Prevalence and Prognostic Significance of Pulmonary Function Test Abnormalities in Hospitalized Patients With Acute Decompensated Heart Failure With Preserved and Reduced Ejection Fraction

Rika Kawakami; Yasuki Nakada; Yukihiro Hashimoto; Tomoya Ueda; Hitoshi Nakagawa; Taku Nishida; Kenji Onoue; Tsunenari Soeda; Makoto Watanabe; Yoshihiko Saito

doi:10.1253/circj.CJ-20-1069

Abstract

Background: This study evaluated the prevalence and prognostic impact of lung function abnormalities in patients with acute decompensated heart failure (ADHF) with reduced ejection fraction (HFrEF) and preserved ejection fraction (HFpEF).

Methods and Results: Of the 1,012 consecutive patients who were admitted to Nara Medical University with ADHF between 2011 and 2018, 657 routinely underwent spirometry (pulmonary function test [PFT]) before discharge. Lung function was classified as normal or abnormal (restrictive, obstructive, or mixed). Abnormal PFTs were seen in 63.0% of patients with ADHF (36.7%, 13.1%, and 13.2% for restrictive, obstructive, and mixed, respectively). The prevalence of abnormal PFT increased with age (P<0.001). Overall, abnormal PFT was an independent predictor of the composite endpoint of cardiovascular mortality or hospitalization for HF (adjusted hazard ratio [HR] 1.402; 95% confidence interval [CI] 1.039–1.914; P=0.027). Abnormal PFT (adjusted HR 2.294; 95% CI 1.368–4.064; P=0.001), as well as the restrictive (HR 2.299; 95% CI 1.322–4.175; P=0.003) and mixed (HR 2.784; 95% CI 1.399–5.581; P=0.004) patterns, were predictive of the composite endpoint in HFpEF, but not in HFrEF.

Conclusions: Abnormal PFT was prevalent and associated with poor outcomes in ADHF. Spirometry may be a useful tool in patients with ADHF, especially in those with HFpEF, to identify those at higher risk of a poorer outcome.

The prevalence of heart failure (HF) increases with age, and so these patients are more likely to have comorbidities, including chronic kidney disease, anemia, frailty, and chronic obstructive pulmonary disease (COPD).¹^,² Generally, the prognosis of HF is poorer with an increasing number of comorbidities.¹^,³ Another feature of aged patients with HF is the higher frequency of HF with preserved ejection fraction (HFpEF), the prognosis of which is as poor as that of HF with reduced EF (HFrEF).⁴^–⁶ A specific treatment for HFpEF has not yet been established, with current management limited to diuretics and the treatment of comorbidities. To better manage HF, especially HFpEF, it is necessary to understand its pathophysiology, with a focus on the impact of comorbidities.

Editorial p 1435

Earlier studies reported that COPD as a comorbidity of HF was associated with poor prognosis in patients with HF.¹^,⁷^,⁸ However, lung function abnormalities, both at rest and during exercise, are frequently observed in patients with HF, despite the absence of clinically overt respiratory diseases.⁹^,¹⁰ In fact, it has been reported that ventilation and diffusion abnormalities with a reduction in lung volumes on pulmonary function tests (PFTs) are present in up to 60% of patients with end-stage HFrEF.¹¹^–¹³ Furthermore, the percentage of predicted forced expiratory volume in 1 s (FEV₁) measured 24–72 h after admission was an independent predictor of mortality in the setting of acute decompensated HF (ADHF).¹⁴ However, lung function is not routinely investigated in patients with ADHF; therefore, the prevalence and type of lung function abnormalities in this population remain unknown. Moreover, the effect of lung function abnormalities on long-term outcomes has not been fully investigated in patients with ADHF, especially those with HFpEF.

Thus, the purpose of the present study was to evaluate the prevalence, type, and prognostic impact of abnormal PFT in HF subtypes (HFpEF, HFrEF) using spirometry, which is a useful and routine tool for assessing lung impairment and breathing pattern disorders.

Methods

Patient Selection

The Nara Registry and Analyses for Heart Failure 4 (NARA-HF4) study is a single-centre registry that has collected data on 1,012 consecutive patients who were emergency admissions to the cardiology wards or the coronary care unit at our hospital for ADHF (either acute new-onset or acute on chronic HF) between April 2011 and December 2018. The diagnosis of HF was based on the Framingham criteria for HF.¹⁵ Exclusion criteria were patient age <20 years and acute de novo HF after acute coronary syndrome, acute myocarditis, and acute HF with acute pulmonary embolism. In all, 355 patients were excluded because they were lost to follow-up (n=17), lacked PFT results (n=290), or died during hospitalization (n=48; Supplementary Figure). For each patient, baseline data were collected, including age, sex, body mass index (BMI), cause of HF, medical history, vital signs, laboratory and echocardiographic data, and medications at discharge.

This study was approved by the Nara Medical University Institutional Ethics Committee and was performed in accordance with the 1975 Declaration of Helsinki guidelines for clinical research protocols. Informed consent was provided by all patients.

Lung Function Measurement

All 657 patients were tested for lung function using spirometry before discharge. ADHF patients underwent spirometry evaluation, including forced vital capacity (FVC) and assessment of FEV₁. Patients were asked to perform at least 2 FVC maneuvers according to the recommended method.¹⁶ The highest FEV₁ and FVC values were recorded. Data are expressed as a ratio or as a percentage of predicted values (i.e., FEV₁/FVC, %FVC, and %FEV₁). Predicted values for lung flow were calculated with established reference equations, which were adjusted by absolute values for age, sex, and height. Reference values for %FVC and %FEV₁ were derived from the Japanese criteria established by the Japanese Respiratory Society.¹⁷

Definitions

Lung function was classified as normal or abnormal (restrictive, obstructive, and mixed patterns). Normal lung function was defined as %FVC ≥80% and FEV₁/FVC ≥70%. The different abnormal patterns of lung fucntion were defined as follows: restrictive pattern, %FVC <80% and FEV₁/FVC ≥70%; obstructive pattern, %FVC ≥80% and FEV₁/FVC <70%; and mixed pattern, %FVC <80% and FEV₁/FVC <70%. According to baseline lung function, patients were divided into 2 groups, those with normal or abnormal PFTs. Self-reported COPD was considered present when patients reported the diagnosis or when a diagnosis of COPD was reported in the medical records, regardless of spirometry findings.

