- Email: [email protected]
Prognostic implications of arterial blood gases in acute decompensated heart failure
European Journal of Internal Medicine 22 (2011) 489–494
Contents lists available at ScienceDirect
European Journal of Internal Medicine j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e j i m
Original article
Prognostic implications of arterial blood gases in acute decompensated heart failure Gema Miñana a,⁎, Julio Núñez a, Pilar Bañuls b, Juan Sanchis a, Eduardo Núñez a, Rocio Robles a, Beatriz Mascarell a, Patricia Palau a, Francisco J. Chorro a, Angel Llàcer a a b
Cardiology Department, Hospital Clínico Universitario, INCLIVA, Universitat de Valencia, Valencia, Spain Respiratory Medicine Department, Hospital Clínico Universitario, Universitat de Valencia, Valencia, Spain
a r t i c l e
i n f o
Article history: Received 13 November 2010 Received in revised form 18 January 2011 Accepted 21 January 2011 Available online 25 February 2011 Keywords: Arterial blood gases Acute decompensated heart failure Prognosis
a b s t r a c t Background: The prognostic value of arterial blood gases (ABG) in patients with acute decompensated heart failure (ADHF) is not well-established. We therefore conducted the present study to determine the relationship between ABG on admission and long-term mortality in patients with ADHF. Methods: We studied 588 patients consecutively admitted to our department with ADHF. ABG and classical prognostic variables were determined at patients' arrival to the emergency department. The independent association among the main variables of ABG (pO2, pCO2 and pH) and mortality was assessed with Cox regression analysis. Results: At a median follow-up of 23 months, 221 deaths (37.6%) were registered. 308 (52.4%), 54 (9.2%) and 50 (8.5%) patients showed hypoxemia (pO2 b 60 mm Hg), hypercapnia (pCO2 N 50 mm Hg) and acidosis (pH b 7.35), respectively. Patients with hypoxemia, hypercapnia and acidosis did not show higher mortality rates (38% vs. 37.1%, 42.6% vs. 37.1%, and 48% vs. 36.6%, respectively; p-value = ns for all comparisons). In multivariate analysis, after adjusting for well-known prognostic covariates, pO2, pCO2 and pH did not show a significant association with mortality. Hazard ratios (HR) for these variables were: pO2, per increase in 10 mm Hg: 0.99 (95% CI: 0.90–1.09), p = 0.861; pCO2, per increase in 10 mm Hg: 1.12 (95% CI: 0.91–1.39), p = 0.262; pH per increase in 0.1: 1.01 (95% CI: 0.99–1.04), p = 0.309. When dichotomizing these variables according to established cut-points, the HR were: hypoxemia (pO2 b 60 mm Hg):1.07 (95% CI: 0.81–1.40), p = 0.637; hypercapnia (pCO2 N 50 mm Hg): 0.98 (95% CI: 0.62–1.57), p = 0.952; acidosis (pH b 7.35): 1.38 (95% CI: 0.87–2.19), p = 0.173. Conclusion: In patients admitted with ADHF, admission arterial pO2, pCO2 and pH were not associated with all-cause long-term mortality. © 2011 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
1. Introduction Approximately 10% to 15% of all emergency department presentations are due to shortness of breath, mostly secondary to heart failure (HF) or lung disease. The most frequent complain of patients with acute HF is dyspnea [1,2]. Recently, the European Society of Cardiology (ESC) has proposed a clinical classification of acute HF into five well-established categories. Among them, acute decompensated HF (ADHF), which is the most common presentation in acute HF, was defined as a progressive worsening of a known chronic HF on treatment and evidence of systemic and pulmonary congestion. As expected, dyspnea is the most frequent complaint in patients with ADHF [3].
⁎ Corresponding author at: Servicio de Cardiología, Hospital Clínico Universitario, Avda. Blasco Ibáñez 17, 46010 Valencia, Spain. Tel.: + 34 617551562; fax: + 34 963862658. E-mail address: [email protected] (G. Miñana).
In patients with acute dyspnea, arterial blood gases (ABG) are one of the main tools for decision making in the emergency department. ABG are a basic tool in the diagnosis work-up for these patients and also for guiding therapy in patients with chronic obstructive pulmonary disease (COPD) exacerbation [4,5], asthma [6], pneumonia [7] or pulmonary embolism [8,9]. However, its prognostic value in patients with ADHF is not well-established. The ESC heart failure guidelines [3] recommend assessing ABG in patients with acute HF, with the rationale that acidosis, due to poor tissue perfusion or CO2 retention is associated with a poor prognosis, especially in those cases with a clinical abrupt-onset presentation, such as acute pulmonary edema, hypertensive emergency or cardiogenic shock. However, we believe this rationale is merely speculative in ADHF, since there is no evidence in the literature that has demonstrated its association with poor prognosis. The present study was conducted to determine the relationship between the main variables traditionally measured in the ABG sample (pO2, pCO2 and pH) at hospital admission and long-term mortality in patients with ADHF.
0953-6205/$ – see front matter © 2011 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.ejim.2011.01.014
490
G. Miñana et al. / European Journal of Internal Medicine 22 (2011) 489–494
2. Methods
2.2. Statistical analysis
2.1. Study population and protocol
Continuous variables were expressed as mean (95% CI) or median (interquartile range) as appropiate. Discrete variables were presented as percentages. Baseline characteristics were compared among those patients with and without hypoxemia, hypercapnia and acidosis. Allcause mortality rates were depicted among time using the Kaplan– Meier method, and their differences tested by the Peto-Peto Prentice test. The independent association between each component of the ABG (pO2, pCO2 and pH) and mortality was assessed using the Cox regression analysis. Candidate covariates for the initial multivariable model were selected based on previous medical knowledge. All variables listed in Table 1 were tested with prognostic purposes. Then, a reduced and parsimonious model was derived by using backward stepdown selection, leading to a final model with the following covariates: gender, age, previous admission for acute HF, last NYHA class measured under stable conditions, dyslipidemia, heart rate on admission, systolic blood pressure on admission, serum creatinine, serum sodium and haemoglobin. The proportionality assumption for the hazard function over time was tested by means of the Schoenfeld residuals. The discrimination and calibration of the model were assessed using the Harrell's C-statistics and the Gronnesby and Borgan test [10] respectively. A 2-sided p-value of b 0.05 was considered to be statistically significant for all analyses. All analyses were performed using STATA 11 (StataCorp. 2009. Stata Statistical Software: Release 11. College Station, TX: StataCorp LP).
We studied a cohort of 1181 patients consecutively admitted in our institution from January 2003 to July 2009, with a diagnosis of acute HF. Following current guidelines [3], diagnosis of acute HF was defined as a rapid onset of symptoms and signs secondary to abnormal cardiac function plus any objective evidence of structural or functional abnormality of heart at rest (cardiomegaly, third heart sound, cardiac murmur, abnormality of the echocardiogram or raised natriuretic peptide levels); as a subtype of acute HF, ADHF was defined as a progressive worsening of a known chronic HF on treatment and evidence of systemic and pulmonary congestion. A trained cardiologist confirmed the diagnosis of ADHF in all patients at hospital arrival. Therefore, we excluded for the present analysis acute HF cases other than ADHF, such as acute pulmonary edema (n = 258), hypertensive acute HF (n = 98), cardiogenic shock (n = 14) and right ventricular HF (n = 7). Among 804 patients with ADHF, 216 were also excluded for the following reasons: 190 (23.6%) because ABG determinations were not available (the physician in charge did not consider it necessary) and 26 (3.2%) because ABG were assessed after starting treatment. Thus, the final study cohort included 588 ADHF patients (see flow chart, Fig. 1). Demographic information, medical history, vital signs, 12-lead electrocardiogram, laboratory and drug utilization were routinely collected in the emergency department and throughout the hospital course following pre-established registry questionnaires. ABG were determined at patients' arrival to the emergency department with a FiO2 of 21%. Left ventricular ejection fraction (LVEF) was assessed with two-dimensional echocardiography in all patients during hospitalization (Agilent Sonos 5500-Phillips). All patients received intravenous treatment with furosemide at least during the first 48 hours of admission. Treatment with angiotensin converting enzyme inhibitors (ACEI), angiotensin II receptor blockers (ARB), betablockers, aldosterone antagonists, anticoagulants and other therapeutic strategies was individualized following established guidelines [3]. Patients were followed until death. They were censored, however, by having undergone cardiac transplantation, cardiac valve replacement surgery, or lost to follow-up. All-cause mortality was selected as the main clinical endpoint. Patients' clinical status and endpoint ascertainment were routinely evaluated by trained cardiologists either during recurrent hospitalizations or outpatient visits.
