Sympathetic dysfunction is associated with worse fatigue and early and subtle symptoms in heart failure: an exploratory sex-stratified analysis

Abstract

Aims

Physical symptoms impact patients with heart failure (HF) despite treatment advancements; however, our understanding of the pathogenic mechanisms underlying HF symptoms remains limited, including sex differences therein. The objective of this study was to quantify associations between sympathetic markers [norepinephrine (NE) and 3,4-dihydroxyphenylglycol (DHPG)] and physical symptoms in patients with HF and to explore sex differences in these associations.

Methods and results

We performed a secondary analysis of combined data from two studies: outpatients with HF (n = 111), and patients prior to left ventricular assist device implantation (n = 38). Physical symptoms were measured with the Heart Failure Somatic Perception Scale (HFSPS) dyspnoea and early/subtle symptom subscales and the Functional Assessment in Chronic Illness Therapy Fatigue Scale (FACIT-F) to capture dyspnoea, early symptoms of decompensation, and fatigue. Norepinephrine and DHPG were measured with high-performance liquid chromatography with electrochemical detection. Multivariate linear regression was used to quantify associations between symptoms and sympathetic markers. The sample (n = 149) was 60.8 ± 15.7 years, 41% women, and 71% non-ischaemic aetiology. Increased plasma NE and NE:DHPG ratio were associated with worse FACIT-F scores (P = 0.043 and P = 0.013, respectively). Increased plasma NE:DHPG ratio was associated with worse HFSPS early/subtle symptoms (P = 0.025). In sex-stratified analyses, increased NE:DHPG ratio was associated with worse FACIT-F scores (P = 0.011) and HFSPS early/subtle scores (P = 0.022) among women but not men.

Conclusion

In patients with HF, sympathetic dysfunction is associated with worse fatigue and early/subtle physical symptoms with associations stronger in women than men.

Graphical Abstract

Graphical Abstract.

Novelty.

• Sympathetic markers (i.e. norepinephrine and its main metabolite) were associated with fatigue and early/subtle symptoms in patients with moderate to advanced heart failure (HF).

• In sex-stratified analyses, worse sympathetic dysfunction was associated with worse fatigue and early/subtle symptoms in women but not men.

• With further clinical and research validation, a multidisciplinary clinical team may be able to use these biomarkers and pathways to implement targeted strategies to mitigate symptom burden in HF.

Introduction

Heart failure (HF) is a complex and heterogeneous syndrome that persists as a leading cause of morbidity and mortality.¹ Heart failure symptom burden is an important component of diagnosing HF and guiding HF classification and treatment strategy.² Patients with HF frequently present with burdensome physical and affective symptoms,³ which adversely affect quality of life and mortality risk.^4,5 Furthermore, exacerbation of symptoms represents a significant driver of health care utilization and rehospitalizations.^5,6 In particular, physical symptoms (e.g. dyspnoea, fatigue, and exercise intolerance) are commonly reported in HF;^3,7 yet, it remains poorly understood how physical symptoms relate to the underlying pathophysiological mechanisms of HF. Elucidating mechanisms that underpin physical HF symptoms could guide the development of targeted interventions to ameliorate symptom burden. Most prior symptom biology research has focused on conventional clinical markers (e.g. ejection fraction), but the general conclusion is that there is little-to-no association between these markers and symptoms.^8,9 Thus, symptom biology research has transitioned to examining more nuanced biochemical markers measured in the blood.^10–12

As one potential mechanism, sympathetic nervous system activation is a key component of HF pathophysiology as evidenced by substantially increased sympathetic outflow.¹³ When chronically upregulated, catecholamine excess leads to impaired left ventricular function, cardiac and vascular remodelling, cardiac injury and apoptosis, and arrhythmias.^14,15 These cardiac and peripheral effects may in turn contribute to the development and persistence of physical symptoms. Sympathetic function can be measured with plasma levels of the catecholamine norepinephrine (NE),¹⁶ which is a strong predictor of poor prognosis in HF.¹⁷ Norepinephrine is deaminated by monoamine oxidase into its primary metabolite, 3,4-dihydroxyphenylglycol (DHPG), a process that occurs following NE reuptake into sympathetic nerve endings. Under normal physiologic conditions, the majority of NE is taken up and metabolized into DHPG; however, NE reuptake is impaired in HF with significant NE spillover into plasma.^18–20 Thus, measures of NE and DHPG, including the ratio of NE:DHPG, provide insight into the sympathetic milieu. In small samples, we have previously demonstrated associations between sympathetic dysfunction and both symptoms¹² and health-related quality of life.²¹

