regardless of the baseline PNI level. A study involving 141 patients with AHF demonstrated that elevated PNI during hospitalization was independently associated with a favorable outcome (HR= 0.30, 95% CI: 0.14 - 0.57, P = 0.0006). Peng et al. proposed that decreased postoperative PNI, rather than low baseline PNI, was an independent risk factor for overall survival and recurrent-free survival in patients with hepatocellular carcinoma [37]. These evidences suggested that it is necessary to improve the nutritional status of patients to prolong their survival. In our study, we additionally found that PNI may have a greater impact on men. Knowledge of the risk factors of adverse events is important not only to improve the clinical prognosis of patients, but also to optimize the treatment of patients.This study has clearly demonstrated that the baseline level and changes of the PNI are significantly associated with heart failure hospitalization and all - cause death in patients with cardiac pacemaker implantation. However, the practical implications of these findings in clinical practice need further exploration. Based on our results, patients with a lower baseline PNI or a significant decline in PNI may face a higher risk of adverse outcomes, so a stricter follow - up protocol for these high - risk patients is strongly recommended, like increasing the frequency of outpatient visits in the first few months after pacemaker implantation and conducting a comprehensive assessment at each visit, including physical examination, laboratory tests (e.g., repeated measurement of PNI - related parameters such as serum albumin and lymphocyte count), and electrocardiogram (ECG) monitoring to facilitate early detection of signs of heart failure exacerbation or other complications for timely intervention. For patients with abnormal PNI values, outpatient follow-up for pacemaker programming should be enhanced, and key indicators should be reviewed more carefully, including heart rate recording and the detection of arrhythmias such as atrial fibrillation and ventricular tachycardia. In addition, pacing algorithms should be adjusted. For example, the application of the RVPm algorithm may reduce the risk of developing persistent/permanent atrial fibrillation (PerAF), cardiovascular (CV) hospitalization, and HFH [38]. Although it is difficult to directly obtain accurate cardiac function indicators such as ejection fraction through remote monitoring, cardiac function can be indirectly evaluated by monitoring exercise tolerance and the degree of dyspnea. Monitoring fluid balance indicators such as body weight and blood pressure can help identify the early signs of fluid retention. Regarding infection prevention, before PM implantation, a comprehensive assessment of patients' underlying diseases like diabetes should be carried out and existing infection foci actively treated, and nutritional support including a nutrient - rich diet or supplements should be provided for patients with a low PNI; after PM implantation, the wound should be closely observed, kept clean and dry with regular dressing changes, and any bleeding or exudation promptly addressed, vital signs monitored as fever may signal infection for immediate treatment, and attention paid to indicators like blood routine and C - reactive protein as abnormal increases may suggest infection for early intervention, aiming to reduce the infection risk in patients with a low PNI through multi - step prevention and control.

Limitations

There were still some limitations in our research. This was a single-center, retrospective study with a limited number of patients. And we did not risk stratify according to PNI, we simply divided patients into a high PNI group and a low PNI group based on the optimal cut-off value of the baseline PNI. Our study was limited to the population with successful pacemaker implantation, and patients who died during hospitalization after pacemaker implantation were excluded. Due to the limited number of patients, the optimal cut-off value of PNI calculated by us may not be generalizable to a larger population. The impact of PNI changes on the prognosis of patients in our study was only one month after discharge, and the changes in the nutritional status of patients over a longer time were not explored. Meanwhile, the infection indicators after the implantation of the device were not included. This is also a very important outcome indicator, which helps to clarify whether the differences in PNI are related to postoperative infections. Currently, there are quite a few problems with pacemaker algorithms in safeguarding patients' health. These algorithms struggle to fully adapt to the complex medical conditions and lifestyles of patients [39]. For example, in patients with multiple chronic diseases, their unique physiological characteristics and differences in daily activities make it impossible for existing algorithms to precisely adjust pacing parameters. As a result, the treatment effect is suboptimal, affecting the recovery of patients' cardiac function and their quality of life.

Faced with complex arrhythmias and subtle changes in cardiac function, the algorithms are unable to identify them accurately and in a timely manner. This may delay the diagnosis of the disease, causing patients to miss the optimal treatment window, increasing the risk of cardiac disease deterioration, and in severe cases, endangering their lives. The multi - site pacing algorithms also have flaws. The setting of optimal parameters is not precise enough, there is insufficient personalized adaptation to different patients, and the flexibility is poor under complex cardiac conditions. This leads to limited effectiveness of cardiac resynchronization therapy and fails to fully improve the cardiac function of heart failure patients.

Future pacemaker algorithms will place greater emphasis on tailoring pacing strategies for each patient based on factors such as the patient's individual cardiac physiological characteristics, disease conditions, and lifestyle. Through comprehensive monitoring and analysis of patients' physiological parameters, such as cardiac electrophysiological properties, myocardial function, and exercise capacity, the algorithms will be able to automatically adjust parameters such as pacing frequency, pacing sequence, and pacing intensity [40]. This will achieve more precise and effective treatment, better meet the personalized needs of patients, and improve treatment outcomes and quality of life. In the future, more prospective studies are needed to demonstrate the impact of nutritional status of patients with cardiac pacemaker on prognosis.

Conclusion

Both baseline preoperative PNI and ΔPNI ≤ 0 during the follow-up were independent risk factors for HFH and all-cause death in pacemaker patients. Improving patients' nutritional status during hospitalization may prolong overall survival and complication-free survival. Therefore, the PNI calculated through routine biochemical and hematological examinations has the potential to serve as a convenient tool for the risk stratification of pacemaker patients and the implementation of nutritional interventions. This may contribute to more effective treatment of pacemaker patients. Certainly, it remains necessary for us to conduct prospective studies to further demonstrate the impact of patients' nutritional status on clinical outcomes.