3. Results 3.1. Baseline characteristics The mean age of the study population was 74 ± 11 years; 306 patients (52%) were female, 244 (41.5%) had been previously admitted for acute HF and 317 (53.9%) exhibited left ventricular ejection fraction N50%. Three hundred thirty-six (57.1%) patients showed an abnormal ABG analysis, as follows: 308 (52.4%), 54 (9.2%) and 50 (8.5%) patients showed hypoxemia (pO2 b 60 mm Hg), hypercapnia (pCO2 N 50 mm Hg) and acidosis (pH b 7.35), respectively. Tables 1, 2 and 3 show the clinical characteristics of the study population according to the presence of hypoxemia, hypercapnia and acidosis, respectively. Hypoxemia was more frequent among females, those with higher means of age, systolic blood pressure and LVEF and without peripheral edema. Likewise, lower means of serum creatinine and brain natriuretic peptide (BNP) were observed in patients with hypoxemia. Interestingly, no differences in cardiovascular risk factors, etiology or comorbidity were found among patients with or without hypoxemia. Hypercapnia was more prevalent in patients with NYHA class III/IV (last measurement under clinically stable conditions, and before the index admission), females, diabetics, COPD patients and in those with peripheral edema. Likewise, patients with hypercapnia showed higher means of systolic blood pressure, left ventricular ejection fraction and serum sodium and lower means of BNP. Finally, patients with acidosis showed higher means of systolic blood pressure and serum creatinine and were treated less frequently with spironolactone and ACEI. 3.2. Long-term mortality and arterial blood gases in ADHF
Fig. 1. Flow chart describing patients excluded. ADHF: acute decompensated heart failure; APE: acute pulmonary edema; HTA-HF: hypertensive heart failure; RVHF: right ventricle heart failure; ABG: arterial blood gases.
At a median follow-up of 23 months (IQR: 7–43), 221 deaths (37.6%) were registered: 32 occurred during the index hospitalization and 189 after discharge. Crude mortality rates were not different for patients with hypoxemia, hypercapnia or acidosis (38% vs. 37.1%, p = 0.833; 42.6% vs. 37.1%, p = 0.426; and 48% vs. 36.6%, p = 0.112, respectively). Kaplan–Meier curves confirmed that cumulative risk of
G. Miñana et al. / European Journal of Internal Medicine 22 (2011) 489–494 Table 1 Baseline characteristics stratified by the presence of hypoxemia.
491
Table 2 Baseline characteristics stratified by the presence of hypercapnia.
pO2 b 60 mm Hg
pO2 ≥ 60 mm Hg p-value
pCO2 N 50 mm Hg pO2 ≤ 50 mm Hg p-value
n = 308 (52.4%)
n = 280 (47.6%)
n = 54 (9.2%)
Demographic and medical history Age, years Gender (male), n (%) Debut AHF, n (%) Hypertension, n (%) Dyslipidemia, n (%) Current smoker, n (%) Diabetes mellitus, n (%) Ischemic heart disease, n (%) Valvular heart disease, n (%) NYHA class III/IV, n (%)a COPD, n (%) Stroke, n (%) PAD, n (%) Radiological pleural effusion, n (%) Peripheral edema, n (%) Previous use of diuretics, n (%) Previous use of beta-blockers, n (%) Previous use of ACEi/ARB, n (%) Previous use of statins, n (%)
75 134 182 233 132 22 134 100 83 65 79 39 19 156 183 196 77 140 83
73 148 162 205 112 24 114 105 74 64 61 22 21 154 194 193 57 126 68
Vital signs Heart rate, bpm Systolic blood pressure, mm Hg Diastolic blood pressure, mm Hg
97 (94–100) 147 (143–150) 79 (77–81)
ECG Atrial fibrillation, n (%) QRS N 120 ms, n (%)
145 (47.1) 91 (29.6)
Biomarkers Haemoglobin, g/dl Serum creatinine, mg/dl Uric acid, mg/dl Sodium, mEq/l Troponin I, ng/ml BNP, pg/mlb
n = 534 (90.8%)
0.024 0.023 0.762 0.499 0.482 0.527 0.493 0.211 0.908 0.608 0.271 0.055 0.522 0.291 0.013 0.175 0.179 0.912 0.460
Demographic and medical history Age, years 74 (72–77) Gender (male), n (%) 17 (31.5) Debut AHF, n (%) 28 (51.8) Hypertension, n (%) 45 (83.3) Dyslipidemia, n (%) 25 (46.3) Current smoker, n (%) 0 Diabetes mellitus, n (%) 30 (55.6) Ischemic heart disease, n (%) 20 (37.0) Valvular heart disease, n (%) 12 (22.2) NYHA class III/IV, n (%)a 19 (35.2) COPD, n (%) 25 (46.3) Stroke, n (%) 5 (9.3) PAD, n (%) 3 (5.6) Radiological pleural effusion, n (%) 26 (48.1) Peripheral edema, n (%) 44 (81.5) Previous use of diuretics, n (%) 38 (70.4) Previous use of beta-blockers, n (%) 13 (24.1) Previous use of ACEi/ARB, n (%) 27 (50.0) Previous use of statins, n (%) 18 (33.3)
73 265 316 393 219 46 218 185 145 110 115 56 37 284 333 351 121 239 133
101 (98–105) 140 (136–144) 79 (77–81)
0.067 0.006 0.885
Vital signs Heart rate, bpm Systolic blood pressure, mm Hg Diastolic blood pressure, mm Hg
100 (97–102) 142 (140–145) 79 (77–80)
0.449 0.007 0.171
144 (51.4) 78 (27.9)
0.292 0.651
ECG Atrial fibrillation, n (%) QRS N 120 ms, n (%)
257 (48.1) 153 (28.6)
0.118 0.880
12.5 (12.3–12.7) 12.6 (12.4–12.8) 1.26 (1.19–1.33) 1.38 (1.31–1.45) 8 (7.7–8.2) 8 (7.7–8.3) 139 (139–140) 139 (138–139) 0.23 (0.11–0.36) 0.27 (0.15–0.38) 138 (74–254) 154 (86–328)
0.485 0.022 0.894 0.209 0.705 0.036
Biomarkers Haemoglobin, g/dl Serum creatinine, mg/dl Uric acid, mg/dl Sodium, mEq/l Troponin I, ng/ml BNP, pg/mlb
12.8 1.27 7.7 141 0.19 107
12.5 (12.4–12.7) 1.32 (1.27–1.37) 8 (7.8–8.2) 139 (138–139) 0.26 (0.16–0.35) 152 (82–294)
0.276 0.563 0.289 0.005 0.688 0.011
Echocardiography LVEF, % LVEF b 40%, n (%) LAD, mm LVDD, mm Septum, mm Posterior wall, mm
53 66 44 55 11.6 11.4
(52–55) 50 (48–52) (21.4) 90 (32.1) (43–45) 45 (44–46) (54–56) 56 (55–58) (11.3–11.8) 11.2 (10.9–11.5) (11.2–11.6) 11.1 (10.9–11.4)
0.020 0.003 0.140 0.017 0.060 0.098
Echocardiography LVEF, % LVEF b 40%, n (%) LAD, mm LVDD, mm Septum, mm Posterior wall, mm
60 (56–63) 6 (11.1) 45 (42–47) 52 (50–55) 12 (11.2–12.7) 11.7 (11.0–12.4)
Medical treatment Beta-blockers, n (%) Diuretics, n (%) Spironolactone, n (%) ACEI, n (%) ARB, n (%) Statins, n (%) Oral anticoagulants, n (%) Nitrates, n (%) Digoxin, n (%)
137 303 44 130 93 99 131 60 89
(44.5) (98.4) (14.3) (42.2) (30.2) (32.1) (42.9) (19.5) (28.9)
0.320 0.847 0.341 0.649 0.360 0.782 0.753 0.464 0.707
Medical treatment Beta-blockers, n (%) Diuretics, n (%) Spironolactone, n (%) ACEI, n (%) ARB, n (%) Statins, n (%) Oral anticoagulants, n (%) Nitrates, n (%) Digoxin, n (%)
(73–76) (43.5) (59.1) (75.6) (42.9) (7.2) (43.5) (32.6) (26.9) (21.1) (25.6) (12.7) (6.2) (50.6) (59.4) (63.6) (25.0) (45.4) (26.9)
136 276 48 113 75 93 123 48 77
(71–74) (52.9) (57.9) (73.2) (40.0) (8.6) (40.7) (37.5) (26.5) (22.9) (21.8) (7.9) (7.5) (55.0) (69.3) (68.9) (20.4) (45.0) (24.3)
(48.6) (98.6) (17.1) (40.4) (26.8) (33.2) (44.2) (17.1) (27.5)
96 (89–104) 154 (145–163) 82 (77–87)
32 (59.3) 16 (29.6)
(12.4–13.2) (1.11–1.43) (7.0–8.3) (139–142) (0–0.42) (64–186)
21 (39.0) 54 (100) 6 (11.1) 17 (31.5) 16 (29.6) 13 (24.1) 22 (40.7) 12 (22.2) 21 (38.9)
(73–74) (49.6) (59.2) (73.6) (41.0) (8.6) (40.8) (34.7) (27.2) (20.6) (21.5) (10.5) (7.0) (53.2) (62.4) (65.7) (22.7) (44.8) (24.9)
51 150 44 56 11.3 11.2
(50–52) (28.1) (44–45) (55–57) (11.1–11.5) (11.1–11.4)
252 525 86 226 152 179 232 96 145
(47.2) (98.3) (16.1) (42.3) (28.5) (33.5) (43.9) (18.0) (27.1)
0.498 0.010 0.301 0.103 0.454 0.026 0.038 0.733 0.422 0.019 b 0.001 0.775 0.695 0.480 0.003 0.488 0.814 0.462 0.188
b 0.001 0.004 0.657 0.011 0.068 0.102
0.241 0.336 0.316 0.118 0.857 0.148 0.659 0.453 0.076
AHF: acute heart failure; NYHA: New York Heart Association; COPD: chronic pulmonary obstructive disease; PAD: peripheral arterial disease; ACEI: angiotensin converting enzyme inhibitor; ARB: angiotensin II receptor antagonist; BNP: brain natriuretic peptide; LVEF: left ventricular ejection fraction; LAD: left atrial diameter; LVDD: left ventricular diastolic diameter. Values are expressed as mean (95% CI). Categorical variables are presented as percentages. a Last NYHA functional class measured under clinically stable conditions. b Presented as median (interquartile range).