One important consideration when exploring mechanisms of symptoms is potential sex differences.²² Evidence has shown that there may be physiologic and pathophysiologic sexual dimorphisms in sympathetic function; for example, women have increased cardiac-specific sympathetic activation in both HF and healthy conditions.²³ One hypothesis is that women have increased sympathetic activation to compensate for smaller left ventricular diameters.²³ However, potential sex differences in sympathetic function, and autonomic function more broadly, are nuanced and highly variable depending on the methodology used to assess sympathetic activity and contextual factors such as age.^24,25 Sex differences in HF symptoms are less clear, but evidence seems to indicate that there may be minimal differences.²⁶ However, the biological mechanisms underlying symptoms could be different between women and men, as we have previously shown.²⁷ Building on these signals, and to further advance this area, the purpose of this study was to (i) quantify associations between sympathetic markers and physical symptoms in patients with moderate to advanced HF and (ii) explore sex differences in these associations.

Methods

Study design and sample

This is a secondary analysis of data sourced from two different studies. The first study included patients with chronic stable HF enrolled from outpatient HF and general cardiology clinics at a single centre in the Pacific Northwest of the United States between May 2018 and February 2020 (i.e. ‘chronic HF study’).²⁸ Inclusion criteria were (i) age 21 years or older, (ii) ability to read and comprehend 5th grade English, and (iii) diagnosed with New York Heart Association (NYHA) functional classification I–IV HF. Exclusion criteria were (i) documented major cognitive impairment (e.g. Alzheimer’s disease) or active psychosis that would preclude study participation, (ii) prior heart transplantation or durable mechanical circulatory support, (iii) major and uncorrected hearing dysfunction, or (iv) were otherwise unable to complete the requirements of the study (e.g. life-threatening illness). The second study included patients prior to left ventricular assist device (LVAD) implantation enrolled from the same academic medical centre between April 2012 and May 2016 (i.e. ‘LVAD study’).²⁹ Inclusion criteria were (i) age 21 years or older, (ii) ability to read and comprehend 5th grade English or Spanish, and (iii) eligibility for implantation of a commercially available, U.S. Food and Drug Administration-approved continuous flow LVAD as a bridge to transplantation/decision or as destination therapy. Exclusion criteria were (i) documented major cognitive impairment (e.g. Alzheimer’s disease), major psychiatric illness, prior heart transplantation, or mechanical circulatory support or (ii) if they had a concomitant terminal illness that impeded participation in a 6-month study. The Institutional Review Board approved both studies, and written informed consent was obtained from all participants. The data reported on in this study only include those who consented to storing data and samples in a biorepository for future research.

Measurement

Sociodemographic and clinical data

Data collection procedures were identical across both studies. Data on sex and race, among other variables, were collected using sociodemographic questionnaires. We performed a medical record review to collect data on HF history, aetiology, NYHA functional class, clinical and laboratory data, and treatment of HF. The Charlson Comorbidity Index³⁰ was used to summarize comorbid conditions. The Seattle Heart Failure Model (SHFM) 1-year projected survival was calculated based on the model developed by Levy et al.³¹ and includes relevant clinical characteristics such as age, ejection fraction, lab values, medications, and other clinical parameters.

Physical symptoms

Overall fatigue was measured using the 13-item Functional Assessment of Chronic Illness Therapy Fatigue Scale (FACIT-F; v.4).³² The FACIT-F captures self-reported tiredness, weakness, and inability to perform activities of daily living as a result of fatigue. The 13 items are rated from 0 (not at all) to 4 (very much), and cumulative scores range from 0 to 52 with lower scores indicating more fatigue. The FACIT-F was only measured in the first study,²⁸ and Cronbach’s α of the FACIT-F in this sample was 0.94.

