Plasma Alpha-Human Atrial Natriuretic Peptide Concentration in Patients with Acute Lung Injury1-3
CHIEKO MITAKA, YUKIO HIRATA, TAKASHI NAGURA, NOBUE SAKANISHI, YUKIO TSUNODA, and KEISUKE AMAHA
Introduction
Alpha-human atrial natriuretic peptide (a-hANP) is a novel cardiac hormone involved in homeostasis of body fluid and blood pressure (1). a-hANP primarily acts on blood vessels, kidneys, and adrenal glands to induce vasodilation, diuresis/natriuresis, and inhibition of aldosterone secretion, respectively. The biologic actions of a-hANP are initiated by binding to biologic (B) receptor to stimulate the particulate guanylate cyclase in target tissues, whereas clearance (C) receptor has been reported to function for clearance and metabolism of a-hANP (2). Therefore, cGMP has been shown to function as a second messenger for the biologic effects of a-hANP. In addition to the blood vessels, kidney (3), and adrenal glands (4), lungs also contain abundant B-receptor (5), although its physiologic significance in the lung remains obscure. We have shown in a preliminary study that plasma a-hANP concentrations were elevated in patients with acute respiratory failure associated with sepsis (6). However, the pathophysiologic significance of such increased levels of circulating a-hANP in acute lung injury remains unknown. Therefore, we have attempted to measure plasma concentrations of a-hANP in patients with acute lung injury to know whether circulating a-hANP is related to the severity of lung injury, diuresis/natriuresis, and fluid balance. Methods Subjects Fifteen patients (mean age, 59 ± 15 yr) with severe lung injury were studied. The severity of lung injury was classified by using the scoring system for lung injury of Murray and coworkers (7). This system is based on four components: the chest roentgenogram, oxygenation (Pao2/FI02), positive end-expiratory pressure (PEEP), and lung compliance, each scoring from zero to 4 points. The final value is obtained by dividing the aggregate sum by the number of components used. A score
SUMMARY To elucidate the pathophysiologic role of a-human atrial natriuretic peptide (a-hANP) In acute lung Injury, plasma a-hANP concentrations were measured In 15 patients with severe lung Injury, and the relationships of plasma a-hANP levels to the severity of lung Injury, diuresis/natriuresis, and fluid balance were examined. The mean concentretlons of plasma a-hANP (188.0 ::I: 94.6 pg/ml) In patients with severe lung Injury at the entry Into the atudy were significantly (p < 0.001) higher than those In normal subJects (31.7 ::I: 12.0 pglml). Plasma a-hANP levels decreased In parallel with the Improvement of lung Injury In nine of 15 patients, whereas they changed little, If any, In the patients who did not recover. Plasma a-hANP concentrations correlated positively with urine volume, urinary sodium excretion, and excreted fraction of filtered sodium, but they correlated negatively with fluid balance atthe onset olthe disease as well as during the clinical course. It Is suggested that elevetlon of circulatory a-hANP may reflect an adaptstlve mechanism to remove exceulve fluid retention and reduce pulmonary hypertension for acute lung Injury. AM REV RESPIR DIS 1"2; 148:43-48
more than 2.5 represents severe injury, a score between 0.1 and 2.5 represents mild-tomoderate injury, and a zero score means no lung injury. The lung injury score of each patient was more than 2.5 at entry into the present study. The patients' clinical data are summarized in table 1. Most of the patients developed severelung injury as a result of sepsis. Chest roentgenograms showed diffuse pulmonary infiltrations in all patients. No patient had any evidence of preexisting lung, heart, or renal diseases, and serum creatinine was less than 2.0 mg/dl during the study. All patients were treated with mechanical ventilation with positive end-expiratory pressure (PEEP), usually less than 12 em H 20 , and FIo2and PEEP were adjusted to ensure a Pao2 of at least 80 mm Hg. Pulmonary capillary wedge pressure (Ppcw) was less than 18 mm Hg, and fluids were administered intravenously to maintain an adequate cardiac output while keeping a normal Ppcw. Diuretics were not used during the study. Informed consent was obtained from each patient's closest relatives.