AHF: acute heart failure; NYHA: New York Heart Association; COPD: chronic pulmonary obstructive disease; PAD: peripheral arterial disease; ACEI: angiotensin converting enzyme inhibitor; ARB: angiotensin II receptor antagonist; BNP: brain natriuretic peptide; LVEF: left ventricular ejection fraction; LAD: left atrial diameter; LVDD: left ventricular diastolic diameter. Values are expressed as mean (95% CI). Categorical variables are presented as percentages. a Last NYHA functional class measured under clinically stable conditions. b Presented as median (interquartile range).
death did not differ among patients with hypoxemia, hypercapnia or acidosis (Fig. 2 a, b and c). In multivariable Cox regression analysis, pO2, pCO2 and pH did not show a significant association with all-cause mortality. Evaluated in a continuous metric, the hazard ratios (HR) for pO2 (per 10 mm Hg), pCO2 (per 10 mm Hg) and pH (per 0.1 scale) were: 0.99 (95% CI: 0.90–1.09), p = 0.861; 1.12 (95% CI: 0.91–1.39), p = 0.262 and; 1.01
(0.99–1.04), p = 0.309, respectively. The Harrell's C-statistics of the multivariate model that included pO2, pCO2 and pH as continuous showed high discriminative ability of 0.741; the Gronnesby and Borgan test of goodness-of-fit showed a good model calibration (p = 0.802).
492
G. Miñana et al. / European Journal of Internal Medicine 22 (2011) 489–494
Table 3 Baseline characteristics stratified by the presence of acidosis.
Demographic and medical history Age, years Gender (male), n (%) Debut AHF, n (%) Hypertension, n (%) Dyslipidemia, n (%) Current smoker, n (%) Diabetes mellitus, n (%) Ischemic heart disease, n (%) Valvular heart disease, n (%) NYHA class III/IV, n (%)a COPD, n (%) Stroke, n (%) PAD, n (%) Radiological pleural effusion, n (%) Peripheral edema, n (%) Previous use of diuretics, n (%) Previous use of beta-blockers, n (%) Previous use of ACEi/ARB, n (%) Previous use of statins, n (%) Vital signs Heart rate, bpm Systolic blood pressure, mm Hg Diastolic blood pressure, mm Hg ECG Atrial fibrillation, n (%) QRS N 120 ms, n (%)
pH b 7.35
pH ≥ 7.35
n = 50 (8.5%)
n = 538 (91.5%)
76 (74–79) 21 (42.0) 33 (66.0) 41 (82.0) 24 (48.0) 3 (6.0) 21 (42.0) 22 (44.0) 9 (18.0) 14 (28.0) 17 (34.0) 6 (12.0) 5 (10.0) 25 (50.0) 30 (60.0) 29 (58.0) 9 (18.0) 22 (44.0) 13 (26.0)
98 (89–106) 154 (144–164) 82 (77–87)
20 (40.0) 19 (38.0)
p-value
73 (72–74) 261 (48.5) 311 (57.8) 397 (73.8) 220 (40.9) 43 (8.0) 227 (42.2) 183 (34.1) 148 (27.6) 115 (21.4) 123 (22.9) 55 (10.2) 35 (6.5) 285 (53.0) 347 (64.5) 360 (66.9) 125 (23.2) 244 (45.3) 138 (25.6)
0.063 0.377 0.256 0.188 0.332 0.601 0.979 0.166 0.129 0.292 0.088 0.699 0.375 0.687 0.529 0.209 0.387 0.854 0.957
99 (97–102) 143 (140–145) 79 (77–80)
0.683 0.013 0.181
269 (50.0) 150 (27.9)
0.175 0.140
Biomarkers Haemoglobin, g/dl Serum creatinine, mg/dl Uric acid, mg/dl Sodium, mEq/l Troponin I, ng/ml BNP, pg/mlb
12.6 (12.1–13.1) 12.5 (12.4–12.7) 0.816 1.49 (1.29–1.68) 1.30 (1.25–1.35) 0.039 8 (7.3–8.6) 8 (7.8–8.2) 0.927 140 (139–141) 139 (138–139) 0.159 0.38 (0.10–0.66) 0.24 (0.15–0.33) 0.374 182 (96–373) 143 (81–281) 0.198
Echocardiography LVEF, % LVEF b 40%, n (%) LAD, mm LVDD, mm Septum, mm Posterior wall, mm
53 (50–57) 52 (50–53) 8 (16.0) 148 (27.5) 43 (41–45) 44 (44–45) 54 (52–56) 56 (55–56) 11.9 (11.3–12.5) 11.3 (11.1–11.6) 12 (11.4–12.6) 11.2 (11–11.4)
Medical treatment Beta-blockers, n (%) Diuretics, n (%) Spironolactone, n (%) ACEI, n (%) ARB, n (%) Statins, n (%) Oral anticoagulants, n (%) Nitrates, n (%) Digoxin, n (%)
21 (42.0) 50 (100) 2 (4.0) 14 (28.0) 14 (28.0) 14 (28.0) 16 (32.0) 13 (26.0) 13 (26.0)
252 (46.8) 529 (98.3) 90 (16.7) 229 (42.6) 154 (28.6) 178 (33.1) 238 (44.6) 95 (17.7) 153 (28.4)
0.461 0.064 0.237 0.348 0.142 0.006
0.510 0.357 0.006 0.041 0.925 0.458 0.103 0.162 0.712
AHF: acute heart failure; NYHA: New York Heart Association; COPD: chronic pulmonary obstructive disease; PAD: peripheral arterial disease; ACEI: angiotensin converting enzyme inhibitor; ARB: angiotensin II receptor antagonist; BNP: brain natriuretic peptide; LVEF: left ventricular ejection fraction; LAD: left atrial diameter; LVDD: left ventricular diastolic diameter. Values are expressed as mean (95% CI). Categorical variables are presented as percentages. a Last NYHA functional class measured under clinically stable conditions. b Presented as median (interquartile range).
Similarly, non-significant results were observed by dichotomizing these variables with established cut points. The adjusted HRs for hypoxemia (pO2 b 60 mm Hg), hypercapnia (pCO2 N 50 mm Hg) and acidosis (pH b 7.35) were: 1.07 (95% CI: 0.81–1.40), p = 0.637; 0.98 (95% CI: 0.62–1.57), p = 0.952; and 1.38 (0.87–2.19), p = 0.173, respectively (Fig. 3).
Fig. 2. Overall survival stratified according to pO2 (a), pCO2 (b) and pH (c) values.
No significant interactions were found between hypoxemia, hypercapnia and acidosis with any of the most representative subgroups such as being older than 65, female, those with preserved systolic function, ischemic heart disease and previous admission for AHF, by showing not only the lack of significance but a similar direction of the effect. Because the availability of ABG may be considered a potential surrogate of clinical severity, we tested the prognostic value of having performed an ABG at patients' arrival, but this time using the entire population of patients with ADHF (n = 804). The adjusted HR for mortality showed no significant differences between patients with or without ABG measurements [HR: 1.25 (95% CI: 0.92–1.70), p = 0.162], a result that diminished the likelihood of significant selection bias in our results.
G. Miñana et al. / European Journal of Internal Medicine 22 (2011) 489–494
Fig. 3. All-cause mortality and ABG variables. Multivariate model adjusted by gender, age, first admission for AHF, last NYHA class measured under stable conditions, hypercholesterolemia, heart rate, systolic blood pressure, serum creatinine and serum sodium.
4. Discussion The present study showed, in a large and non-selected population with ADHF and after a thorough multivariable adjustment, that ABG (pO2, pCO2 and pH) determined at patients' arrival to the emergency department are not associated with all-cause long-term mortality. We believe that the present results may bring to the notice of the emergency physicians to avoid the routine use of ABG results for prognostic assessment in patients during an ADHF episode.
493
47 patients with severe chronic congestive HF before and one year after heart transplantation. These authors showed that restrictive ventilatory defect induced by chronic HF was reversible, whereas the exception of the reduction in diffusion lung capacity for carbon monoxide (DLCO) was not improved, which probably reflected permanent changes in the lung vasculature. In addition, widely used treatments in patients with HF such as ACEI or ARB have been shown to have pleiotropic beneficial effects on lung function [12]. These changes have been mainly attributed to the reduction of levels of angiotensin II at the level of the pulmonary microvasculature and increase exposure to prostaglandins, especially prostacyclin and nitric oxide, as a result of increased local kinin concentrations. Even considering the limitation of these studies, which include small sample sizes and selected patients, they reinforce the idea that gas exchange disturbances in ADHF are mainly transitory, unlike the dyspnea that is consequence of a primary lung disease. Such reversibility is what made our results relevant. Thus, we speculate that most of patients with ADHF show reversible lung function disturbances that are limited to acute episodes. 4.3. Clinical implications Due to a lack of evidence linking ABG with long-term mortality, long-term prognosis assessment should not be based on data from admission arterial pO2, pCO2 and pH. As stated by guidelines, a thorough clinical, electrocardiographic, radiologic and biomarker evaluation should be standard medical practice in these patients.