The 18-item Heart Failure Somatic Perception Scale (HFSPS; v.3) was used to measure dyspnoea and early/subtle symptoms.³³ The HFSPS asks questions about participants’ physical symptoms of HF and provides six response options ranging from 0 (I did not have this symptom) to 5 (extremely bothersome). The six-item HFSPS-Dyspnoea (HFSPS-D) subscale focuses on symptoms of dyspnoea, including orthopnoea, dyspnoea on exertion, and paroxysmal nocturnal dyspnoea; scores range from 0 to 30 with higher scores indicating worse dyspnoea. The seven-item HFSPS-Early/Subtle (HFSPS-ES) subscale focuses on symptoms that indicate early signs of decompensation or symptoms that are subtle, including upset stomach, cough, feeling tired, clothes feeling tighter, waking up to urinate, needing to rest more than usual, and not feeling like eating; scores range from 0 to 35 with higher scores indicating worse symptoms. The HFSPS was measured in both studies,^28,29 and Cronbach’s α on the HFSPS-D and HFSPS-ES subscales were 0.93 and 0.81, respectively, in the combined samples. We chose to only use the dyspnoea and early/subtle subscale, excluding the other two subscales (i.e. chest discomfort and oedema), because these have sufficient number of items, are well validated, and predict clinical outcomes in the HFSPS v.3.^33,34 as well as discriminate between moderate to advanced HF.¹⁰

Plasma catecholamines

For both studies, plasma was prepared from whole blood samples collected from participants and stored at −80°C. Plasma samples were thawed and centrifuged for processing, and there were no repeated freeze-thaw cycles. We used high-performance liquid chromatography with electrochemical detection to measure plasma NE and DHPG as described previously.³⁵ Briefly, we added an internal standard, dihydroxybenzylamine, to plasma samples and to standards, and the catechols were pre-purified by alumina extraction. The catechols were separated by reversed-phase chromatography on C18 column (Agilent Microsorb, 150 × 4.6 mm, 5 μm) and quantified by electrochemical detection (Coulochem III; ESA, Bedford, MA, USA). Electrode potentials were: guard cell at +300 mV, detector 1 at +150 mV, and detector 2 at −350 mV. The oxidation–reduction protocol was used to eliminate irreversibly oxidized uric acid, which co-eluted with DHPG.³⁵ In each assay, we included samples of pooled plasma from healthy adults as a comparison. The average intra-assay and inter-assay coefficients of variation were 9.0% and 10.8% for the first study (n = 111). The average intra-assay and inter-assay coefficients of variation were 5.3% and 4.6% for the second study (n = 38).

Statistical analysis

Sample characteristics are presented using standard descriptive statistics including measures of central tendency, variability, and frequency. Comparative statistics (Student’s t-test, Mann–Whitney U test, Fischer’s exact test, or the Pearson χ² test) were used to determine significant differences in characteristics between women and men, as well as between the two study groups (i.e. ‘chronic HF study’ vs. ‘LVAD study’). The catecholamine data were natural log (ln) transformed to address the skewed nature of the raw data (based on histograms and the skewness test in Stata). Multivariate linear regression was performed to measure the association of the natural log (ln) transformed levels of NE, DHPG, and NE:DHPG ratio with FACIT-F and HFSPS subscale scores, adjusting for the SHFM 1-year survival score to account for HF severity and treatment. We performed analyses on both the total sample and stratified by sex. We also used interaction testing to determine if the results differed by sex. Stata/MP version 17MP (StataCorp, College Station, TX, USA) was utilized for all statistical analyses.