Echocardiography

HFrEF was defined as a left ventricular ejection fraction (LVEF) <40% on an echocardiogram, and HFpEF was diagnosed based on clinical guidelines and LVEF ≥50%.¹⁸^,¹⁹ Echocardiograms were performed before discharge; LVEF was calculated from the conventional apical 2- and 4-chamber images using the modified Simpson’s method. Systolic pulmonary artery pressure (PAP) was estimated from a tricuspid regurgitation (TR) peak velocity (TRV) by continuous-wave Doppler using the modified Bernoulli equation in the absence of right ventricular outflow tract (RVOT) obstruction. Normal resting values are usually defined as a TRV of ≤2.8 to 2.9 m/s, or a peak systolic pressure of 35–36 mmHg, assuming a right atrium (RA) pressure of 3–5 mmHg.²⁰^,²¹ A TR gradient of >2.9 m/s was used as a definition of the possibility of pulmonary hypertension.

Outcomes

The primary endpoint of the study was a composite of cardiovascular (CV) mortality or hospitalization for HF. Survival status and the date of death or last follow-up of patients were obtained from the medical records. When this information was unavailable in the medical record, patients (or their families) and their family doctors were interviewed by telephone or were additionally asked to complete a postal questionnaire to screen for outcome events. The status of all patients was surveyed at least 6 months after discharge, and the following information was obtained: (1) survival; (2) cause of death; and (3) rehospitalization for exacerbation of HF that required more than continuation of the usual therapy from the prior admission.

Statistical Analysis

Data are presented as the mean±SD, numbers and percentages for categorical variables, or as the median with interquartile range (IQR). The comparison of continuous variables between the 2 groups was done using the Student-t test for normally distributed variables or the Mann-Whitney U test for non-normally distributed variables, and the χ² test for categorical variables. Logistic regression analysis was used to identify univariate predictors of abnormal PFT; backward stepwise logistic regression was used to identify independent predictors of abnormal PFT. Variables with P<0.10 in the univariate model were entered into a multivariate analysis. Cumulative event-free rates during follow-up were derived using the Kaplan-Meier method.

Multivariate Cox proportional hazards regression models were used to determine the association of abnormal PFT with CV mortality or hospitalization for HF. In subsequent multivariate Cox proportional hazard analyses to identify independent predictors, 3 approaches were used. Model 1 adjusted for age and sex; Model 2 adjusted for all variables in Model 1 plus other baseline clinical variables known to have an association with the outcome (i.e., B-type natriuretic peptide [BNP], hemoglobin [Hb], estimated glomerular filtration rate [eGFR], and sodium); and Model 3 was established from variables with P<0.10 in the univariate analysis, using a stepwise selection process to identify relevant variables. The hazard ratios (HRs) for outcomes in the abnormal PFT group were compared with those for the normal PFT group, which served as the reference group.

A value of P<0.05 was considered significant. All statistical analyses were performed using JMP version 13 for Windows (SAS Institute, Cary, NC, USA).

Results

Patient Characteristics

The overall mean age of the cohort was 73.3±12.0 years, 59.4% were men, and the median length of hospital stay was 19 days (IQR 14–27 days). The study population comprised 657 patients (243 with normal and 414 with abnormal PFTs). Of the 414 patients with abnormal PFTs, 86 (13.1%), 241 (36.7%), and 87 (13.2%) patients demonstrated obstructive, restrictive, and mixed patterns, respectively (Figure 1). Compared with patients with normal PFTs, those with abnormal PFTs were significantly older and had significantly higher LVEF, lower eGFR, and lower Hb levels (Table 1). Furthermore, patients with abnormal PFTs were more likely to have atrial fibrillation (AF), self-reported COPD, lung disease (e.g., COPD, interstitial pneumonia, prior pulmonary tuberculosis, asthma, and pneumonectomy), and TRV >2.9 m/s. There were no significant differences in BNP levels and the prevalence of hypertension, male sex, and current and former smokers between patients with normal PFT and those with abnormal PFT.

Figure 1.

Prevalence of abnormal pulmonary function tests (PFTs). (Right) The bar-of-pie chart shows the prevalence of the different types of abnormal PFTs in the overall population (n=657). (Left) The smaller bar-of-pie charts show the prevalence of the different types of abnormal PFTs in heart failure with preserved ejection fraction (HFpEF; n=215) and in heart failure with reduced ejection fraction (HFrEF; n=320). *P<0.01 vs. HFrEF.