Measurements of a-hANP and cGMP Blood samples for a-hANP, cOMP, and blood gas determinations were obtained from an indwelling radial arterial catheter. The initial blood samples were obtained in most patients at the early time point (1 to 2 days) of the severelung injury, although three patients (patients 8, 11, and 15) had been treated before they were referred to the intensive care unit, which moved blood sampling to a later point (3 to 4 days) of the disease. After the initial
blood sampling, frequent blood samplings (every 1 to 3 days) were performed during the hospital course, and the initial and the later samples of patients who clinically improved or worsened were compared. Blood samples for a-hANP and cGMP were collected into chilled tubes containing EDTA-2K and immediately centrifuged at 4 0 C. Plasma was stored at -80 0 C until assayed. Plasma a-hANP concentrations were measured by a specific and sensitive radioimmunoassay (RIA) as reported (8). Plasma cGMP concentrations were determined by a commercial cGMP radioimmunoassay kit (Yamasa, Choshi, Japan). Serum and urinary sodium and creatinine were measured by an autoanalyzer (736-60; Hitachi, Tokyo, Japan). Urine volume (UV) and urinary sodium excretion (UNaV) were measured, and the excreted fraction of filtered
(Received in original form March 9, 1991 and in revised form March 5, 1992) 1 From the Section of Intensive Care and the Second Department of Internal Medicine, Tokyo Medical and Dental University, Tokyo, Japan. 2 Supported in part by Research Grants 02304055, 03454512,03268102, and 03454218 from the Ministry of Education, Science and Culture of Japan, and by funds from the Uehara Memorial Foundation (Tokyo). 3 Correspondence and requests for reprints should be addressed to Chieko Mitaka, M.D., The Section of Intensive Care, Tokyo Medical and Dental University, 1-5-45,Yushima, Bunkyo-ku, Tokyo, 113, Japan.
43
44
MITAKA, HIRATA, NAGURA, SAKANISHI, TSUNODA, AND AMAHA
TABLE 1 CLINICAL CHARACTERISTICS OF 15 PATIENTS WITH LUNG INJURY AT THE ENTRY INTO THE STUDY Patient No.
Age
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Mean ± SO
74 76 46 59 83 59 27 52 69 47 50 46
(yT) Sex
77 68 54
M M M M M M F M M M M F M M M
Consolidation by Chest Radiograph
Diagnosis
4quadrants(4) * 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4} 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4 ± 0 (4 ± 0)
Sepsis after esophagogastrectomy, OIC Sepsis after pancreatoduodenectomy Malignant lymphoma, peritonitis, sepsis Pneumonia, sepsis Septic shock of unknown origin Sepsis after pancreatoduodenectomy Sepsis after trauma, multiple blood transfusion Malignant lymphoma, sepsis, OIC Pneumonia after bypass grafts Septic shock after esophagogastrectomy Peritonitis after esophagogastrectomy Systemic lupus erythematosus, sepsis Aspiration pneumonia Aspiration pneumonia Malignant lymphoma, pneumonia, OIC
PaoiFlo.
PEEP (em H2O)
Compliance (m/fcm H2O)
Final Score
u-hANP (pgfm~
152(3) 10(2) 35(3) 3 100 114(3) 12(3) 3.25 30(3) 360 200(2) 3 258 6(1) 138(3) 2.66 148.3 110(3} 3.5 228.5 202(2) 10(2) 2.66 118.3 220(2) 10(2) 23(3) 2.75 186.2 10(2) 107(3) 24(3) 3 115.6 6(1) 170(3) 2.66 121.5 215(2) 10(2) 2.75 25(3) 279.5 98(4) 12(3) 21(3) 3.5 158.1 10(2) 166(3) 26(3) 3 66.9 10(2) 202(2) 27(3) 2.75 244 141(3) 10(2) 38(3) 3 353 6(1) 203(2) 18(4) 2.75 82.5 162 ± 42 9 ± 2 26 ± 6 2.9 ± 0.2 188.0 ± 94.6 (2.6 ± 0.6) (1.9 ± 0.6) (3.1 ± 0.3)
Outcome S
0
0 0 S S S
0 0 S
0 0 S S
0
Definitionof abbreviations: Pao,lF1o, = arterialoxygen tensionto inspired oxygen concentration ratio; PEEP s positive end-explratory pressure; a-hANP = a-human atrial natriuretic peptide; ole - disseminated intravascular coagulation; 0 • died; S - survived. Two patients (Nos. 3 and 12), although recovered from respiratory failure, died of hepatic failure. * Shownin parentheses arethe lung injury scoresusingthe scoringsystem of Murrayand coworkers (7). Thefinal score is obtained by dividingthe aggregate sum by the number of components thet were used.
sodium (FENa) was calculated according to the standard formula, Daily fluid intake, including blood products, total parenteral nutrition, intravenously administered fluids, and antibiotics, were recorded, Fluid loss other than UV, such as nasogastric output and insensible loss, was not included. Fluid balance was calculated as fluid intake minus UV. These measurements wererecorded at the entry into the study and continued during the entire course, Statistical Analysis Values are expressed as mean ± SD. Linear regression analysis was used to determine the correlations between plasma a-hANP concentrations and other variables. Statistical analysis was performed by paired and unpaired t tests; p < 0.05 was considered a statistically significant change.
