©1991 S. K argcr A G , Basel 0025-7931/91/0584-0141 $ 2.75/0
Respiration 1991;58:141-144
Erythropoietin in Chronic Obstructive Pulmonary Disease Relationship between Serum Erythropoietin, Blood Hemoglobin and Lung Function Effect of the Calcium Antagonist Isradipine on Serum Erythropoietin Niels Graudalb, Anders Mikael Gall0ea, OveJuul Nielsenc * Department of Pulmonary Medicine P, Chest Clinic, Bispebjerg Hospital, Copenhagen; b Department of Internal Medicine B, Bispebjerg Hospital, Copenhagen; c Department of Medicine P, Division of Nephrology, Rigshospitalet, Copenhagen, Denmark
Key Words. Anemia • Calcium channel blockers • Lung diseases, obstructive • Pulmonary circulation • Polycythemia Abstract. The relations between serum-erythropoietin (se-EPO), blood hemoglobin and lung function were investigated in patients with chronic obstructive pulmonary disease (COPD). In a randomized, double-blind, placebo-controllcd design, the hypothesis was tested that se-EPO might increase after treatment with a calcium antagonist. In 18 patients with COPD, median se-EPO (19.2 mU ml-1) was in the low normal range. The correla tions between se-EPO and blood hemoglobin (p = -0.63, p = 0.024) and between se-EPO and lung function in dices were negative (n.s.) and the correlation between blood hemoglobin and lung function indices (n.s.) were positive. The hypothesis is proposed that the normal hemoglobin found in most patients with COPD is a result of a balance between a trend towards a decreased red cell mass, as found in chronic diseases, and a trend to wards an increased red cell mass due to the erythropoietic effect of EPO. Se-EPO was not changed by the cal cium antagonist, isradipine.
In patients with severe chronic obstructive pulmo nary disease (COPD) polycythemia may develop as a result of an increase in erythropoietin (EPO) due to hypoxia. However, in 1963 Vanier et al. [1] found that hemoglobin and hematocrit were abnormally low in hypoxic emphysematous patients at sea level when compared to those of normal high altitude dwellers exposed to a similar degree of hypoxia. Only few em physematous patients suffered from polycythemia. Recently, scrum erythropoietin (se-EPO) has been analysed in selected polycythemic and nonpolycythemic patients with severe COPD and various degrees of hypoxia. The results are diverging and do not show an unequivocal trend towards a raised EPO [2-4], The first purpose of the present study was to characterize the level of se-EPO and to analyze the relation be tween se-EPO, blood hemoglobin and lung function in consecutive patients with severe COPD, who were not selected with respect to hemoglobin. A beneficial effect of calcium antagonists on the
pulmonary hemodynamics of patients with COPD has been indicated [5]. Thus, calcium antagonists may prove to be of importance in the treatment of these patients. The calcium antagonist verapamil has been found to enhance the EPO production in rats exposed to hypoxia [6]. The implication of this may be that es pecially anemic COPD patients may benefit from treatment with calcium antagonists and that calcium antagonists are relatively contraindicated in polycyt hemic COPD patients. The second purpose of the present study was to test the hypothesis, that se-EPO might increase after treatment with the calcium an tagonist, isradipine. The study was performed in a randomized, double-blind, placebo-controlled design.
Patients and Methods Patients below 70 years were included if forced expiratory vol ume in the first second (FEV,) was below 1.0 liter and FEV,/FVC was below 70% (FVC: forced vital capacity). Patients with asthma, serious diseases, ongoing treatment with calcium antagonists, or known allergy to calcium antagonists, were excluded. Eighteen
Downloaded by: King's College London 137.73.144.138 - 12/11/2017 1:56:05 PM
Introduction
Graudal/Galloe/Nielsen
142
Baseline
2h
3 months
Hemoglobin P
-0.63 0.024
-0.73 0.008
Hematocrit P
-0.63 0.024
-0.69 0.014
FEV, P
-0.19 n.s.
-0.26 n.s.
-0.37 n.s.
FVC P
-0.17 n.s.
-0.19 n.s.