Table 1. Baseline Characteristics

	All			HFpEF			HFrEF
	Normal PFT (n=243)	Abnormal PFT (n=414)	P value	Normal PFT (n=68)	Abnormal PFT (n=147)	P value	Normal PFT (n=141)	Abnormal PFT (n=179)	P value
Age (years)	69.2±12.9	75.7±10.7	<0.001	73.7±11.7	77.4±9.5	0.016	65.5±13.0	73.8±11.3	<0.001
Male sex	133 (54.7)	259 (62.1)	0.065	25 (36.8)	83 (56.5)	0.007	91 (64.5)	120 (67.0)	0.640
Prior HF hospitalization	49 (20.2)	102 (24.6)	0.185	10 (14.7)	37 (24.7)	0.089	33 (23.4)	46 (25.7)	0.636
Ischemic etiology	103 (42.4)	139 (33.6)	0.024	17 (25.0)	21 (14.3)	0.061	65 (46.1)	81 (45.3)	0.880
LOS (days)	19 (14, 26)	19 (14, 27)	0.806	19 (13, 26)	18 (12, 25)	0.529	20 (14, 27)	20 (16, 29)	0.620
BMI at discharge (kg/m²)	21.9±3.8	21.4±4.4	0.139	22.5±3.4	21.6±4.7	0.170	21.9±4.1	21.2±4.2	0.701
SBP (mmHg)	112.1±18.0	110.9±17.0	0.416	119.1±17.7	114.3±16.8	0.057	107.5±16.8	106.3±15.2	0.478
Heart rate (beats/min)	70.7±10.1	70.9±11.3	0.981	71.1±10.5	71.2±11.2	0.963	71.0±10.0	70.6±12.0	0.760
CTR (%)	54.3±6.4	57.3±7.8	<0.001	54.8±7.2	58.2±7.9	0.003	53.8±5.9	56.7±6.9	<0.001
Medical history
MI	58 (23.9)	106 (25.6)	0.619	10 (14.7)	15 (10.2)	0.347	39 (27.7)	61 (34.1)	0.217
Hypertension	186 (76.5)	307 (74.2)	0.493	57 (83.8)	117 (79.6)	0.458	100 (70.9)	121 (67.6)	0.523
Diabetes	109 (44.9)	156 (37.7)	0.071	33 (48.5)	51 (34.7)	0.054	62 (44.0)	64 (335.8)	0.136
Atrial fibrillation	73 (30.0)	202 (48.8)	<0.001	24 (35.3)	92 (62.6)	<0.001	42 (29.8)	76 (42.5)	0.019
Self-reported COPD	2 (0.8)	66 (15.9)	<0.001	1 (1.47)	22 (15.0)	<0.001	1 (0.7)	32 (17.9)	<0.001
Any lung disease	26 (10.7)	126 (30.4)	<0.001	9 (13.2)	49 (33.3)	<0.001	12 (8.5)	57 (31.8)	<0.001
Current or former smoker	145 (59.7)	251 (60.6)	0.809	29 (42.7)	80 (54.4)	0.120	97 (68.8)	120 (67.0)	0.739
Discharge laboratory data
eGFR (mL/min/1.73 m²)	45.9±23.7	41.8±22.5	0.028	43.1±25.3	39.0±22.4	0.232	49.4±22.1	44.8±21.9	0.063
Hemoglobin (g/dL)	12.1±2.2	11.6±2.0	0.005	11.1±1.7	11.0±1.9	0.743	12.7±2.3	12.2±1.9	0.037
Sodium (mEq/L)	138.5±3.3	137.8±4.0	0.013	138.7±3.7	137.9±3.8	0.186	138.7±3.1	137.3±4.1	<0.001
BNP (pg/mL)	261 [127–478]	261 [148–503]	0.223	211 [100–414]	187 [116–364]	0.998	274 [133–487]	354 [171–573]	0.015
Echocardiographic data at discharge
LVEF (%)	39.7±16.3	45.0±17.0	<0.001	62.1±7.6	64.6±7.8	0.025	27.9±6.5	29.2±6.7	0.084
E’ (cm/s)	4.7±1.8	4.9±2.0	0.381	5.5±2.4	5.5±2.4	0.894	4.4±1.4	4.5±1.6	0.864
E/E’ ratio	17.2±7.3	18.1±8.4	0.236	18.6±7.4	20.2±9.6	0.269	16.6±7.6	16.6±7.6	0.997
LAVI (mL/m²)	41.0±24.7	44.3±26.8	0.126	47.7±35.9	51.7±33.7	0.434	38.1±19.1	30.5±22.1	0.300
LVMI (g/m²)	150.0±39.1	138.4±44.6	<0.001	135.3±40.3	125.3±40.3	0.097	158.7±37.7	149.6±47.1	0.063
TRV >2.9 m/s	33 (13.6)	86 (20.8)	0.011	14 (20.6)	47 (36.7)	0.054	16 (11.3)	27 (15.1)	0.297
Discharge medication
ACEI/ARBs	225 (92.6)	355 (85.8)	0.007	58 (85.3)	114 (76.0)	0.178	137 (97.2)	170 (95.0)	0.316
β-blockers	197 (81.1)	289 (70.2)	0.002	44 (64.7)	73 (49.7)	0.038	129 (91.5)	149 (83.7)	0.036
Diuretics	192 (79.0)	341 (82.4)	0.291	52 (76.5)	119 (81.0)	0.453	115 (81.6)	154 (86.0)	0.279
Aldosterone antagonists	105 (43.2)	174 (42.0)	0.768	24 (35.3)	49 (33.3)	0.778	72 (51.1)	89 (49.7)	0.811
Bronchodilators	6 (2.5)	42 (10.1)	<0.001	2 (2.9)	17 (11.6)	0.024	3 (2.1)	18 (10.1)	0.003
PFT
FVC (% predicted)	93.7±10.6	68.9±18.6	<0.001	92.2±10.4	64.7±17.7	<0.001	93.6±10.2	73.1±19.2	<0.001
FEV₁ (% predicted)	100.8±17.2	72.2±22.0	<0.001	102.3±19.3	69.9±23.4	<0.001	99.0±15.2	74.1±21.1	<0.001
FEV₁/FVC ratio (%)	79.6±5.6	73.6±14.1	<0.001	80.1±5.6	74.2±14.1	<0.001	79.7±5.6	72.9±14.5	<0.001

Values are given as the mean±SD, median [interquartile range] or n (%). ACEI, angiotensin-converting enzyme inhibitor; ARBs, angiotensin II receptor blockers; BMI, body mass index; BNP, B-type natriuretic peptide; COPD, chronic obstructive pulmonary disease; CTR, cardiothoracic ratio; e’, early diastolic mitral annular tissue velocity; E/e’, early diastolic transmitral velocity to E’; eGFR, estimated glomerular filtration rate; FEV₁, forced expiratory volume in 1 s; FVC, forced vital capacity; HF, heart failure; HFpEF, HF with preserved ejection fraction; HFrEF, HF with reduced ejection fraction; LAVI, left atrial volume index; LOS, length of stay; LVEF, left ventricular ejection fraction; LVMI, left ventricular mass index; MI, myocardial infarction; PFT, pulmonary function test; SBP, systolic blood pressure; TRV, tricuspid regurgitation peak velocity.

Approximately half of the study population had reduced EF (320 patients had HFrEF, 215 had HFpEF). In both the HFrEF and HFpEF groups, patients with abnormal PFTs were significantly older and had a higher prevalence of AF, self-reported COPD and lung disease than those with normal PFTs (Table 1). HFpEF patients with abnormal PFTs were more commonly women than those with normal PFTs, whereas HFrEF patients with abnormal PFTs were more likely to have lower Hb, sodium, and BNP levels than those with normal PFTs.