Results
The mean plasma concentrations of a-hANP (188.0 ± 94.6pg/ml) and cGMP (16.9 ± 11.8 pmol/ml) in 15patients with acute lung injury at the entry into the study weresignificantly (p < 0.001) higher than those in normal subjects (31.7 ± 12.0 pg/ml, n = 124, and 3.2 ± 1.07 pmol/ml, n = 20, respectively). Plasma a-hANP concentrations did not correlate with Ppcw (r = 0.336) or right atrial pressure (r = 0.207) throughout the clinical course. After intensive care for acute respiratory failure, nine of the 15 patients showed amelioration of the respiratory failure, but the other six patients did not recover. Plasma a-hANP concentrations at the entry into the study and after treat-
ments for all 15 patients are illustrated in figure 1. Plasma a-hANP concentrations in patients after successful treatment significantly (p < 0.001)decreased, from 203.8 ± 93.7 to 79.3 ± 52.1 pg/ml in relation to the decrease in the lung injury score, whereas those in patients who did not recover did not significantly change (164.3 ± 99.5versus 165.2 ± 66.8 pg/ml), The relationship between circulating a-hANP levels and severity of the adult respiratory distress syndrome in each patient during the hospital course was examined. In those patients who recovered, plasma a-hANP concentrations correlated well with the lung injury score (figure 2). In contrast, none of those who did not recover showed correlation between plasma a-hANP levels and the lung injury score. There existed a significant correlation (r = 0.477, P < 0.01) between plasma a-hANP and cGMP concentrations frequently obtained from all the patients (data not shown). The relationships be-
tween plasma a-hANP levels and UV, UNaV, FENa, and fluid balance in all lf patients at the early time point are shown in figure 3. Plasma a-hANP concentrations correlated positively with UV (A: r = 0.564,p
CHIEKO MITAKA, YUKIO HIRATA, TAKASHI NAGURA, NOBUE SAKANISHI, YUKIO TSUNODA, and KEISUKE AMAHA
Introduction
Alpha-human atrial natriuretic peptide (a-hANP) is a novel cardiac hormone involved in homeostasis of body fluid and blood pressure (1). a-hANP primarily acts on blood vessels, kidneys, and adrenal glands to induce vasodilation, diuresis/natriuresis, and inhibition of aldosterone secretion, respectively. The biologic actions of a-hANP are initiated by binding to biologic (B) receptor to stimulate the particulate guanylate cyclase in target tissues, whereas clearance (C) receptor has been reported to function for clearance and metabolism of a-hANP (2). Therefore, cGMP has been shown to function as a second messenger for the biologic effects of a-hANP. In addition to the blood vessels, kidney (3), and adrenal glands (4), lungs also contain abundant B-receptor (5), although its physiologic significance in the lung remains obscure. We have shown in a preliminary study that plasma a-hANP concentrations were elevated in patients with acute respiratory failure associated with sepsis (6). However, the pathophysiologic significance of such increased levels of circulating a-hANP in acute lung injury remains unknown. Therefore, we have attempted to measure plasma concentrations of a-hANP in patients with acute lung injury to know whether circulating a-hANP is related to the severity of lung injury, diuresis/natriuresis, and fluid balance. Methods Subjects Fifteen patients (mean age, 59 ± 15 yr) with severe lung injury were studied. The severity of lung injury was classified by using the scoring system for lung injury of Murray and coworkers (7). This system is based on four components: the chest roentgenogram, oxygenation (Pao2/FI02), positive end-expiratory pressure (PEEP), and lung compliance, each scoring from zero to 4 points. The final value is obtained by dividing the aggregate sum by the number of components used. A score
SUMMARY To elucidate the pathophysiologic role of a-human atrial natriuretic peptide (a-hANP) In acute lung Injury, plasma a-hANP concentrations were measured In 15 patients with severe lung Injury, and the relationships of plasma a-hANP levels to the severity of lung Injury, diuresis/natriuresis, and fluid balance were examined. The mean concentretlons of plasma a-hANP (188.0 ::I: 94.6 pg/ml) In patients with severe lung Injury at the entry Into the atudy were significantly (p < 0.001) higher than those In normal subJects (31.7 ::I: 12.0 pglml). Plasma a-hANP levels decreased In parallel with the Improvement of lung Injury In nine of 15 patients, whereas they changed little, If any, In the patients who did not recover. Plasma a-hANP concentrations correlated positively with urine volume, urinary sodium excretion, and excreted fraction of filtered sodium, but they correlated negatively with fluid balance atthe onset olthe disease as well as during the clinical course. It Is suggested that elevetlon of circulatory a-hANP may reflect an adaptstlve mechanism to remove exceulve fluid retention and reduce pulmonary hypertension for acute lung Injury. AM REV RESPIR DIS 1"2; 148:43-48