-0.64 0.05
-
_
Table 2. Se-EPO in the isradipine and the placebo group at baseline (before ingestion of medicine), 2 h after ingestion of the first dose, and after 3 months of treatment with isradipine or pla cebo and change in se-EPO in relation to the baseline value 2 h af ter ingestion of the first dose and after 3 months Isradipine
Se-EPO, U/l Baseline (n = 18) 14.4 (10.6-21.8) 2 h (after first dose) 15.5 (9.1-22.3) (n = 18) 3 months (n = 12) 15.9 (13.9-22.0)
Placebo
Prob ability
22.0 (9.5-30.4)
0.27
18.8 (9.9-29.1) 12.3 (10.1-25.0)
0.29 0.60
Change in se-EPO, % 2 h/baseline 93.7 (74.8-132.8) 94.2 (68.3-118.9) 3 months/baseline 105.1 (71.3-157.6) 91.8 (42.4-112.6)
0.47 0.25
Median and 10-90 percentiles are shown in parentheses.
consecutive patients were included. All patients gave their in formed consent, and the study was approved by the Local Ethics Committee and the National Board of Health. The patients were assigned to isradipine or placebo by means of a single sequence of computer-generated random numbers. Coded drugs were prepared by Sandoz A/S. Isradipine is a calcium antagonist of the dihydropyridine group. Eighty percent of maxi mal efficacy on blood pressure is achieved by a dose of 2.5 mg. Study Procedure After allocation, the therapy was continued with placcbo/isradipine 2.5 mg twice daily for 12 weeks. After 4 and 8 weeks, the blood pressure was controlled. If systolic blood pressure was above 110 mm Hg, the dose was increased to 5 mg twice daily (all patients who terminated the 3-month study received 5 mg of isradipine).
Se-EPO. A blood sample for se-EPO analysis was obtained be fore randomization (‘baseline’ value), 2 h and 12 weeks after in gestion of isradipine or placebo (‘2-hour’ and ‘3-month’ values). Samples were obtained, stored and analyzed as described previ ously (7). Pulmonary Function Tests. Lung function indices (FEV, and FVC) were obtained immediately before the blood samples for EPO analysis. They were determined as the best of three consec utive graphic recordings from a Vitalograph. Hemoglobin and Hematocrit. These were obtained with the first EPO sample. Statistical Methods Data are reported as medians and ranges. For each patient the ratios 2-hour valuc/baseline value and 3-month value/baseline val ue were calculated. These ratios reflect the changes of se-EPO in the observation period. The ratios of the isradipine group were compared to the ratios of the placebo group by means of the Mann-Whitney test as were all unpaired comparisons. Multiplepaired comparisons were performed by means of the Friedman test. Correlation coefficients were calculated as Spearman’s p. The level of significance was 5%.
Results
Ten of the 18 patients in the acute study received placebo and 8 isradipine. These patients participated in the 2-hour study of the first day. None of the pa tients had side-effects (tachycardia, peripheral ede ma, flushing) after the first dose. Seven patients on placebo and 5 on isradipine participated in the 3month study. Three patients on placebo left the study, two due to exhaustion and one because of dizziness. Three patients on isradipine left the study, one be cause of a rash, one due to peripheral edema and one due to exhaustion. The baseline level was 19.2 mli ml-1 (range 8.7-30.4), which was generally in the range of 99 normal persons who had a se-EPO of 24.8 mU ml-1 (range 12^18). The population was characterized by the following parameters (median and ranges): age = 65.5 years (47-69), FEV, = 0.6 I (0.3-0.93), FVC = 1.26 1(0.8-2.77), FEV,/FVC = 42% (29-68), and blood hemoglobin = 9.1 mmol l-1 (7.7-10.2). No significant differences in age, FEV,, FVC, hemoglobin, hemato crit and baseline se-EPO values between the isradi pine group and the placebo group were found. None of the patients were on long-term oxygen treatment. The correlation coefficients between se-EPO and blood hemoglobin, hematocrit, FEV, and FVC were all negative (table 1). The correlation coefficients be tween FEV, and blood hemoglobin at baseline (p = 0.32) and after 2 h (p = 0.37), and between FVC and
Downloaded by: King's College London 137.73.144.138 - 12/11/2017 1:56:05 PM
Table 1. Correlation coefficients (Spearman’s p) between seEPO and hemoglobin, hematocrit, FEV, and FVC
Erythropoietin in COPD
blood hemoglobin at baseline (p = 0.31) and after 2 h (p = 0.28) were all positive, but not statistically signif icant. The EPO level at 2 h was 16.6 U/l (0.8-24.1) and at 3 months it was 15.2 U/I (10.2-24.2). A multiple com parison of the 3 EPO levels (baseline, 2 h, 3 months) showed no statistically significant difference (p = 0.58). Significant differences in changes in se-EPO be tween the isradipine group and the placebo group were not found, neither after 2 h (p = 0.47) nor after 3 months (p = 0.25; table 2).