4.1. ADHF prognosis 4.4. Strengths and limitations Recent ESC heart failure guidelines [3] encourage the use of acute HF classification. ADHF syndrome has become the most prevalent presentation from all acute HF syndromes, with 1-year mortality in most registries about 20–30% [11,12]; these figures frequently exceed the mortality rates observed in other more abrupt dyspneic presentations of acute HF, such as acute pulmonary edema or hypertensive acute HF [13,14]. Thus, it has been recommended to base the initial risk stratification of patients with ADHF on a rigorous clinical, radiologic, electrocardiographic and biomarker evaluation [3]. Based on our results, the addition of ABG results did not provide any additional prognostic utility to the ESC recommendations. Moreover, uniform results were observed in the most important subgroups of patients, such as women, preserved systolic function, previous admissions for AHF and ischemic etiology. In addition, the fact that patients with no ABG (because the physician in charge did not consider them necessary) have similar prognosis as patients with ABG assessment further supports our results.
At present, this is the first study performed in a large and representative population of patients with ADHF, and analyzed with an appropriate methodology, which suggests that the main variables of ABG did not have prognostic implications on long-term mortality. Some limitations, however, should be addressed. First, determination of BNP and other biomarker-related variables was assessed during patient's hospitalization and not on patient's arrival to the emergency department; this fact precludes their use in the multivariable model. Second, due to its observational nature, the possibility of selection bias and/or residual confounding from unknown or unmeasured covariates cannot be excluded. Third, this is a singlecenter registry which may limit the transportability of our conclusion to other populations. Fourth, having a low sample size in certain subgroups precludes obtaining robust estimates regarding a possible differential effect of ABG in these subgroups. 4.5. Conclusions
4.2. Biological interpretation of the present results In acute HF, lung function disturbances are related to reduction in vital and total lung capacity, lung diffusion capacity and lung compliance. Consequently, resistance to air flow is moderately increased, while total and functional residual volume remain normal [15]. Usually, patients with chronic left ventricular failure exhibited a restrictive ventilatory defect [16,17] characterized by a reduction in vital capacity resulting from the replacement of air in the lungs by blood or interstitial fluid. Previous studies have shown that lung function abnormalities can be reduced, or even resolved, with different treatments, including ACEI [18–20], ultrafiltration [21,22] and heart transplantation [23,24]. In a classic study, Light et al. [25] evaluated the changes in pulmonary function in 28 patients hospitalized for congestive HF. Initially, patients had both obstructive and restrictive ventilatory dysfunction that rapidly improved with treatment initiation. In another study, Niset et al. [23] evaluated the reversibility of the lung dysfunction in
In patients with ADHF, admission arterial pO2, pCO2 and pH were not associated with all-cause long-term mortality. Learning points • In patients admitted with acute decompensated heart failure, admission arterial pO2, pCO2 and pH are not associated with allcause long-term mortality. • Due to a lack of evidence linking ABG with long-term mortality, long-term prognosis assessment should not be based on data from admission arterial pO2, pCO2 and pH. As stated by guidelines, a thorough clinical, electrocardiographic, radiologic and biomarker evaluation should be standard medical practice in these patients. Conflict of interest None to declare.
494
G. Miñana et al. / European Journal of Internal Medicine 22 (2011) 489–494
References [1] Adams Jr KF, Fonarow GC, Emerman CL, LeJemtel TH, Costanzo MR, Abraham WT, et al, ADHERE Scientific Advisory Committee and Investigators. Characteristics and outcomes of patients hospitalized for heart failure in the United States: rationale, design, and preliminary observations from the first 100,000 cases in the Acute Decompensated Heart Failure National Registry (ADHERE). Am Heart J 2005;149: 209–16. [2] Fonarow GC, ADHERE Scientific Advisory Committee. The Acute Decompensated Heart Failure National Registry (ADHERE): opportunities to improve care of patients hospitalized with acute decompensated heart failure. Rev Cardiovasc Med 2003;4(Suppl 7):S21–30. [3] European Society of Cardiology; Heart Failure Association of the ESC (HFA); European Society of Intensive Care Medicine (ESICM), Dickstein K, Cohen-Solal A, Filippatos G, McMurray JJ, Ponikowski P, Poole-Wilson PA, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the diagnosis and treatment of acute and chronic heart failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur J Heart Fail 2008;10:933–89. [4] Rabe KF, Hurd S, Anzueto A, Barnes PJ, Buist SA, Calverley P, et al , Global Initiative for Chronic Obstructive Lung Disease. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med 2007;176:532–55. [5] Garland A, Dawson NV, Altmann I, Thomas CL, Phillips RS, Tsevat J, et al, SUPPORT Investigators. Outcomes up to 5 years after severe, acute respiratory failure. Chest 2004;126:1897–904. [6] Camargo Jr CA, Rachelefsky G, Schatz M. Managing asthma exacerbations in the emergency department: summary of the National Asthma Education and Prevention Program Expert Panel Report 3 guidelines for the management of asthma exacerbations. J Emerg Med 2009;37:S6–S17. [7] Lim WS, Baudouin SV, George RC, Hill AT, Jamieson C, Le Jeune I, et al, Pneumonia Guidelines Committee of the BTS Standards of Care Committee. BTS guidelines for the management of community acquired pneumonia in adults: update. 2009. Thorax 2009;64(Suppl 3):iii1–iii55. [8] Wicki J, Perrier A, Perneger TV, Bounameaux H, Junod AF. Predicting adverse outcome in patients with acute pulmonary embolism: a risk score. Thromb Haemost 2000;84:548–52. [9] Aujesky D, Roy PM, Le Manach CP, Verschuren F, Meyer G, Obrosky DS, et al. Validation of a model to predict adverse outcomes in patients with pulmonary embolism. Eur Heart J 2006;27:476–81. [10] May S, Hosmer DW. Hosmer and Lemeshow type goodness-of-fit statistics for the Cox proportional hazards model. In: Balakrishnana N, Rao CR, editors. Advances in survival analysis: handbook of statistics, Vol 23. Amsterdam: Elsevier, NorthHolland; 2004. p. 383–94.
[11] Cleland JG, Swedberg K, Follath F, Komajda M, Cohen-Solal A, Aguilar JC, et al, Study Group on Diagnosis of the Working Group on Heart Failure of the European Society of Cardiology. The EuroHeart Failure survey programme—a survey on the quality of care among patients with heart failure in Europe. Eur Heart J 2003;24: 442–63. [12] Siirilä-Waris K, Lassus J, Melin J, Peuhkurinen K, Nieminen MS, FINN-AKVA Study Group. Characteristics, outcomes, and predictors of 1-year mortality in patients hospitalized for acute heart failure. Eur Heart J 2006;27:3011–7. [13] Núñez J, Núñez E, Fonarow GC, Sanchis J, Bodí V, Bertomeu-González V, et al, Differential prognostic effect of systolic blood pressure on mortality according to left-ventricular function in patients with acute heart failure. Eur J Heart Fail 2010;12:38–44. [14] Harjola VP, Follath F, Nieminen MS, Brutsaert D, Dickstein K, Drexler H, et al. Characteristics, outcomes, and predictors of mortality at 3 months and 1 year in patients hospitalized for acute heart failure. Eur J Heart Fail 2010;12:239–48. [15] Johnson BD, Beck KC, Olson LJ, O'Malley KA, Allison TG, Squires RW, et al. Pulmonary function in patients with reduced left ventricular function: influence of smoking and cardiac surgery. Chest 2001;120:1869–76. [16] Naum CC, Sciurba FC, Rogers RM. Pulmonary function abnormalities in chronic severe cardiomyopathy preceding cardiac transplantation. Am Rev Respir Dis 1992;145:1334–8. [17] Chua TP, Coats AJS. The lungs in chronic heart failure. Eur Heart J 1995;16:882–7. [18] Guazzi M, Marenzi G, Alimento M, Contini M, Agostoni P. Improvement of alveolar capillary membrane diffusing capacity with enalapril in chronic heart failure and counteracting effect of aspirin. Circulation 1997;95:1930–6. [19] Guazzi M, Melzi G, Agostoni PG. Comparison of change in respiratory function and exercise oxygen uptake with losartan vs. enalapril in congestive heart failure secondary to ischemic or idiopathic cardiomyopathy. Am J Cardiol 1997;80:1572–6. [20] Guazzi M, Palermo P, Pontone G, Susini F, Agostoni P. Synergistic efficacy of enalapril and losartan on exercise performance and oxygen consumption at peak exercise in congestive heart failure. Am J Cardiol 1999;84:1038–43. [21] Agostoni PG, Marenzi GC, Pepi M, Doria E, Salvioni A, Perego G. Isolated ultrafiltration in moderate heart failure. J Am Coll Cardiol 1993;21:424–31. [22] Agostoni P, Marenzi G, Lauri G, Perego G, Schianni M, Sganzerla P, et al. Sustained improvement in functional capacity after removal of body fluid with isolated ultrafiltration in chronic cardiac insufficiency. Failure of furosemide provide same result. Am J Med 1994;96:191–6. [23] Niset G, Ninane V, Antoine M, Yernault JC. Respiratory dysfunction in congestive heart failure: correction after heart transplantation. Eur Respir J 1993;6: 1197–201. [24] Hosenpud JD, Stibolt TA, Atwal K, Shelley D. Abnormal pulmonary function specifically related to congestive heart failure: comparison of patients before and after cardiac transplantation. Am J Med 1990;88:493–6. [25] Light RW, George RB. Serial pulmonary function in patients with acute heart failure. Arch Intern Med 1983;143:429–33.