Prevalence of Abnormal PFT by HF Subtype

The prevalence of abnormal PFT was significantly higher in HFpEF than HFrEF patients (68.4% vs. 55.9%, respectively; P=0.004; Figure 1). The restrictive pattern constituted the majority of abnormal PFTs in the setting of HFrEF and HFpEF; however, the prevalence of the restrictive pattern was higher in HFpEF than HFrEF (43.3% vs. 30.0%, respectively; P=0.002; Figure 1). HFrEF patients tended to have a higher prevalence of the obstructive spirometry pattern than HFpEF patients, but the difference was not statistically significant.

The types of abnormal PFT according to age are shown in Figure 2. The prevalence of abnormal PFT increased with age in overall ADHF (P<0.001) and its subtypes (i.e., HFpEF [P=0.077] and HFrEF [P<0.001]). The proportion of all abnormal PFT patterns increased progressively from ages <60 to ≥80 years in overall ADHF and for HFrEF. In contrast, in HFpEF, the obstructive pattern declined with age (except in those aged <60 years), from 24.1% in patients aged 60–69 years, to 9.3% and 7.4% for those aged 70–79 and ≥80 years, respectively (P=0.027).

Figure 2.

Percentage distribution of normal and abnormal (restrictive, obstructive, mixed) pulmonary function tests (PFTs) by age. HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction.

Predictors of Abnormal PFT

Univariate and multivariate predictors of abnormal PFT are presented in Table 2. Multivariate logistic regression analysis revealed that advanced age, male sex, AF, lung disease, and the cardiothoracic ratio (CTR) were significant independent predictors of abnormal PFT in our study population. Smoking was a strong predictor of the obstructive pattern, whereas CTR was the only independent predictor of the restrictive pattern. In addition, advanced age was found to be a significant predictor of both mixed and obstructive patterns, but not the restrictive pattern. The use of angiotensin-converting enzyme inhibitors/angiotensin receptor blockers or β-blockers was associated with a decreased risk of abnormal PFTs on univariate logistic regression analysis, with odds ratios of 0.549 (95% confidence interval [CI] 0.371–0.801; P=0.002) and 0.549 (95% CI 0.371–0.801; P=0.002), respectively, but not on multivariate analysis.

Table 2. Univariate and Multivariate Predictors of Abnormal Pulmonary Function Tests in the Overall Population

Factors	Univariate		Multivariate
Factors	OR (95% CI)	P value	OR (95% CI)	P value
Age (per 1 year)	1.047 (1.033–1.063)	<0.001	1.036 (1.020–1.051)	<0.001
Male sex	1.354 (0.982–1.867)	0.065	2.041 (1.387–3.025)	<0.001
BMI (per 1 kg/m²)	0.972 (0.936–1.009)	0.140	–	–
Atrial fibrillation	2.219 (1.592–3.114)	<0.001	1.602 (1.103–2.334)	0.013
Current or former smoker	1.041 (0.752–1.437)	0.809	–	–
Any lung disease	3.651 (2.346–5.876)	<0.001	3.165 (1.978–5.217)	<0.001
BNP (per 10 pg/mL)	1.003 (0.999–1.007)	0.144	–	–
Cardiothoracic ratio (per 1%)	1.061 (1.037–1.088)	<0.001	1.065 (1.036–1.097)	<0.001
TRV >2.9 m/s	1.753 (1.137–2.756)	0.011	–	–

CI, confidence interval; OR, odds ratio. Other abbreviations as in Table 1.

Overall Long-Term Prognostic Impact of Abnormal PFTs

During a median follow-up of 960 days (IQR 518–1,647 days), 264 deaths (40.2%) occurred; 126 (19.2%) due to CV causes, 117 (17.8%) due to non-CV causes, and 21 (3.2%) due to unknown causes. The primary composite endpoint of CV mortality or hospitalization for HF occurred in 189 patients (28.8%) with abnormal PFTs and in 77 patients (11.7%) with normal PFTs (P<0.001). The Kaplan-Meier curves revealed that the rates of the primary composite endpoint of CV mortality or hospitalization for HF were signiﬁcantly higher in patients with abnormal than normal PFTs (log-rank P<0.001; Figure 3A). Moreover, compared with the normal pattern, the incidence of the primary composite endpoint was signiﬁcantly higher in the restrictive and mixed patterns (log-rank P=0.001 and P<0.001, respectively; Figure 3B). There was no signiﬁcant difference in the incidence of the primary composite endpoint between the normal and obstructive patterns (P=0.099).

Figure 3.

Kaplan-Meier curves for the primary composite outcome of cardiovascular mortality or hospitalization for heart failure (HF) according to normal or abnormal (restrictive, obstructive, mixed) pulmonary function tests (PFTs). (A) Overall population. (B) Overall population classified by spirometry patterns. (C) Patients with HF with preserved ejection fraction (HFpEF). (D) Patients with HF with reduced ejection fraction (HFrEF). P values were calculated using log-rank tests.

In univariate Cox proportional regression analysis, abnormal PFT was an independent predictor for the primary endpoint (HR 1.689; 95% CI 1.301–2.216; P<0.001; Table 3). Among the overall population, abnormal PFT was associated with an increased risk of CV mortality or hospitalization for HF after adjusting for: age and sex (HR 1.395; 95% CI 1.066–1.845; P=0.015; Model 1); for age, sex, Hb, eGFR, BNP, and sodium (HR 1.345; 95% CI 1.026–1.780; P=0.032; Model 2); and for the variables that were significant in the univariate Cox analysis (HR 1.402; 95% CI 1.039–1.914; P=0.027; Model 3; Table 4). In addition, the mixed pattern was associated with a significant adjusted HR for CV mortality or hospitalization for HF of 1.718 (95% CI 1.147–2.559; P=0.009; Model 3; Table 4).