more than 2.5 represents severe injury, a score between 0.1 and 2.5 represents mild-tomoderate injury, and a zero score means no lung injury. The lung injury score of each patient was more than 2.5 at entry into the present study. The patients' clinical data are summarized in table 1. Most of the patients developed severelung injury as a result of sepsis. Chest roentgenograms showed diffuse pulmonary infiltrations in all patients. No patient had any evidence of preexisting lung, heart, or renal diseases, and serum creatinine was less than 2.0 mg/dl during the study. All patients were treated with mechanical ventilation with positive end-expiratory pressure (PEEP), usually less than 12 em H 20 , and FIo2and PEEP were adjusted to ensure a Pao2 of at least 80 mm Hg. Pulmonary capillary wedge pressure (Ppcw) was less than 18 mm Hg, and fluids were administered intravenously to maintain an adequate cardiac output while keeping a normal Ppcw. Diuretics were not used during the study. Informed consent was obtained from each patient's closest relatives.
Measurements of a-hANP and cGMP Blood samples for a-hANP, cOMP, and blood gas determinations were obtained from an indwelling radial arterial catheter. The initial blood samples were obtained in most patients at the early time point (1 to 2 days) of the severelung injury, although three patients (patients 8, 11, and 15) had been treated before they were referred to the intensive care unit, which moved blood sampling to a later point (3 to 4 days) of the disease. After the initial
blood sampling, frequent blood samplings (every 1 to 3 days) were performed during the hospital course, and the initial and the later samples of patients who clinically improved or worsened were compared. Blood samples for a-hANP and cGMP were collected into chilled tubes containing EDTA-2K and immediately centrifuged at 4 0 C. Plasma was stored at -80 0 C until assayed. Plasma a-hANP concentrations were measured by a specific and sensitive radioimmunoassay (RIA) as reported (8). Plasma cGMP concentrations were determined by a commercial cGMP radioimmunoassay kit (Yamasa, Choshi, Japan). Serum and urinary sodium and creatinine were measured by an autoanalyzer (736-60; Hitachi, Tokyo, Japan). Urine volume (UV) and urinary sodium excretion (UNaV) were measured, and the excreted fraction of filtered
(Received in original form March 9, 1991 and in revised form March 5, 1992) 1 From the Section of Intensive Care and the Second Department of Internal Medicine, Tokyo Medical and Dental University, Tokyo, Japan. 2 Supported in part by Research Grants 02304055, 03454512,03268102, and 03454218 from the Ministry of Education, Science and Culture of Japan, and by funds from the Uehara Memorial Foundation (Tokyo). 3 Correspondence and requests for reprints should be addressed to Chieko Mitaka, M.D., The Section of Intensive Care, Tokyo Medical and Dental University, 1-5-45,Yushima, Bunkyo-ku, Tokyo, 113, Japan.
43
44
MITAKA, HIRATA, NAGURA, SAKANISHI, TSUNODA, AND AMAHA
TABLE 1 CLINICAL CHARACTERISTICS OF 15 PATIENTS WITH LUNG INJURY AT THE ENTRY INTO THE STUDY Patient No.
Age
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Mean ± SO
74 76 46 59 83 59 27 52 69 47 50 46
(yT) Sex
77 68 54
M M M M M M F M M M M F M M M
Consolidation by Chest Radiograph
Diagnosis
4quadrants(4) * 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4} 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4quadrants(4) 4 ± 0 (4 ± 0)
Sepsis after esophagogastrectomy, OIC Sepsis after pancreatoduodenectomy Malignant lymphoma, peritonitis, sepsis Pneumonia, sepsis Septic shock of unknown origin Sepsis after pancreatoduodenectomy Sepsis after trauma, multiple blood transfusion Malignant lymphoma, sepsis, OIC Pneumonia after bypass grafts Septic shock after esophagogastrectomy Peritonitis after esophagogastrectomy Systemic lupus erythematosus, sepsis Aspiration pneumonia Aspiration pneumonia Malignant lymphoma, pneumonia, OIC
PaoiFlo.