Results

Sample characteristics are presented in Table 1. Our sample was a majority male (59%), predominantly Non-Hispanic Caucasian (83%), and 60.8 ± 15.7 years of age, on average. Most were NYHA Functional Class III/IV (61%) with predominantly non-ischaemic aetiologies (71%). Those in the ‘LVAD study’ were significantly younger, had more men than women, and had a few significant clinical differences common to advanced HF compared with those in the ‘chronic HF study’ (see Supplementary material online, Table S1). Additionally, there were significant differences in sympathetic markers and symptoms, allowing for a greater range of biomarker values and symptoms for combined analysis. In comparison with men, women were significantly older, had higher body mass indices, smaller left ventricles, higher left ventricular ejection fractions, and better SHFM 1-year survival rates (all P < 0.05). Women had significantly lower (i.e. worse) FACIT-T scores and lower DHPG levels compared with men (both P < 0.05).

Table 1.

Characteristics of the sample

	Total (n = 149)	Women (n = 61)	Men (n = 88)	P-value
Patient characteristics
Age (years)	60.8 ± 15.7	64.2 ± 15.6	58.5 ± 15.5	0.030
Non-Hispanic Caucasian	124 (83%)	52 (85%)	72 (82%)	0.582
Clinical characteristics
Body mass index (kg/m2)	30.7 ± 7.8	32.1 ± 8.7	29.6 ± 6.9	0.035
Charlson Comorbidity Index (weighted)	3.1 ± 1.9	3.1 ± 2.0	3.1 ± 1.8	0.965
Atrial fibrillation	74 (50%)	32 (53%)	42 (48%)	0.616
Stage 3 chronic kidney disease	41 (28%)	18 (30%)	23 (26%)	0.650
Type 2 diabetes	62 (42%)	26 (43%)	36 (41%)	0.835
Heart failure characteristics
Time with heart failure (years)	3.8 [1.4–9]	3.2 [1.1–7.1]	4 [1.7–10]	0.167
New York Heart Association functional Class III/IV	91 (61%)	33 (54%)	58 (66%)	0.146
Non-ischaemic aetiology	106 (71%)	51 (84%)	55 (63%)	0.005
Left ventricular end-diastolic diameter (cm)	5.9 ± 1.4	5.3 ± 1.1	6.3 ± 1.4	<0.001
Left ventricular ejection fraction (%)	37.4 ± 17.0	42.9 ± 17.5	33.5 ± 15.5	0.001
Serum sodium (mEq/L)	137.3 ± 3.8	137.8 ± 2.9	137.0 ± 4.3	0.156
Serum haemoglobin (g/dL)	12.7 ± 2.0	12.5 ± 1.8	12.8 ± 2.2	0.444
Serum BUN:creatinine ratio	21.3 ± 7.7	22.3 ± 8.0	20.5 ± 7.5	0.168
Prescribed a β-blocker	110 (74%)	44 (72%)	66 (75%)	0.695
Prescribed an angiotensin-converting enzyme-inhibitor or angiotensin II receptor blocker	109 (73%)	44 (72%)	65 (74%)	0.814
Prescribed an aldosterone antagonist	74 (50%)	34 (56%)	40 (45%)	0.217
SHFM 1-year projected survival (%)	92 [83–97]	96 [89–97]	90 [73–96]	0.008
Biomarkers
lnNE (pmol/mL)	1.51 ± 0.62	1.44 ± 0.55	1.55 ± 0.66	0.233
lnDHPG (pmol/mL)	2.72 ± 0.40	2.63 ± 0.39	2.79 ± 0.39	0.014
lnNE:lnDHPG ratio	0.55 ± 0.23	0.55 ± 0.23	0.55 ± 0.23	0.991
Symptoms
FACIT-F	34.5 ± 11.3	31.6 ± 12.4	37.0 ± 9.6	0.013
HFSPS-Dyspnoea	5 [1–14]	5 [1–12]	6 [0–14]	0.712
HFSPS-Early/Subtle	11.6 ± 6.9	12.1 ± 7.5	11.2 ± 6.6	0.475

Reported as mean ± SD, n (%), or median [IQR].

BUN, blood urea nitrogen; DHPG, dihydroxyphenylglycol; ln, natural log; FACIT-F, Functional Assessment of Chronic Illness Therapy Fatigue; HFSPS, heart failure somatic perception scale; NE, norepinephrine; SHFM, Seattle Heart Failure Model.