Table 3. Univariate Analysis for the Prediction of Cardiovascular Mortality or Hospitalization for HF in the Overall Population and in HF Patients With Preserved (HFpEF) or Reduced (HFrEF) Ejection Fraction Separately

	Overall (n=657)		HFpEF (n=215)		HFrEF (n=320)
	HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value
Age (per 1 year)	1.032 (1.020–1.044)	<0.001	1.027 (1.006–1.049)	0.010	1.036 (1.020–1.054)	<0.001
Male sex	1.293 (1.010–1.667)	0.042	1.361 (0.917–2.029)	0.127	1.541 (1.046–2.328)	0.028
BMI (per 1 kg/m²)	0.981 (0.951–1.010)	0.202	0.946 (0.898–0.995)	0.030	1.019 (0.976–1.060)	0.389
Atrial fibrillation	1.522 (1.196–1.937)	<0.001	1.415 (0.951–2.125)	0.087	1.733 (1.213–2.469)	0.003
Prior HF admission	2.112 (1.634–2.712)	<0.001	1.688 (1.083–2.570)	0.022	2.679 (1.862–3.824)	<0.001
Self-reported COPD	1.640 (1.129–2.311)	0.011	1.602 (0.851–2.773)	0.136	1.412 (0.772–2.387)	0.247
Abnormal PFT	1.689 (1.301–2.216)	<0.001	2.144 (1.357–3.531)	<0.001	1.731 (1.201–2.531)	0.003
Spirometry patterns		<0.001		0.006		0.014
Normal	1 (Reference)		1 (Reference)		1 (Reference)
Restrictive	1.596 (1.190–2.151)	0.002	2.149 (1.315–3.627)	0.002	1.744 (1.135–2.677)	0.011
Obstructive	1.418 (0.940–2.098)	0.095	1.613 (0.703–3.400)	0.245	1.427 (0.822–2.393)	0.200
Mixed	2.275 (1.583–3.240)	<0.001	2.520 (1.348–4.665)	0.004	2.202 (1.252–3.731)	0.007
Hemoglobin (per 1 g/dL)	0.858 (0.807–0.912)	<0.001	0.979 (0.876–1.092)	0.709	0.826 (0.757–0.899)	<0.001
eGFR (per 1 mL/min/1.73 m²)	0.989 (0.983–0.994)	<0.001	0.998 (0.989–1.006)	0.578	0.977 (0.968–0.986)	<0.001
Sodium (per 1 mEq/L)	0.994 (0.963–1.028)	0.724	1.015 (0.964–1.074)	0.580	0.977 (0.932–1.026)	0.355
BNP (per 10 pg/mL)	1.004 (1.002–1.006)	<0.001	1.003 (1.000–1.006)	0.026	1.006 (1.003–1.009)	<0.001
LVEF (per 1%)	1.005 (0.998–1.012)	0.171	0.998 (0.974–1.022)	0.886	1.003 (0.976–1.031)	0.839
TRV >2.9 m/s	1.800 (1.345–2.381)	<0.001	2.001 (1.290–3.084)	0.002	1.682 (1.047–2.602)	0.032

HR, hazard ratio. Other abbreviations as in Tables 1,2.

Table 4. Multivariate Cox Proportional Hazards Regression Model Analysis for Cardiovascular Mortality or Hospitalization for HF in the Overall Population and in HF Patients With Preserved (HFpEF) or Reduced (HFrEF) Ejection Fraction Separately

	Overall population		HFpEF		HFrEF
	HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value
Abnormal PFT
Model 1	1.395 (1.066–1.845)	0.015	1.919 (1.206–3.182)	0.005	1.319 (0.896–1.967)	0.163
Model 2	1.345 (1.026–1.780)	0.032	1.938 (1.212–3.224)	0.005	1.318 (0.891–1.974)	0.169
Model 3	1.402 (1.039–1.914)	0.027	2.294 (1.368–4.064)	0.001	1.344 (0.911–2.008)	0.138
Spirometry patterns
Model 1		0.017		0.026		0.327
Normal	1 (Reference)		1 (Reference)		1 (Reference)
Restrictive	1.368 (1.015–1.853)	0.040	1.942 (1.182–3.295)	0.008	1.405 (0.902–2.186)	0.132
Obstructive	1.134 (0.746–1.690)	0.550	1.386 (0.598–2.962)	0.428	1.061 (0.602–1.810)	0.833
Mixed	1.777 (1.225–2.558)	0.003	2.243 (1.192–4.183)	0.013	1.507 (0.840–2.614)	0.164
Model 2		0.047		0.017		0.387
Normal	1 (Reference)		1 (Reference)		1 (Reference)
Restrictive	1.280 (0.947–1.738)	0.109	1.993 (1.204–3.398)	0.007	1.311 (0.838–2.050)	0.235
Obstructive	1.180 (0.772–1.768)	0.437	1.237 (0.512–2.764)	0.622	1.132 (0.636–1.954)	0.664
Mixed	1.697 (1.171–2.439)	0.006	2.392 (1.262–4.496)	0.008	1.599 (0.892–2.774)	0.112
Model 3		0.049		0.009		0.513
Normal	1 (Reference)		1 (Reference)		1 (Reference)
Restrictive	1.382 (0.995–1.932)	0.054	2.299 (1.322–4.175)	0.003	1.364 (0.871–2.133)	0.174
Obstructive	1.134 (0.699–1.790)	0.601	1.634 (0.588–3.950)	0.323	1.268 (0.718–2.171)	0.404
Mixed	1.718 (1.147–2.559)	0.009	2.784 (1.399–5.581)	0.004	1.393 (0.771–2.437)	0.265

Model 1 was adjusted for age and sex. Model 2 was adjusted for all variables in Model 1 and further adjusted for other baseline clinical variables known to have an association with the outcome (BNP, haemoglobin, eGFR, and sodium). Model 3 was adjusted for variables related with the outcome at P<0.10 in the univariate analysis. Abbreviations as in Tables 1–3.