PEEP (em H2O)
Compliance (m/fcm H2O)
Final Score
u-hANP (pgfm~
152(3) 10(2) 35(3) 3 100 114(3) 12(3) 3.25 30(3) 360 200(2) 3 258 6(1) 138(3) 2.66 148.3 110(3} 3.5 228.5 202(2) 10(2) 2.66 118.3 220(2) 10(2) 23(3) 2.75 186.2 10(2) 107(3) 24(3) 3 115.6 6(1) 170(3) 2.66 121.5 215(2) 10(2) 2.75 25(3) 279.5 98(4) 12(3) 21(3) 3.5 158.1 10(2) 166(3) 26(3) 3 66.9 10(2) 202(2) 27(3) 2.75 244 141(3) 10(2) 38(3) 3 353 6(1) 203(2) 18(4) 2.75 82.5 162 ± 42 9 ± 2 26 ± 6 2.9 ± 0.2 188.0 ± 94.6 (2.6 ± 0.6) (1.9 ± 0.6) (3.1 ± 0.3)
Outcome S
0
0 0 S S S
0 0 S
0 0 S S
0
Definitionof abbreviations: Pao,lF1o, = arterialoxygen tensionto inspired oxygen concentration ratio; PEEP s positive end-explratory pressure; a-hANP = a-human atrial natriuretic peptide; ole - disseminated intravascular coagulation; 0 • died; S - survived. Two patients (Nos. 3 and 12), although recovered from respiratory failure, died of hepatic failure. * Shownin parentheses arethe lung injury scoresusingthe scoringsystem of Murrayand coworkers (7). Thefinal score is obtained by dividingthe aggregate sum by the number of components thet were used.
sodium (FENa) was calculated according to the standard formula, Daily fluid intake, including blood products, total parenteral nutrition, intravenously administered fluids, and antibiotics, were recorded, Fluid loss other than UV, such as nasogastric output and insensible loss, was not included. Fluid balance was calculated as fluid intake minus UV. These measurements wererecorded at the entry into the study and continued during the entire course, Statistical Analysis Values are expressed as mean ± SD. Linear regression analysis was used to determine the correlations between plasma a-hANP concentrations and other variables. Statistical analysis was performed by paired and unpaired t tests; p < 0.05 was considered a statistically significant change.
Results
The mean plasma concentrations of a-hANP (188.0 ± 94.6pg/ml) and cGMP (16.9 ± 11.8 pmol/ml) in 15patients with acute lung injury at the entry into the study weresignificantly (p < 0.001) higher than those in normal subjects (31.7 ± 12.0 pg/ml, n = 124, and 3.2 ± 1.07 pmol/ml, n = 20, respectively). Plasma a-hANP concentrations did not correlate with Ppcw (r = 0.336) or right atrial pressure (r = 0.207) throughout the clinical course. After intensive care for acute respiratory failure, nine of the 15 patients showed amelioration of the respiratory failure, but the other six patients did not recover. Plasma a-hANP concentrations at the entry into the study and after treat-
ments for all 15 patients are illustrated in figure 1. Plasma a-hANP concentrations in patients after successful treatment significantly (p < 0.001)decreased, from 203.8 ± 93.7 to 79.3 ± 52.1 pg/ml in relation to the decrease in the lung injury score, whereas those in patients who did not recover did not significantly change (164.3 ± 99.5versus 165.2 ± 66.8 pg/ml), The relationship between circulating a-hANP levels and severity of the adult respiratory distress syndrome in each patient during the hospital course was examined. In those patients who recovered, plasma a-hANP concentrations correlated well with the lung injury score (figure 2). In contrast, none of those who did not recover showed correlation between plasma a-hANP levels and the lung injury score. There existed a significant correlation (r = 0.477, P < 0.01) between plasma a-hANP and cGMP concentrations frequently obtained from all the patients (data not shown). The relationships be-
tween plasma a-hANP levels and UV, UNaV, FENa, and fluid balance in all lf patients at the early time point are shown in figure 3. Plasma a-hANP concentrations correlated positively with UV (A: r = 0.564,p