After adjusting for SHFM 1-year survival score, higher lnNE and higher lnNE:lnDHPG ratio were significantly associated with lower FACIT-F scores indicating worse fatigue [P = 0.043 and P = 0.013, respectively (Table 2)]. When stratifying on sex, higher lnNE:lnDHPG ratio was significantly associated with worse fatigue in women (P = 0.011), but there were no significant associations for men (Table 2). The interaction tests by sex for the associations between each of the sympathetic markers and FACIT-F scores are presented in Figure 1; the only significant interaction was for lnDHPG and FACIT-F scores.

Table 2.

Associations between markers of sympathetic function and physical symptoms

	FACIT-Fa		HFSPS-Dyspnoea Subscalea		HFSPS-Early and Subtle symptomsa
	β ± SE	P-value	β ± SE	P-value	β ± SE	P-value
Total	(n = 111)		(n = 149)		(n = 149)
lnNE	−3.82 ± 1.86	0.043	2.06 ± 1.11	0.065	1.64 ± 0.97	0.095
lnDHPG	3.32 ± 3.17	0.298	1.97 ± 1.67	0.240	−1.46 ± 1.47	0.321
lnNE:lnDHPG ratio	−11.80 ± 4.68	0.013	4.86 ± 2.79	0.084	5.63 ± 2.43	0.022
Female	(n = 52)		(n = 61)		(n = 61)
lnNE	−5.98 ± 3.02	0.054	2.05 ± 1.80	0.260	1.90 ± 1.75	0.282
lnDHPG	9.69 ± 4.93	0.055	0.41 ± 2.57	0.874	−5.81 ± 2.37	0.017
lnNE:lnDHPG ratio	−18.78 ± 7.10	0.011	5.20 ± 4.22	0.223	9.29 ± 3.96	0.023
Male	(n = 59)		(n = 88)		(n = 88)
lnNE	−1.67 ± 2.16	0.443	2.01 ± 1.43	0.164	1.52 ± 1.14	0.188
lnDHPG	−3.66 ± 3.77	0.336	3.19 ± 2.28	0.167	2.53 ± 1.82	0.168
lnNE:lnDHPG ratio	−2.98 ± 5.71	0.604	4.52 ± 3.80	0.238	2.39 ± 3.05	0.435

DHPG, dihydroxyphenylglycol; FACIT-F, Functional Assessment of Chronic Illness Therapy Fatigue; HFSPS, Heart Failure Somatic Perception Scale; ln, natural log; NE, norepinephrine; SE, standard error.

^aAdjusting for Seattle Heart Failure Model (SHFM) 1-year survival score (a composite of clinical variables and heart failure treatments).

Figure 1.

Relationship between fatigue symptoms and sympathetic markers in women vs. men. Associations between Functional Assessment of Chronic Illness Therapy Fatigue Scale scores and the natural log of plasma norepinephrine (A), dihydroxyphenylglycol (B), and norepinephrine:dihydroxyphenylglycol ratio (C) are presented in women and men, after adjusting for Seattle Heart Failure Model Score. Interaction testing P-values are presented for each marker. DHPG, dihydroxyphenylglycol; FACIT-F, Functional Assessment of Chronic Illness Therapy Fatigue Scale; ln, natural log; NE, norepinephrine.

After adjusting for SHFM 1-year survival score, there were no significant associations between any of the sympathetic markers and the HFSPS-D scores, either in the total sample or sex-stratified (Table 2). The interaction tests by sex for the associations between each of the sympathetic markers and HFSPS-D scores are presented in Figure 2.

Figure 2.

Relationship between dyspnoea symptoms and sympathetic markers in women vs. men. Associations between HFSPS-Dyspnoea scores and the natural log of plasma norepinephrine (A), dihydroxyphenylglycol (B), and norepinephrine:dihydroxyphenylglycol ratio (C) are presented in women and men, after adjusting for Seattle Heart Failure Model Score. Interaction testing P-values are presented for each marker. DHPG, dihydroxyphenylglycol; HFSPS, Heart Failure Somatic Perception Scale; ln, natural log; NE, norepinephrine.