Abnormal PFT and Outcome in HFpEF and HFrEF

Subsequently, survival curves were estimated separately for the HFpEF (n=215) and HFrEF (n=320) groups using the Kaplan-Meier method (Figure 3C,D). Abnormal PFT was associated with a higher risk of CV mortality or hospitalization for HF in HFpEF (unadjusted HR 2.144; 95% CI 1.357–3.531; P<0.001; Table 3). After adjusting for significant variables in the univariate analysis (Model 3), abnormal PFT was an independent predictor of the primary endpoint in HFpEF (HR 2.294; 95% CI 1.368–4.064; P=0.001; Table 4). Conversely, in HFrEF, abnormal PFT was a predictor of the primary endpoint in univariate analysis (unadjusted HR 1.731; 95% CI 1.201–2.531; P=0.003), but not after multivariate analysis (P=0.169 in Model 2 and P=0.138 in Model 3). When we examined whether spirometric patterns of abnormal PFT predict the risk of the primary endpoint, restrictive and mixed patterns, but not the obstructive pattern, were strongly associated with CV mortality or hospitalization for HF in HFpEF (HR 2.299 [95% CI 1.322–4.175; P=0.003] and 2.784 [95% CI 1.399–5.581; P=0.004], respectively; Model 3). In HFrEF, restrictive and mixed patterns were also associated with poor prognosis in univariate analyses; however, after adjustment for covariates, the prognostic impact of these patterns was no longer observed (Table 4).

Discussion

The major findings of this study are as follows: (1) an abnormal PFT, particularly the restrictive pattern, was prevalent in ADHF and its subtypes (HFpEF and HFrEF); (2) age, AF, lung disease, and CTR were significant predictors of abnormal PFT; (3) abnormal PFTs, as well as the restrictive and mixed patterns, were independent predictors of the primary composite endpoint in patients with HFpEF, but not in those with HFrEF.

It is possible that lung function abnormalities may be due to either HF itself or to the presence of concomitant comorbid lung diseases. Indeed, any lung disease was found to predict abnormal PFTs or the mixed pattern in this study. However, although the prevalence of abnormal PFTs was 63.0% in this study, only 152 of all 657 patients (23.1%) and 126 of 414 patients with abnormal PFTs (30.4%) had any lung disease. We therefore believe that even though the results in some patients were possibly affected by lung diseases, HF itself can lead to abnormal PFTs. Although the mechanism underlying HF-induced abnormalities in lung function remains unclear, these abnormalities are thought to be linked to congestion and increased heart size, which promote airway obstruction and lung restriction.²²^–²⁵ Previous studies have strongly suggested that at least some of the impairment in lung function has been linked to pulmonary congestion because of a significant improvement in lung function after treatment of the acute HF episode, transplantation, or valve replacement.¹²^,¹³^,²⁶ As the left ventricular filling pressure increases, pulmonary congestion and interstitial edema develop, causing a reduction in lung volume and lung diffusing capacity for carbon monoxide (D_LCO), whereas the FEV₁/FVC remains normal.²⁵^,²⁷^,²⁸ Thus, the FVC may decline with even moderate congestion, whereas low a FEV₁/FVC ratio may emerge in decompensated HF due to bronchial wall edema. However, PFTs were performed in the non-acute setting before discharge in this study; at that time, almost all patients except for 4 (0.6%) were found not to have residual pulmonary congestion on chest radiographs.

In addition, we investigated the effects of BNP and TRV on respiratory function, which are not direct markers of congestion but are associated with invasive pulmonary capillary wedge pressure (PCWP) and left ventricular filling pressure. TRV >2.9 m/s and BNP at discharge were not associated with abnormal PFTs. Moreover, the length of hospital stay in patients with HF is markedly longer in Japan (15–21 days) than in Western countries (4–9 days).²⁹ The prolonged length of stay for HF hospitalization likely contributes to improved pulmonary congestion before discharge. Therefore, even if a few patients present with residual pulmonary congestion, we believe that this would have had only a minor effect on the results of this study.

Another factor to consider is the possible interaction between increased heart size and lung function. We demonstrated that CTR was an independent predictor of abnormal PFT. Cardiomegaly resulted in reductions in lung volumes and contributed to the overall restrictive pattern often reported in patients with HF and adult congenital heart disease.²²^–²⁴ Patients with HF commonly have chronic ventricular volume and/or pressure overload, presenting with enlargement of the cardiac chambers, which is reflected by increased CTR on chest radiographs. In our population, cardiomegaly on chest radiographs was more likely to relate to left and right atrial enlargement, right ventricular enlargement, and right ventricular dysfunction (Supplementary Table). Conversely, although AF was also a predictor of abnormal PFT, left atrial volume index was not associated with abnormal PFT. AF may affect pulmonary function, or vice versa, because impaired pulmonary function was associated with a higher incidence of AF in previous studies.³⁰^,³¹

Together, these observations indicate that, in HF, abnormal PFTs before discharge can be caused by HF itself, as well as coexisting lung disease.

Lung function declines slowly throughout adult life, even in healthy people.³² Previous studies have shown that the lung function parameters FEV₁ and FVC, but not the FEV₁/FVC ratio, were related to all-cause mortality and an increased risk of CV disease in population-based individuals without lung disease.³³^–³⁵ These declines in lung function were associated with comorbidities including diabetes, renal dysfunction, and neurocognitive disease.³⁴ In addition, it has been shown that low-grade systemic inflammation is present among healthy men and older adults with abnormal spirometric findings without established lung disease.³⁵^,³⁶ It is speculated that lung function abnormalities may be an important indicator of vulnerability or inherent susceptibility to the development of chronic diseases. Alternatively, lung function abnormalities may be causally related to systemic inflammatory pathways, with multiple organ effects.³⁴

Many comorbidities may also be involved in the pathophysiology of HFpEF. Although these comorbidities are common among patients with HFpEF and identify those at higher risk of adverse outcomes, the relative contribution of comorbidities to outcomes in HFpEF vs. HFrEF remains contentious.³⁷ However, in HFpEF, abnormal PFT showed a strong effect on CV mortality or hospitalization for HF in both univariate and multivariable analysis, whereas abnormal PFT was no longer significantly associated with adverse outcome in HFrEF after adjusting for other important variables. Prior HF hospitalization, male sex, reduced eGFR, low Hb, and the prevalence of AF were prognostic factors in patients with HFrEF. Although the pathophysiological mechanisms underlying the worsening of prognosis in patients with HFpEF and abnormal PFT are not yet fully understood, the burden of abnormal PFT can play a more significant role in outcomes in HFpEF than in HFrEF.