143
Pa02 FEV I
Discussion
Fig. 1. Curve 1 shows decreasing lung function and Pa02 with time. Curve 2 shows that the basal level of EPO is normal for a long time, hut, in the period of intermittent hypoxia. EPO spikes can be registered (grating area). In the period of continuous hyp oxia, the basal level of EPO rises, but in phase III it decreases again due to a feedback mechanism caused by polycythemia. Spikes can still be registered in case of intermittent deteriorations. Curve 3 shows that hemoglobin is maintained in the normal area by EPO-spikes and, in the late phase of continuous hypoxia, it may rise secondary to a rise in EPO. Curve 4 shows the hypothetical decrease in hemoglobin with time in a progressive chronic disease. Curve 5 shows the net effect of EPO on hemoglobin, i.e. the dif ference between curve 3 and 4.
nal phase of the disease, the hypoxic drive may be so strong that the effect of EPO may result in polycythe mia. When polycythemia has developed, a feedback mechanism results in a relative decrease in EPO. The hypothesis is illustrated in figure 1. It may explain the divergent results of the different studies. Our patients and the patients of Miller et al. [2] and Wedzicha et al. [3], who had a normal hemoglobin and a normal se-EPO, could belong to the phase of intermittent hypoxia or to phase I of a continuous hypoxia. The patients of Guidet et al. [4] with a normal hemoglobin and a raised EPO could belong to phase II of a con tinuous hypoxia, whereas the polycythemic patients with a raised EPO of Wedzicha et al. [3] and Guidet et al. [4] could belong to phase III of a continuous hy poxia.
Downloaded by: King's College London 137.73.144.138 - 12/11/2017 1:56:05 PM
Miller et al. [2] found a mean EPO level of 19.6 mU ml-1 in patients with COPD and no polycythemia and a mean EPO level of 18.5 ml) ml"1 in normal subjects. Wedzicha et al. [3] found a mean EPO level of 19 mU ml"1 in patients with COPD and no polycythemia, and a raised mean EPO level of 30 mU ml"1 in patients with COPD and polycythemia. Guidet et al. [4] found that the mean se-EPO level was 60 mU ml"1 in those who had a normal hemoglobin, which was higher than in those who were polycythemic (mean se-EPO = 40 mU ml"1). Our finding of a normal se-EPO in patients with a severely reduced lung function and a normal hemo globin corresponds to the results of Miller et al. [2] and Wedzicha et al. [3]. The negative correlation coef ficient between se-EPO and hemoglobin should have been expected to be positive if EPO had been raised due to hypoxia. The negative correlations between the lung function parameters and se-EPO indicate that EPO increases with decreasing lung function within the population. The positive correlation coefficients between lung function (FEV, and FVC) and hemo globin indicate that hemoglobin decreases with de creasing lung function. We propose the hypothesis that, in parallel to other chronic diseases, the hemo globin decreases in patients with COPD as the chron ic disease progresses, that is as the FEV1( FVC and Pa02 decrease. In the early phase of respiratory fail ure, intermittent hypoxia during sleep, exhaustion and acute exacerbations result in frequent spikes of EPO which maintain a normal level of hemoglobin. Later, during continuous hypoxia, the basal level of EPO ris es, but still it only maintains an otherwise decreased hemoglobin on the normal level. Rarely, in the termi-
Graudal/Galloe/Nielsen
144
The unchanged se-EPO in the isradipine group af ter 2 h and after 3 months compared to the placebo group may be the net effect of a direct stimulatory ef fect of the calcium antagonist and an indirect inhib itory effect of an improved oxygen uptake. Alterna tively, isradipine has no effect on the EPO-production or the oxygen uptake. At any rate polycythemia does not seem to be a contraindication to treatment with a calcium antagonist. Acknowledgements The authors are grateful to Ernst and Vibeke Husmans Foun dation and Sandoz A/S for financial support.