Contents lists available at ScienceDirect
European Journal of Internal Medicine j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e j i m
Original article
Prognostic implications of arterial blood gases in acute decompensated heart failure Gema Miñana a,⁎, Julio Núñez a, Pilar Bañuls b, Juan Sanchis a, Eduardo Núñez a, Rocio Robles a, Beatriz Mascarell a, Patricia Palau a, Francisco J. Chorro a, Angel Llàcer a a b
Cardiology Department, Hospital Clínico Universitario, INCLIVA, Universitat de Valencia, Valencia, Spain Respiratory Medicine Department, Hospital Clínico Universitario, Universitat de Valencia, Valencia, Spain
a r t i c l e
i n f o
Article history: Received 13 November 2010 Received in revised form 18 January 2011 Accepted 21 January 2011 Available online 25 February 2011 Keywords: Arterial blood gases Acute decompensated heart failure Prognosis
a b s t r a c t Background: The prognostic value of arterial blood gases (ABG) in patients with acute decompensated heart failure (ADHF) is not well-established. We therefore conducted the present study to determine the relationship between ABG on admission and long-term mortality in patients with ADHF. Methods: We studied 588 patients consecutively admitted to our department with ADHF. ABG and classical prognostic variables were determined at patients' arrival to the emergency department. The independent association among the main variables of ABG (pO2, pCO2 and pH) and mortality was assessed with Cox regression analysis. Results: At a median follow-up of 23 months, 221 deaths (37.6%) were registered. 308 (52.4%), 54 (9.2%) and 50 (8.5%) patients showed hypoxemia (pO2 b 60 mm Hg), hypercapnia (pCO2 N 50 mm Hg) and acidosis (pH b 7.35), respectively. Patients with hypoxemia, hypercapnia and acidosis did not show higher mortality rates (38% vs. 37.1%, 42.6% vs. 37.1%, and 48% vs. 36.6%, respectively; p-value = ns for all comparisons). In multivariate analysis, after adjusting for well-known prognostic covariates, pO2, pCO2 and pH did not show a significant association with mortality. Hazard ratios (HR) for these variables were: pO2, per increase in 10 mm Hg: 0.99 (95% CI: 0.90–1.09), p = 0.861; pCO2, per increase in 10 mm Hg: 1.12 (95% CI: 0.91–1.39), p = 0.262; pH per increase in 0.1: 1.01 (95% CI: 0.99–1.04), p = 0.309. When dichotomizing these variables according to established cut-points, the HR were: hypoxemia (pO2 b 60 mm Hg):1.07 (95% CI: 0.81–1.40), p = 0.637; hypercapnia (pCO2 N 50 mm Hg): 0.98 (95% CI: 0.62–1.57), p = 0.952; acidosis (pH b 7.35): 1.38 (95% CI: 0.87–2.19), p = 0.173. Conclusion: In patients admitted with ADHF, admission arterial pO2, pCO2 and pH were not associated with all-cause long-term mortality. © 2011 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
1. Introduction Approximately 10% to 15% of all emergency department presentations are due to shortness of breath, mostly secondary to heart failure (HF) or lung disease. The most frequent complain of patients with acute HF is dyspnea [1,2]. Recently, the European Society of Cardiology (ESC) has proposed a clinical classification of acute HF into five well-established categories. Among them, acute decompensated HF (ADHF), which is the most common presentation in acute HF, was defined as a progressive worsening of a known chronic HF on treatment and evidence of systemic and pulmonary congestion. As expected, dyspnea is the most frequent complaint in patients with ADHF [3].
⁎ Corresponding author at: Servicio de Cardiología, Hospital Clínico Universitario, Avda. Blasco Ibáñez 17, 46010 Valencia, Spain. Tel.: + 34 617551562; fax: + 34 963862658. E-mail address: [email protected] (G. Miñana).
In patients with acute dyspnea, arterial blood gases (ABG) are one of the main tools for decision making in the emergency department. ABG are a basic tool in the diagnosis work-up for these patients and also for guiding therapy in patients with chronic obstructive pulmonary disease (COPD) exacerbation [4,5], asthma [6], pneumonia [7] or pulmonary embolism [8,9]. However, its prognostic value in patients with ADHF is not well-established. The ESC heart failure guidelines [3] recommend assessing ABG in patients with acute HF, with the rationale that acidosis, due to poor tissue perfusion or CO2 retention is associated with a poor prognosis, especially in those cases with a clinical abrupt-onset presentation, such as acute pulmonary edema, hypertensive emergency or cardiogenic shock. However, we believe this rationale is merely speculative in ADHF, since there is no evidence in the literature that has demonstrated its association with poor prognosis. The present study was conducted to determine the relationship between the main variables traditionally measured in the ABG sample (pO2, pCO2 and pH) at hospital admission and long-term mortality in patients with ADHF.
0953-6205/$ – see front matter © 2011 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.ejim.2011.01.014
490
G. Miñana et al. / European Journal of Internal Medicine 22 (2011) 489–494
2. Methods
2.2. Statistical analysis
2.1. Study population and protocol
Continuous variables were expressed as mean (95% CI) or median (interquartile range) as appropiate. Discrete variables were presented as percentages. Baseline characteristics were compared among those patients with and without hypoxemia, hypercapnia and acidosis. Allcause mortality rates were depicted among time using the Kaplan– Meier method, and their differences tested by the Peto-Peto Prentice test. The independent association between each component of the ABG (pO2, pCO2 and pH) and mortality was assessed using the Cox regression analysis. Candidate covariates for the initial multivariable model were selected based on previous medical knowledge. All variables listed in Table 1 were tested with prognostic purposes. Then, a reduced and parsimonious model was derived by using backward stepdown selection, leading to a final model with the following covariates: gender, age, previous admission for acute HF, last NYHA class measured under stable conditions, dyslipidemia, heart rate on admission, systolic blood pressure on admission, serum creatinine, serum sodium and haemoglobin. The proportionality assumption for the hazard function over time was tested by means of the Schoenfeld residuals. The discrimination and calibration of the model were assessed using the Harrell's C-statistics and the Gronnesby and Borgan test [10] respectively. A 2-sided p-value of b 0.05 was considered to be statistically significant for all analyses. All analyses were performed using STATA 11 (StataCorp. 2009. Stata Statistical Software: Release 11. College Station, TX: StataCorp LP).
We studied a cohort of 1181 patients consecutively admitted in our institution from January 2003 to July 2009, with a diagnosis of acute HF. Following current guidelines [3], diagnosis of acute HF was defined as a rapid onset of symptoms and signs secondary to abnormal cardiac function plus any objective evidence of structural or functional abnormality of heart at rest (cardiomegaly, third heart sound, cardiac murmur, abnormality of the echocardiogram or raised natriuretic peptide levels); as a subtype of acute HF, ADHF was defined as a progressive worsening of a known chronic HF on treatment and evidence of systemic and pulmonary congestion. A trained cardiologist confirmed the diagnosis of ADHF in all patients at hospital arrival. Therefore, we excluded for the present analysis acute HF cases other than ADHF, such as acute pulmonary edema (n = 258), hypertensive acute HF (n = 98), cardiogenic shock (n = 14) and right ventricular HF (n = 7). Among 804 patients with ADHF, 216 were also excluded for the following reasons: 190 (23.6%) because ABG determinations were not available (the physician in charge did not consider it necessary) and 26 (3.2%) because ABG were assessed after starting treatment. Thus, the final study cohort included 588 ADHF patients (see flow chart, Fig. 1). Demographic information, medical history, vital signs, 12-lead electrocardiogram, laboratory and drug utilization were routinely collected in the emergency department and throughout the hospital course following pre-established registry questionnaires. ABG were determined at patients' arrival to the emergency department with a FiO2 of 21%. Left ventricular ejection fraction (LVEF) was assessed with two-dimensional echocardiography in all patients during hospitalization (Agilent Sonos 5500-Phillips). All patients received intravenous treatment with furosemide at least during the first 48 hours of admission. Treatment with angiotensin converting enzyme inhibitors (ACEI), angiotensin II receptor blockers (ARB), betablockers, aldosterone antagonists, anticoagulants and other therapeutic strategies was individualized following established guidelines [3]. Patients were followed until death. They were censored, however, by having undergone cardiac transplantation, cardiac valve replacement surgery, or lost to follow-up. All-cause mortality was selected as the main clinical endpoint. Patients' clinical status and endpoint ascertainment were routinely evaluated by trained cardiologists either during recurrent hospitalizations or outpatient visits.