Clinical Implications

In patients with HF, it is important to identify those with the highest risk. In addition to known risk factors in patients with HF, especially those with HFpEF, the prognostic importance of abnormal PFT is new information. Considering the high prevalence of abnormal PFT in patients with ADHF, these results should be seen as a key step in understanding the influence of HF on lung function and its implications for prognosis. Given the impaired respiratory system in HF, the lung could thus be considered a target organ in HF. In addition, specifically targeting comorbidities in this population may not only reduce the burden of the comorbidities themselves, but may also allow more targeted and successful HFpEF clinical trials. The restoration and maintenance of normal sinus rhythm in the treatment of AF is likely to improve lung function; however, further research is required to address its diagnostic and prognostic importance in this patient population.

Study Limitations

This study has several limitations. First, the present study was conducted in a single centre with a relatively small registry-based patient cohort. The validity of our findings in other populations remains to be established. Second, the diagnosis of restriction was based on FVC and not on total lung capacity. Thus, we may have overestimated the number of patients with reduced lung volumes,³⁸ especially those with air trapping; however, fewer patients with severe airway obstruction reduced this possibility. Third, spirometry was performed without a bronchodilator, except for some patients with COPD and asthma, because spirometry was used to assess lung function before discharge in patients with HF, but not to diagnose and decide on the treatment of the lung disease. Therefore, the FEV₁/FVC ratio may be underestimated, so that patients with COPD may be overdiagnosed. Even after excluding patients with the obstructive pattern from the analysis, abnormal PFTs with a restrictive or mixed pattern had a higher risk of poor outcomes compared with normal PFT (data not shown). Fourth, more than half the patients admitted for ADHF were >75 years old in this study. Comorbidities, including cognitive impairment and apraxia, may influence the quality of spirometry. Haynes³⁹ studied whether elderly patients (age ≥80 years) were able to achieve spirometry results comparable to those of a younger adult population (40–50 years). Overall, 92.6% of the elderly group and 91.5% of the control group spirometry tests satisfied all the American Thoracic Society/European Respiratory Society acceptability and reproducibility criteria (P=0.84). Although it is clear that elderly patients with marked cognitive impairment and apraxia are less likely to perform spirometry correctly, most elderly patients without severe cognitive impairment are able to produce quality spirometry data.³⁹ In the present study, spirometry was not performed in patients with marked cognitive impairment and bedridden patients at the discretion of the attending physician. Finally, the optimal timing of PFT to identify the prevalence and pattern of lung function abnormalities has not been clarified because the clinical implications for assessment of changes in lung function have not been well studied among HF patients. Further studies are required to validate the prognostic value of PFT measurments across the spectrum of ADHF.

Conclusions

We observed a high prevalence of lung function abnormalities, as measured by spirometry, in patients with ADHF. Abnormal PFT is an independent predictor of CV mortality or HF hospitalizations in ADHF, especially in HFpEF. Spirometry should be considered in all patients with ADHF because it has a remarkable value for risk stratification, even in patients without clinically apparent lung disease.

Acknowledgments

The authors thank Yoko Wada and Rika Nagao for their help with data acquisition and management.

Sources of Funding

This work was supported, in part, by a Nara Medical University Grant-in-Aid for a large-scale prospective cohort study on healthy life expectancy.

Disclosures

Y.S. is a member of Circulation Journal’s Editorial Team. The remaining authors have no conflicts of interest to disclose.

IRB Information

This study was approved by the Nara Medical University Ethics Committee (Reference no. 624).

Data Availability

The deidentified participant data will not be shared.