After adjusting for SHFM 1-year survival score, higher lnNE:lnDHPG ratio was associated with higher HFSPS-ES scores overall indicating worse early/subtle symptoms (P = 0.022; Table 2). When stratifying on sex, lower lnDHPG levels and higher lnNE:lnDHPG ratios were significantly associated with worse early/subtle symptoms in women (P = 0.017 and P = 0.023, respectively), but there were no significant associations for men (Table 2). The interaction tests by sex for the associations between each of the sympathetic markers and HFSPS-ES scores are presented in Figure 3; the only significant interaction was for lnDHPG and HFSPS-ES scores.

Figure 3.

Relationship between early/subtle symptoms and sympathetic markers in women vs. men. Associations between HFSPS-Early/Subtle scores and the natural log of plasma norepinephrine (A), dihydroxyphenylglycol (B), and norepinephrine:dihydroxyphenylglycol ratio (C) are presented in women and men, after adjusting for Seattle Heart Failure Model Score. Interaction testing P-values are presented for each marker. DHPG, dihydroxyphenylglycol; HFSPS, Heart Failure Somatic Perception Scale; ln, natural log; NE, norepinephrine.

Discussion

In this study, we found significant associations between sympathetic dysfunction and physical symptoms of HF. Moreover, when exploring potential sex differences, we found that these associations were only significant among women but not men. Physical symptoms are particularly burdensome in HF, but our understanding of the underlying mechanisms is limited. Our findings show that one of the mechanisms underlying physical symptoms may be sympathetic overactivation, which could be targeted to address symptom burden, particularly among women.

The symptom experience in HF is multifaceted and heterogeneous as physical symptoms can encompass varying degrees of dyspnoea, fatigue, sleepiness, pain, gastrointestinal upset, and many other problematic symptoms.^3,7,33,36 Therefore, it is important to examine symptoms from multiple angles when exploring biological mechanisms. In previous work, using the total HFSPS Score (i.e. combined physical symptoms), we found significant associations between G protein-coupled receptor kinase-2 (as a marker of beta-adrenergic receptor downregulation and internalization) and physical symptoms.¹² Using the Kansas City Cardiomyopathy Questionnaire Overall Summary Score, a measure of health-related quality of life, we found associations between two markers of sympathetic dysfunction (i.e. G protein-coupled receptor kinase-2 and DHPG) and overall summary scores.²¹ This current study adds to these findings by showing that general fatigue as well as the early/subtle symptoms of HF (e.g. upset stomach, cough, need to urinate, fatigue, early satiety, and clothes feeling tighter) are also associated with sympathetic dysfunction. While dyspnoea symptoms are often the primary focus of HF symptom inventories, especially clinically, this study highlights the need to consider the broader constellation of symptoms that women and men may experience in HF. Multiple studies have demonstrated that symptom burden is associated with worse clinical outcomes and is a primary driver of healthcare utilization in patients with HF, as recently highlighted in multiple scientific statements.^37,38 As such, our work highlights the continued need to understand the patient experience of HF symptoms.

One consistent finding is that increased NE:DHPG ratios are significantly associated with both fatigue symptoms and early/subtle HF symptoms, especially in women. Plasma NE levels are determined by the release rate of NE from sympathetic nerve endings, its reuptake, and its breakdown into DHPG by monoamine oxidase. The ratio of NE and DHPG, therefore, complements NE levels alone by providing insight into whether there is normal reuptake of plasma NE level (NE:DHPG ratio stays the same) or diminished NE reuptake (NE:DHPG ratio increases). In a small sample of patients with a left ventricular assist device (n = 39), we previously found that increased NE:DHPG ratios were numerically, but not significantly, increased with worse health-related quality of life.²¹ Taken together, it appears that impaired NE reuptake may contribute to adverse patient-reported outcomes in HF.