3. Results 3.1. Baseline characteristics The mean age of the study population was 74 ± 11 years; 306 patients (52%) were female, 244 (41.5%) had been previously admitted for acute HF and 317 (53.9%) exhibited left ventricular ejection fraction N50%. Three hundred thirty-six (57.1%) patients showed an abnormal ABG analysis, as follows: 308 (52.4%), 54 (9.2%) and 50 (8.5%) patients showed hypoxemia (pO2 b 60 mm Hg), hypercapnia (pCO2 N 50 mm Hg) and acidosis (pH b 7.35), respectively. Tables 1, 2 and 3 show the clinical characteristics of the study population according to the presence of hypoxemia, hypercapnia and acidosis, respectively. Hypoxemia was more frequent among females, those with higher means of age, systolic blood pressure and LVEF and without peripheral edema. Likewise, lower means of serum creatinine and brain natriuretic peptide (BNP) were observed in patients with hypoxemia. Interestingly, no differences in cardiovascular risk factors, etiology or comorbidity were found among patients with or without hypoxemia. Hypercapnia was more prevalent in patients with NYHA class III/IV (last measurement under clinically stable conditions, and before the index admission), females, diabetics, COPD patients and in those with peripheral edema. Likewise, patients with hypercapnia showed higher means of systolic blood pressure, left ventricular ejection fraction and serum sodium and lower means of BNP. Finally, patients with acidosis showed higher means of systolic blood pressure and serum creatinine and were treated less frequently with spironolactone and ACEI. 3.2. Long-term mortality and arterial blood gases in ADHF
Fig. 1. Flow chart describing patients excluded. ADHF: acute decompensated heart failure; APE: acute pulmonary edema; HTA-HF: hypertensive heart failure; RVHF: right ventricle heart failure; ABG: arterial blood gases.
At a median follow-up of 23 months (IQR: 7–43), 221 deaths (37.6%) were registered: 32 occurred during the index hospitalization and 189 after discharge. Crude mortality rates were not different for patients with hypoxemia, hypercapnia or acidosis (38% vs. 37.1%, p = 0.833; 42.6% vs. 37.1%, p = 0.426; and 48% vs. 36.6%, p = 0.112, respectively). Kaplan–Meier curves confirmed that cumulative risk of
G. Miñana et al. / European Journal of Internal Medicine 22 (2011) 489–494 Table 1 Baseline characteristics stratified by the presence of hypoxemia.
491
Table 2 Baseline characteristics stratified by the presence of hypercapnia.
pO2 b 60 mm Hg
pO2 ≥ 60 mm Hg p-value
pCO2 N 50 mm Hg pO2 ≤ 50 mm Hg p-value
n = 308 (52.4%)
n = 280 (47.6%)
n = 54 (9.2%)
Demographic and medical history Age, years Gender (male), n (%) Debut AHF, n (%) Hypertension, n (%) Dyslipidemia, n (%) Current smoker, n (%) Diabetes mellitus, n (%) Ischemic heart disease, n (%) Valvular heart disease, n (%) NYHA class III/IV, n (%)a COPD, n (%) Stroke, n (%) PAD, n (%) Radiological pleural effusion, n (%) Peripheral edema, n (%) Previous use of diuretics, n (%) Previous use of beta-blockers, n (%) Previous use of ACEi/ARB, n (%) Previous use of statins, n (%)
75 134 182 233 132 22 134 100 83 65 79 39 19 156 183 196 77 140 83
73 148 162 205 112 24 114 105 74 64 61 22 21 154 194 193 57 126 68
Vital signs Heart rate, bpm Systolic blood pressure, mm Hg Diastolic blood pressure, mm Hg
97 (94–100) 147 (143–150) 79 (77–81)
ECG Atrial fibrillation, n (%) QRS N 120 ms, n (%)
145 (47.1) 91 (29.6)
Biomarkers Haemoglobin, g/dl Serum creatinine, mg/dl Uric acid, mg/dl Sodium, mEq/l Troponin I, ng/ml BNP, pg/mlb
n = 534 (90.8%)
0.024 0.023 0.762 0.499 0.482 0.527 0.493 0.211 0.908 0.608 0.271 0.055 0.522 0.291 0.013 0.175 0.179 0.912 0.460
Demographic and medical history Age, years 74 (72–77) Gender (male), n (%) 17 (31.5) Debut AHF, n (%) 28 (51.8) Hypertension, n (%) 45 (83.3) Dyslipidemia, n (%) 25 (46.3) Current smoker, n (%) 0 Diabetes mellitus, n (%) 30 (55.6) Ischemic heart disease, n (%) 20 (37.0) Valvular heart disease, n (%) 12 (22.2) NYHA class III/IV, n (%)a 19 (35.2) COPD, n (%) 25 (46.3) Stroke, n (%) 5 (9.3) PAD, n (%) 3 (5.6) Radiological pleural effusion, n (%) 26 (48.1) Peripheral edema, n (%) 44 (81.5) Previous use of diuretics, n (%) 38 (70.4) Previous use of beta-blockers, n (%) 13 (24.1) Previous use of ACEi/ARB, n (%) 27 (50.0) Previous use of statins, n (%) 18 (33.3)
73 265 316 393 219 46 218 185 145 110 115 56 37 284 333 351 121 239 133
101 (98–105) 140 (136–144) 79 (77–81)
0.067 0.006 0.885
Vital signs Heart rate, bpm Systolic blood pressure, mm Hg Diastolic blood pressure, mm Hg
100 (97–102) 142 (140–145) 79 (77–80)
0.449 0.007 0.171
144 (51.4) 78 (27.9)
0.292 0.651
ECG Atrial fibrillation, n (%) QRS N 120 ms, n (%)
257 (48.1) 153 (28.6)
0.118 0.880
12.5 (12.3–12.7) 12.6 (12.4–12.8) 1.26 (1.19–1.33) 1.38 (1.31–1.45) 8 (7.7–8.2) 8 (7.7–8.3) 139 (139–140) 139 (138–139) 0.23 (0.11–0.36) 0.27 (0.15–0.38) 138 (74–254) 154 (86–328)
0.485 0.022 0.894 0.209 0.705 0.036
Biomarkers Haemoglobin, g/dl Serum creatinine, mg/dl Uric acid, mg/dl Sodium, mEq/l Troponin I, ng/ml BNP, pg/mlb
12.8 1.27 7.7 141 0.19 107
12.5 (12.4–12.7) 1.32 (1.27–1.37) 8 (7.8–8.2) 139 (138–139) 0.26 (0.16–0.35) 152 (82–294)
0.276 0.563 0.289 0.005 0.688 0.011
Echocardiography LVEF, % LVEF b 40%, n (%) LAD, mm LVDD, mm Septum, mm Posterior wall, mm
53 66 44 55 11.6 11.4
(52–55) 50 (48–52) (21.4) 90 (32.1) (43–45) 45 (44–46) (54–56) 56 (55–58) (11.3–11.8) 11.2 (10.9–11.5) (11.2–11.6) 11.1 (10.9–11.4)
0.020 0.003 0.140 0.017 0.060 0.098
Echocardiography LVEF, % LVEF b 40%, n (%) LAD, mm LVDD, mm Septum, mm Posterior wall, mm
60 (56–63) 6 (11.1) 45 (42–47) 52 (50–55) 12 (11.2–12.7) 11.7 (11.0–12.4)
Medical treatment Beta-blockers, n (%) Diuretics, n (%) Spironolactone, n (%) ACEI, n (%) ARB, n (%) Statins, n (%) Oral anticoagulants, n (%) Nitrates, n (%) Digoxin, n (%)
137 303 44 130 93 99 131 60 89
(44.5) (98.4) (14.3) (42.2) (30.2) (32.1) (42.9) (19.5) (28.9)
0.320 0.847 0.341 0.649 0.360 0.782 0.753 0.464 0.707
Medical treatment Beta-blockers, n (%) Diuretics, n (%) Spironolactone, n (%) ACEI, n (%) ARB, n (%) Statins, n (%) Oral anticoagulants, n (%) Nitrates, n (%) Digoxin, n (%)
(73–76) (43.5) (59.1) (75.6) (42.9) (7.2) (43.5) (32.6) (26.9) (21.1) (25.6) (12.7) (6.2) (50.6) (59.4) (63.6) (25.0) (45.4) (26.9)
136 276 48 113 75 93 123 48 77
(71–74) (52.9) (57.9) (73.2) (40.0) (8.6) (40.7) (37.5) (26.5) (22.9) (21.8) (7.9) (7.5) (55.0) (69.3) (68.9) (20.4) (45.0) (24.3)
(48.6) (98.6) (17.1) (40.4) (26.8) (33.2) (44.2) (17.1) (27.5)
96 (89–104) 154 (145–163) 82 (77–87)
32 (59.3) 16 (29.6)
(12.4–13.2) (1.11–1.43) (7.0–8.3) (139–142) (0–0.42) (64–186)
21 (39.0) 54 (100) 6 (11.1) 17 (31.5) 16 (29.6) 13 (24.1) 22 (40.7) 12 (22.2) 21 (38.9)
(73–74) (49.6) (59.2) (73.6) (41.0) (8.6) (40.8) (34.7) (27.2) (20.6) (21.5) (10.5) (7.0) (53.2) (62.4) (65.7) (22.7) (44.8) (24.9)
51 150 44 56 11.3 11.2
(50–52) (28.1) (44–45) (55–57) (11.1–11.5) (11.1–11.4)
252 525 86 226 152 179 232 96 145
(47.2) (98.3) (16.1) (42.3) (28.5) (33.5) (43.9) (18.0) (27.1)
0.498 0.010 0.301 0.103 0.454 0.026 0.038 0.733 0.422 0.019 b 0.001 0.775 0.695 0.480 0.003 0.488 0.814 0.462 0.188
b 0.001 0.004 0.657 0.011 0.068 0.102
0.241 0.336 0.316 0.118 0.857 0.148 0.659 0.453 0.076
AHF: acute heart failure; NYHA: New York Heart Association; COPD: chronic pulmonary obstructive disease; PAD: peripheral arterial disease; ACEI: angiotensin converting enzyme inhibitor; ARB: angiotensin II receptor antagonist; BNP: brain natriuretic peptide; LVEF: left ventricular ejection fraction; LAD: left atrial diameter; LVDD: left ventricular diastolic diameter. Values are expressed as mean (95% CI). Categorical variables are presented as percentages. a Last NYHA functional class measured under clinically stable conditions. b Presented as median (interquartile range).