Supplementary Files

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-20-1069

References

1. van Deursen VM, Urso R, Laroche C, Damman K, Dahlström U, Tavazzi L, et al. Comorbidities in patients with heart failure: An analysis of the European Heart Failure Pilot Survey. Eur J Heart Fail 2014; 16: 103–111.
2. Mentz RJ, Kelly JP, von Lueder TG, Voors AA, Lam CS, Cowie MR, et al. Noncardiac comorbidities in heart failure with reduced versus preserved ejection fraction. J Am Coll Cardiol 2014; 64: 2281–2293.
3. Coats AJS. Ageing, demographics, and heart failure. Eur Heart J Suppl 2019; 21(Suppl L): L4–L7.
4. Bhatia RS, Tu JV, Lee DS, Austin PC, Fang J, Haouzi A, et al. Outcome of heart failure with preserved ejection fraction in a population-based study. N Engl J Med 2006; 355: 260–269.
5. Ather S, Chan W, Bozkurt B, Aguilar D, Ramasubbu K, Zachariah AA, et al. Impact of noncardiac comorbidities on morbidity and mortality in a predominantly male population with heart failure and preserved versus reduced ejection fraction. J Am Coll Cardiol 2012; 59: 998–1005.
6. Lam CS, Donal E, Kraigher-Krainer E, Vasan RS. Epidemiology and clinical course of heart failure with preserved ejection fraction. Eur J Heart Fail 2011; 13: 18–28.
7. Macchia A, Monte S, Romero M, D’Ettorre A, Tognoni G. The prognostic influence of chronic obstructive pulmonary disease in patients hospitalised for chronic heart failure. Eur J Heart Fail 2007; 9: 942–948.
8. Sato Y, Yoshihisa A, Oikawa M, Nagai T, Yoshikawa T, Saito Y, et al. Prognostic impact of chronic obstructive pulmonary disease on adverse prognosis in hospitalized heart failure patients with preserved ejection fraction: A report from the JASPER registry. J Cardiol 2019; 73: 459–465.
9. Apostolo A, Giusti G, Gargiulo P, Bussotti M, Agostoni P. Lungs in heart failure. Pulm Med 2012; 2012: 952741.
10. Olson TP, Johnson B, Borlaug BA. Impaired pulmonary diffusion in heart failure with preserved ejection fraction. JACC Heart Fail 2016; 4: 490–498.
11. Gehlbach BK, Geppert E. The pulmonary manifestations of left heart failure. Chest 2004; 125: 669–682.
12. Ulrik CS, Carlsen J, Arendrup H, Aldershvile J. Pulmonary function in chronic heart failure. Changes after heart transplantation. Scand Cardiovasc J 1999; 33: 131–136.
13. Domínguez JM, Keller B, Moisés J, Spitaleri G, Farrero M, Casal J, et al. Changes in pulmonary function in patients with advanced heart failure listed for heart transplantation. Transplant Proc 2019; 51: 3424–3427.
14. Iversen KK, Kjaergaard J, Akkan D, Kober L, Torp-Pedersen C, Hassager C, et al. The prognostic importance of lung function in patients admitted with heart failure. Eur J Heart Fail 2010; 12: 685–691.
15. Ho KK, Pinsky JL, Kannel WB, Levy D. The epidemiology of heart failure: The Framingham Study. J Am Coll Cardiol 1993; 22(4 Suppl A): 6A–13A.
16. The Committee of Pulmonary Physiology, The Japanese Respiratory Society. Guidelines for pulmonary function tests: Spirometry, flow-volume curve, diffusion capacity of the lung. Nihon Kokyuki Gakkai Zasshi 2004; Suppl: 1–56 (in Japanese).
17. Kubota M, Kobayashi H, Quanjer PH, Omori H, Tatsumi K, Kanazawa M. Reference values for spirometry, including vital capacity, in Japanese adults calculated with the LMS method and compared with previous values. Respir Investig 2014; 52: 242–250.
18. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG, Coats AJ, et al. 2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail 2016; 18: 891–975.
19. Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE Jr, Drazner MH, et al. 2013 ACCF/AHA guideline for the management of heart failure: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2013; 62: e147–e239.
20. Rudski LG, Lai WW, Afilalo J, Hua L, Handschumacher MD, Chandrasekaran K, et al. Guidelines for the echocardiographic assessment of the right heart in adults: A report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echocardiogr 2010; 23: 685–713.
21. Ong KC, Geske JB, Hebl VB, Nishimura RA, Schaff HV, Ackerman MJ, et al. Pulmonary hypertension is associated with worse survival in hypertrophic cardiomyopathy. Eur Heart J Cardiovasc Imaging 2016; 17: 604–610.
22. Alonso-Gonzalez R, Borgia F, Diller GP, Inuzuka R, Kempny A, Martinez-Naharro A, et al. Abnormal lung function in adults with congenital heart disease: Prevalence, relation to cardiac anatomy, and association with survival. Circulation 2013; 127: 882–890.
23. Olson TP, Beck KC, Johnson BD. Pulmonary function changes associated with cardiomegaly in chronic heart failure. J Card Fail 2007; 13: 100–107.
24. Agostoni P, Cattadori G, Guazzi M, Palermo P, Bussotti M, Marenzi G. Cardiomegaly as a possible cause of lung dysfunction in patients with heart failure. Am Heart J 2000; 140: e24.
25. Magnussen H, Canepa M, Zambito PE, Brusasco V, Meinertz T, Rosenkranz S. What can we learn from pulmonary function testing in heart failure? Eur J Heart Fail 2017; 19: 1222–1229.
26. Saxena P, Luthra S, Dhaliwal RS, Rana SS, Behera D. Early changes in pulmonary functions after mitral valve replacement. Ann Thorac Med 2007; 2: 111–117.
27. Chase SC, Fermoyle CC, Wheatley CM, Schaefer JJ, Olson LJ, Johnson BD. The effect of diuresis on extravascular lung water and pulmonary function in acute decompensated heart failure. ESC Heart Fail 2018; 5: 364–371.
28. Dalsgaard M, Plesner LL, Schou M, Kjøller E, Vestbo J, Iversen K. Prevalence of airflow obstruction in patients with stable systolic heart failure. BMC Pulm Med 2017; 17: 6.
29. Konishi M, Ishida J, Springer J, von Haehling S, Akashi YJ, Shimokawa H, et al. Heart failure epidemiology and novel treatments in Japan: Facts and numbers. ESC Heart Fail 2016; 3: 145–151.
30. Buch P, Friberg J, Scharling H, Lange P, Prescott E. Reduced lung function and risk of atrial fibrillation in the Copenhagen City Heart Study. Eur Respir J 2003; 21: 1012–1016.
31. Li J, Agarwal SK, Alonso A, Blecker S, Chamberlain AM, London SJ, et al. Airflow obstruction, lung function, and incidence of atrial fibrillation: The Atherosclerosis Risk in Communities (ARIC) study. Circulation 2014; 129: 971–980.
32. Celli BR, Halbert RJ, Isonaka S, Schau B. Population impact of different definitions of airway obstruction. Eur Respir J 2003; 22: 268–73.
33. Magnussen C, Ojeda FM, Rzayeva N, Zeller T, Sinning CR, Pfeiffer N, et al; Gutenberg Health Study investigators. FEV₁ and FVC predict all-cause mortality independent of cardiac function: Results from the population-based Gutenberg Health Study. Int J Cardiol 2017; 234: 64–68.
34. Duong M, Islam S, Rangarajan S, Leong D, Kurmi O, Teo K, et al; PURE investigators. Mortality and cardiovascular and respiratory morbidity in individuals with impaired FEV₁ (PURE): An international, community-based cohort study. Lancet Glob Health 2019; 7: e613–e623.
35. Engström G, Lind P, Hedblad B, Wollmer P, Stavenow L, Janzon L, et al. Lung function and cardiovascular risk: Relationship with inflammation-sensitive plasma proteins. Circulation 2002; 106: 2555–2560.
36. Georgiopoulou VV, Kalogeropoulos AP, Psaty BM, Rodondi N, Bauer DC, Butler AB, et al. Lung function and risk for heart failure among older adults: The Health ABC Study. Am J Med 2011; 124: 334–341.
37. Ergatoudes C, Schaufelberger M, Andersson B, Pivodic A, Dahlström U, Fu M. Non-cardiac comorbidities and mortality in patients with heart failure with reduced vs. preserved ejection fraction: A study using the Swedish Heart Failure Registry. Clin Res Cardiol 2019; 108: 1025–1033.
38. Scarlata S, Pedone C, Fimognari FL, Bellia V, Forastiere F, Incalzi RA. Restrictive pulmonary dysfunction at spirometry and mortality in the elderly. Respir Med 2008; 102: 1349–1354.
39. Haynes JM. Pulmonary function test quality in the elderly: A comparison with younger adults. Respir Care 2014; 59: 16–21.

Corresponding author

Register with J-STAGE for free!