Notably, when exploring these associations by sex, we found that they only persisted in women and not men. Studies informing our understanding of HF have historically been based on male patients, and there is a dearth of women included in HF clinical trials.³⁹ However, there are noteworthy sex differences in objective measures of HF aetiology, risk factors, pathophysiology, and presentation as well as in underlying cardiovascular system function.³⁹ Importantly, women and men may have different physiologic responses to intervention such as mechanical circulatory support.⁴⁰ We have previously shown that having a larger left ventricle was associated with better physical symptoms in women but more severe physical symptoms in men.²⁷ Moreover, women more commonly have HF with preserved ejection fraction, higher rates of frailty, and more comorbidities and depressive symptoms than men,^39,41 which could possibly converge on sex differences in the biological underpinnings of symptoms. Taken together, differences in left ventricular geometry, along with other phenotypic characteristics, and the downstream compensatory mechanisms (i.e. cardiac NE spillover²³) may give rise to worse symptomatology among women.

There are a number of clinical implications of this study. First, it is important to consider the broad constellation of symptoms that patients with HF may experience beyond dyspnoea-related symptoms. Symptoms may be subtle or early indicators of decompensation and should be incorporated into the assessment and management plan for patients. Second, this study highlights that sexual dimorphisms are relevant in understanding the potential causes of symptoms in HF. Based on our findings, sympathetic dysfunction may be linked with the early/subtle symptoms experienced by women, indicating that targeting sympathetic dysfunction may prevent HF decompensation among women. Third, these findings underscore the need to implement and optimize guideline-directed medical therapy, especially among women, to mitigate the adverse sequalae of sympathetic dysfunction.¹⁷

Limitations of this study include the sample size and the demographics of the participants. Even though the sample included over 40% women, the small sample size may be insufficiently powered to detect some significant sex differences. Moreover, although we leveraged data from two previous studies, this was a secondary data analysis that was limited to the data available as part of the merged dataset. Furthermore, study participants were enrolled from a single clinic in the Pacific Northwest and are predominantly non-Hispanic Caucasians. Lastly, we were restricted to identifying associations and not causal relationships due to the study design. Nonetheless, our findings suggest that in patients with moderate to advanced HF, markers of sympathetic activation are associated with worse fatigue and early/subtle HF symptoms.

These findings can serve as a starting point for further research. Future research utilizing larger numbers of male and female participants is needed to yield a better understanding of the sex differences in HF symptom burden and markers of HF pathophysiology. Moreover, other markers of sympathetic dysfunction (e.g. microneurography) may help provide additional mechanistic evidence. Finally, future research could examine these associations alongside interventions (pharmacological and non-pharmacological) that target the sympathetic nervous system, in part; this evidence would help confirm or refute these potential mechanisms.

Conclusions

Our study found that markers of sympathetic dysfunction are associated with worse fatigue and early/subtle physical symptoms in moderate to advanced HF, primarily among women. These findings provide evidence that sympathetic dysfunction could be one mechanism underlying symptoms in HF and that sexual dimorphisms need to be considered when studying symptom biology. Further research is needed to gain a better understanding of sex differences in the relationship between sympathetic overactivation and physical symptoms of HF.

Supplementary material

Supplementary material is available at European Journal of Cardiovascular Nursing online.

Funding

The Office of Research on Women’s Health and the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the NIH (K12HD043488) and the National Institute of Nursing Research (R01NR013492). The work reported in this article was also supported, in part, by the National Center for Advancing Translational Sciences of the NIH (UL1TR002369). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Author contributions

N.S. contributed to the conceptualization, data curation, investigation, validation, and writing—original draft. N.P. contributed to the investigation, methodology, validation, and writing—reviewing and editing. B.A.H. contributed to the funding acquisition, investigation, methodology, resources, supervision, and writing—review and editing. C.Y.J. contributed to methodology, validation, and writing—reviewing and editing. W.R.W. contributed to the data curation, investigation, methodology, resources, supervision, and writing—review and editing. C.S.L. contributed to the funding acquisition, investigation, methodology, resources, supervision, and writing—review and editing. Q.E.D. contributed to the conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, visualization, and writing—original draft. All authors have approved the final version.

Data availability

Data are available on reasonable request to the corresponding author. Code also available upon request to the corresponding author.