AHF: acute heart failure; NYHA: New York Heart Association; COPD: chronic pulmonary obstructive disease; PAD: peripheral arterial disease; ACEI: angiotensin converting enzyme inhibitor; ARB: angiotensin II receptor antagonist; BNP: brain natriuretic peptide; LVEF: left ventricular ejection fraction; LAD: left atrial diameter; LVDD: left ventricular diastolic diameter. Values are expressed as mean (95% CI). Categorical variables are presented as percentages. a Last NYHA functional class measured under clinically stable conditions. b Presented as median (interquartile range).
death did not differ among patients with hypoxemia, hypercapnia or acidosis (Fig. 2 a, b and c). In multivariable Cox regression analysis, pO2, pCO2 and pH did not show a significant association with all-cause mortality. Evaluated in a continuous metric, the hazard ratios (HR) for pO2 (per 10 mm Hg), pCO2 (per 10 mm Hg) and pH (per 0.1 scale) were: 0.99 (95% CI: 0.90–1.09), p = 0.861; 1.12 (95% CI: 0.91–1.39), p = 0.262 and; 1.01
(0.99–1.04), p = 0.309, respectively. The Harrell's C-statistics of the multivariate model that included pO2, pCO2 and pH as continuous showed high discriminative ability of 0.741; the Gronnesby and Borgan test of goodness-of-fit showed a good model calibration (p = 0.802).
492
G. Miñana et al. / European Journal of Internal Medicine 22 (2011) 489–494
Table 3 Baseline characteristics stratified by the presence of acidosis.
Demographic and medical history Age, years Gender (male), n (%) Debut AHF, n (%) Hypertension, n (%) Dyslipidemia, n (%) Current smoker, n (%) Diabetes mellitus, n (%) Ischemic heart disease, n (%) Valvular heart disease, n (%) NYHA class III/IV, n (%)a COPD, n (%) Stroke, n (%) PAD, n (%) Radiological pleural effusion, n (%) Peripheral edema, n (%) Previous use of diuretics, n (%) Previous use of beta-blockers, n (%) Previous use of ACEi/ARB, n (%) Previous use of statins, n (%) Vital signs Heart rate, bpm Systolic blood pressure, mm Hg Diastolic blood pressure, mm Hg ECG Atrial fibrillation, n (%) QRS N 120 ms, n (%)
pH b 7.35
pH ≥ 7.35
n = 50 (8.5%)
n = 538 (91.5%)
76 (74–79) 21 (42.0) 33 (66.0) 41 (82.0) 24 (48.0) 3 (6.0) 21 (42.0) 22 (44.0) 9 (18.0) 14 (28.0) 17 (34.0) 6 (12.0) 5 (10.0) 25 (50.0) 30 (60.0) 29 (58.0) 9 (18.0) 22 (44.0) 13 (26.0)
98 (89–106) 154 (144–164) 82 (77–87)
20 (40.0) 19 (38.0)
p-value
73 (72–74) 261 (48.5) 311 (57.8) 397 (73.8) 220 (40.9) 43 (8.0) 227 (42.2) 183 (34.1) 148 (27.6) 115 (21.4) 123 (22.9) 55 (10.2) 35 (6.5) 285 (53.0) 347 (64.5) 360 (66.9) 125 (23.2) 244 (45.3) 138 (25.6)
0.063 0.377 0.256 0.188 0.332 0.601 0.979 0.166 0.129 0.292 0.088 0.699 0.375 0.687 0.529 0.209 0.387 0.854 0.957
99 (97–102) 143 (140–145) 79 (77–80)
0.683 0.013 0.181
269 (50.0) 150 (27.9)
0.175 0.140
Biomarkers Haemoglobin, g/dl Serum creatinine, mg/dl Uric acid, mg/dl Sodium, mEq/l Troponin I, ng/ml BNP, pg/mlb
12.6 (12.1–13.1) 12.5 (12.4–12.7) 0.816 1.49 (1.29–1.68) 1.30 (1.25–1.35) 0.039 8 (7.3–8.6) 8 (7.8–8.2) 0.927 140 (139–141) 139 (138–139) 0.159 0.38 (0.10–0.66) 0.24 (0.15–0.33) 0.374 182 (96–373) 143 (81–281) 0.198
Echocardiography LVEF, % LVEF b 40%, n (%) LAD, mm LVDD, mm Septum, mm Posterior wall, mm
53 (50–57) 52 (50–53) 8 (16.0) 148 (27.5) 43 (41–45) 44 (44–45) 54 (52–56) 56 (55–56) 11.9 (11.3–12.5) 11.3 (11.1–11.6) 12 (11.4–12.6) 11.2 (11–11.4)
Medical treatment Beta-blockers, n (%) Diuretics, n (%) Spironolactone, n (%) ACEI, n (%) ARB, n (%) Statins, n (%) Oral anticoagulants, n (%) Nitrates, n (%) Digoxin, n (%)
21 (42.0) 50 (100) 2 (4.0) 14 (28.0) 14 (28.0) 14 (28.0) 16 (32.0) 13 (26.0) 13 (26.0)
252 (46.8) 529 (98.3) 90 (16.7) 229 (42.6) 154 (28.6) 178 (33.1) 238 (44.6) 95 (17.7) 153 (28.4)
0.461 0.064 0.237 0.348 0.142 0.006
0.510 0.357 0.006 0.041 0.925 0.458 0.103 0.162 0.712
AHF: acute heart failure; NYHA: New York Heart Association; COPD: chronic pulmonary obstructive disease; PAD: peripheral arterial disease; ACEI: angiotensin converting enzyme inhibitor; ARB: angiotensin II receptor antagonist; BNP: brain natriuretic peptide; LVEF: left ventricular ejection fraction; LAD: left atrial diameter; LVDD: left ventricular diastolic diameter. Values are expressed as mean (95% CI). Categorical variables are presented as percentages. a Last NYHA functional class measured under clinically stable conditions. b Presented as median (interquartile range).
Similarly, non-significant results were observed by dichotomizing these variables with established cut points. The adjusted HRs for hypoxemia (pO2 b 60 mm Hg), hypercapnia (pCO2 N 50 mm Hg) and acidosis (pH b 7.35) were: 1.07 (95% CI: 0.81–1.40), p = 0.637; 0.98 (95% CI: 0.62–1.57), p = 0.952; and 1.38 (0.87–2.19), p = 0.173, respectively (Fig. 3).
Fig. 2. Overall survival stratified according to pO2 (a), pCO2 (b) and pH (c) values.
No significant interactions were found between hypoxemia, hypercapnia and acidosis with any of the most representative subgroups such as being older than 65, female, those with preserved systolic function, ischemic heart disease and previous admission for AHF, by showing not only the lack of significance but a similar direction of the effect. Because the availability of ABG may be considered a potential surrogate of clinical severity, we tested the prognostic value of having performed an ABG at patients' arrival, but this time using the entire population of patients with ADHF (n = 804). The adjusted HR for mortality showed no significant differences between patients with or without ABG measurements [HR: 1.25 (95% CI: 0.92–1.70), p = 0.162], a result that diminished the likelihood of significant selection bias in our results.
G. Miñana et al. / European Journal of Internal Medicine 22 (2011) 489–494
Fig. 3. All-cause mortality and ABG variables. Multivariate model adjusted by gender, age, first admission for AHF, last NYHA class measured under stable conditions, hypercholesterolemia, heart rate, systolic blood pressure, serum creatinine and serum sodium.
4. Discussion The present study showed, in a large and non-selected population with ADHF and after a thorough multivariable adjustment, that ABG (pO2, pCO2 and pH) determined at patients' arrival to the emergency department are not associated with all-cause long-term mortality. We believe that the present results may bring to the notice of the emergency physicians to avoid the routine use of ABG results for prognostic assessment in patients during an ADHF episode.
493
47 patients with severe chronic congestive HF before and one year after heart transplantation. These authors showed that restrictive ventilatory defect induced by chronic HF was reversible, whereas the exception of the reduction in diffusion lung capacity for carbon monoxide (DLCO) was not improved, which probably reflected permanent changes in the lung vasculature. In addition, widely used treatments in patients with HF such as ACEI or ARB have been shown to have pleiotropic beneficial effects on lung function [12]. These changes have been mainly attributed to the reduction of levels of angiotensin II at the level of the pulmonary microvasculature and increase exposure to prostaglandins, especially prostacyclin and nitric oxide, as a result of increased local kinin concentrations. Even considering the limitation of these studies, which include small sample sizes and selected patients, they reinforce the idea that gas exchange disturbances in ADHF are mainly transitory, unlike the dyspnea that is consequence of a primary lung disease. Such reversibility is what made our results relevant. Thus, we speculate that most of patients with ADHF show reversible lung function disturbances that are limited to acute episodes. 4.3. Clinical implications Due to a lack of evidence linking ABG with long-term mortality, long-term prognosis assessment should not be based on data from admission arterial pO2, pCO2 and pH. As stated by guidelines, a thorough clinical, electrocardiographic, radiologic and biomarker evaluation should be standard medical practice in these patients.
4.1. ADHF prognosis 4.4. Strengths and limitations Recent ESC heart failure guidelines [3] encourage the use of acute HF classification. ADHF syndrome has become the most prevalent presentation from all acute HF syndromes, with 1-year mortality in most registries about 20–30% [11,12]; these figures frequently exceed the mortality rates observed in other more abrupt dyspneic presentations of acute HF, such as acute pulmonary edema or hypertensive acute HF [13,14]. Thus, it has been recommended to base the initial risk stratification of patients with ADHF on a rigorous clinical, radiologic, electrocardiographic and biomarker evaluation [3]. Based on our results, the addition of ABG results did not provide any additional prognostic utility to the ESC recommendations. Moreover, uniform results were observed in the most important subgroups of patients, such as women, preserved systolic function, previous admissions for AHF and ischemic etiology. In addition, the fact that patients with no ABG (because the physician in charge did not consider them necessary) have similar prognosis as patients with ABG assessment further supports our results.
At present, this is the first study performed in a large and representative population of patients with ADHF, and analyzed with an appropriate methodology, which suggests that the main variables of ABG did not have prognostic implications on long-term mortality. Some limitations, however, should be addressed. First, determination of BNP and other biomarker-related variables was assessed during patient's hospitalization and not on patient's arrival to the emergency department; this fact precludes their use in the multivariable model. Second, due to its observational nature, the possibility of selection bias and/or residual confounding from unknown or unmeasured covariates cannot be excluded. Third, this is a singlecenter registry which may limit the transportability of our conclusion to other populations. Fourth, having a low sample size in certain subgroups precludes obtaining robust estimates regarding a possible differential effect of ABG in these subgroups. 4.5. Conclusions
4.2. Biological interpretation of the present results In acute HF, lung function disturbances are related to reduction in vital and total lung capacity, lung diffusion capacity and lung compliance. Consequently, resistance to air flow is moderately increased, while total and functional residual volume remain normal [15]. Usually, patients with chronic left ventricular failure exhibited a restrictive ventilatory defect [16,17] characterized by a reduction in vital capacity resulting from the replacement of air in the lungs by blood or interstitial fluid. Previous studies have shown that lung function abnormalities can be reduced, or even resolved, with different treatments, including ACEI [18–20], ultrafiltration [21,22] and heart transplantation [23,24]. In a classic study, Light et al. [25] evaluated the changes in pulmonary function in 28 patients hospitalized for congestive HF. Initially, patients had both obstructive and restrictive ventilatory dysfunction that rapidly improved with treatment initiation. In another study, Niset et al. [23] evaluated the reversibility of the lung dysfunction in
In patients with ADHF, admission arterial pO2, pCO2 and pH were not associated with all-cause long-term mortality. Learning points • In patients admitted with acute decompensated heart failure, admission arterial pO2, pCO2 and pH are not associated with allcause long-term mortality. • Due to a lack of evidence linking ABG with long-term mortality, long-term prognosis assessment should not be based on data from admission arterial pO2, pCO2 and pH. As stated by guidelines, a thorough clinical, electrocardiographic, radiologic and biomarker evaluation should be standard medical practice in these patients. Conflict of interest None to declare.
494
G. Miñana et al. / European Journal of Internal Medicine 22 (2011) 489–494
References [1] Adams Jr KF, Fonarow GC, Emerman CL, LeJemtel TH, Costanzo MR, Abraham WT, et al, ADHERE Scientific Advisory Committee and Investigators. Characteristics and outcomes of patients hospitalized for heart failure in the United States: rationale, design, and preliminary observations from the first 100,000 cases in the Acute Decompensated Heart Failure National Registry (ADHERE). Am Heart J 2005;149: 209–16. [2] Fonarow GC, ADHERE Scientific Advisory Committee. The Acute Decompensated Heart Failure National Registry (ADHERE): opportunities to improve care of patients hospitalized with acute decompensated heart failure. Rev Cardiovasc Med 2003;4(Suppl 7):S21–30. [3] European Society of Cardiology; Heart Failure Association of the ESC (HFA); European Society of Intensive Care Medicine (ESICM), Dickstein K, Cohen-Solal A, Filippatos G, McMurray JJ, Ponikowski P, Poole-Wilson PA, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the diagnosis and treatment of acute and chronic heart failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur J Heart Fail 2008;10:933–89. [4] Rabe KF, Hurd S, Anzueto A, Barnes PJ, Buist SA, Calverley P, et al , Global Initiative for Chronic Obstructive Lung Disease. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med 2007;176:532–55. [5] Garland A, Dawson NV, Altmann I, Thomas CL, Phillips RS, Tsevat J, et al, SUPPORT Investigators. Outcomes up to 5 years after severe, acute respiratory failure. Chest 2004;126:1897–904. [6] Camargo Jr CA, Rachelefsky G, Schatz M. Managing asthma exacerbations in the emergency department: summary of the National Asthma Education and Prevention Program Expert Panel Report 3 guidelines for the management of asthma exacerbations. J Emerg Med 2009;37:S6–S17. [7] Lim WS, Baudouin SV, George RC, Hill AT, Jamieson C, Le Jeune I, et al, Pneumonia Guidelines Committee of the BTS Standards of Care Committee. BTS guidelines for the management of community acquired pneumonia in adults: update. 2009. Thorax 2009;64(Suppl 3):iii1–iii55. [8] Wicki J, Perrier A, Perneger TV, Bounameaux H, Junod AF. Predicting adverse outcome in patients with acute pulmonary embolism: a risk score. Thromb Haemost 2000;84:548–52. [9] Aujesky D, Roy PM, Le Manach CP, Verschuren F, Meyer G, Obrosky DS, et al. Validation of a model to predict adverse outcomes in patients with pulmonary embolism. Eur Heart J 2006;27:476–81. [10] May S, Hosmer DW. Hosmer and Lemeshow type goodness-of-fit statistics for the Cox proportional hazards model. In: Balakrishnana N, Rao CR, editors. Advances in survival analysis: handbook of statistics, Vol 23. Amsterdam: Elsevier, NorthHolland; 2004. p. 383–94.
[11] Cleland JG, Swedberg K, Follath F, Komajda M, Cohen-Solal A, Aguilar JC, et al, Study Group on Diagnosis of the Working Group on Heart Failure of the European Society of Cardiology. The EuroHeart Failure survey programme—a survey on the quality of care among patients with heart failure in Europe. Eur Heart J 2003;24: 442–63. [12] Siirilä-Waris K, Lassus J, Melin J, Peuhkurinen K, Nieminen MS, FINN-AKVA Study Group. Characteristics, outcomes, and predictors of 1-year mortality in patients hospitalized for acute heart failure. Eur Heart J 2006;27:3011–7. [13] Núñez J, Núñez E, Fonarow GC, Sanchis J, Bodí V, Bertomeu-González V, et al, Differential prognostic effect of systolic blood pressure on mortality according to left-ventricular function in patients with acute heart failure. Eur J Heart Fail 2010;12:38–44. [14] Harjola VP, Follath F, Nieminen MS, Brutsaert D, Dickstein K, Drexler H, et al. Characteristics, outcomes, and predictors of mortality at 3 months and 1 year in patients hospitalized for acute heart failure. Eur J Heart Fail 2010;12:239–48. [15] Johnson BD, Beck KC, Olson LJ, O'Malley KA, Allison TG, Squires RW, et al. Pulmonary function in patients with reduced left ventricular function: influence of smoking and cardiac surgery. Chest 2001;120:1869–76. [16] Naum CC, Sciurba FC, Rogers RM. Pulmonary function abnormalities in chronic severe cardiomyopathy preceding cardiac transplantation. Am Rev Respir Dis 1992;145:1334–8. [17] Chua TP, Coats AJS. The lungs in chronic heart failure. Eur Heart J 1995;16:882–7. [18] Guazzi M, Marenzi G, Alimento M, Contini M, Agostoni P. Improvement of alveolar capillary membrane diffusing capacity with enalapril in chronic heart failure and counteracting effect of aspirin. Circulation 1997;95:1930–6. [19] Guazzi M, Melzi G, Agostoni PG. Comparison of change in respiratory function and exercise oxygen uptake with losartan vs. enalapril in congestive heart failure secondary to ischemic or idiopathic cardiomyopathy. Am J Cardiol 1997;80:1572–6. [20] Guazzi M, Palermo P, Pontone G, Susini F, Agostoni P. Synergistic efficacy of enalapril and losartan on exercise performance and oxygen consumption at peak exercise in congestive heart failure. Am J Cardiol 1999;84:1038–43. [21] Agostoni PG, Marenzi GC, Pepi M, Doria E, Salvioni A, Perego G. Isolated ultrafiltration in moderate heart failure. J Am Coll Cardiol 1993;21:424–31. [22] Agostoni P, Marenzi G, Lauri G, Perego G, Schianni M, Sganzerla P, et al. Sustained improvement in functional capacity after removal of body fluid with isolated ultrafiltration in chronic cardiac insufficiency. Failure of furosemide provide same result. Am J Med 1994;96:191–6. [23] Niset G, Ninane V, Antoine M, Yernault JC. Respiratory dysfunction in congestive heart failure: correction after heart transplantation. Eur Respir J 1993;6: 1197–201. [24] Hosenpud JD, Stibolt TA, Atwal K, Shelley D. Abnormal pulmonary function specifically related to congestive heart failure: comparison of patients before and after cardiac transplantation. Am J Med 1990;88:493–6. [25] Light RW, George RB. Serial pulmonary function in patients with acute heart failure. Arch Intern Med 1983;143:429–33.