3 Wedzicha JA, Cotes PM, Empey DW. Newland AC, Royston JP, Tam RC: Serum immunoreactive erythopoictin in hypoxic lung disease with and without polycythaemia. Clin Sci 1985; 69:413-422. 4 Guidet B, Offcnstadt G, Boffa G, Najman A, Baiilou C. Hatzfeld C. Amstutz P: Polycythemia in chronic obstructive pulmo nary disease. A study of serum and urine erythropoietin and medullary erythroid progenitors. Chest 1987;92:867-870. 5 Bratel T, Hedenstierna G. Nyquist O, Ripe E: The use of a vasodilator, felodipine, as an adjuvant to long-term oxygen treatment in COLD patients. Eur Respir J 1990:3:46-54. 6 McGonicle RJS. Brookins J, Pcgram PL. Fisher JW: Enhanced erythropoietin production by calcium entry blockers in rats ex posed to hypoxia. J Pharmacol Exp Thcr 1987;241:428-432. 7 Nielsen OJ: Determination of human erythropoietin by ra dioimmunoassay. Method and clinical data. Clin Chim Acta 1988:176:303-313.
R eferen ces
Received: January 17, 1991 Accepted after revision: May 2, 1991 Niels Graudal, MD Wiedeweltsgade 56 DK-2100 Copenhagen (Denmark)
Downloaded by: King's College London 137.73.144.138 - 12/11/2017 1:56:05 PM
1 Vanier T, Dulfano MJ, Wu C. Desforges JF: Emphysema, hy poxia and the polycythemic response. N Engl J Med 1963; 269:169-178. 2 Miller ME. Garcia JF, Cohen RA, Cronkite EP. Moccia G. Acevedo J: Diurnal levels of immunoreactive erythropoietin in normal subjects and subjects with chronic lung disease. Br J Haematol 1981;49:189-200.
Respiration 1991;58:141-144
Erythropoietin in Chronic Obstructive Pulmonary Disease Relationship between Serum Erythropoietin, Blood Hemoglobin and Lung Function Effect of the Calcium Antagonist Isradipine on Serum Erythropoietin Niels Graudalb, Anders Mikael Gall0ea, OveJuul Nielsenc * Department of Pulmonary Medicine P, Chest Clinic, Bispebjerg Hospital, Copenhagen; b Department of Internal Medicine B, Bispebjerg Hospital, Copenhagen; c Department of Medicine P, Division of Nephrology, Rigshospitalet, Copenhagen, Denmark
Key Words. Anemia • Calcium channel blockers • Lung diseases, obstructive • Pulmonary circulation • Polycythemia Abstract. The relations between serum-erythropoietin (se-EPO), blood hemoglobin and lung function were investigated in patients with chronic obstructive pulmonary disease (COPD). In a randomized, double-blind, placebo-controllcd design, the hypothesis was tested that se-EPO might increase after treatment with a calcium antagonist. In 18 patients with COPD, median se-EPO (19.2 mU ml-1) was in the low normal range. The correla tions between se-EPO and blood hemoglobin (p = -0.63, p = 0.024) and between se-EPO and lung function in dices were negative (n.s.) and the correlation between blood hemoglobin and lung function indices (n.s.) were positive. The hypothesis is proposed that the normal hemoglobin found in most patients with COPD is a result of a balance between a trend towards a decreased red cell mass, as found in chronic diseases, and a trend to wards an increased red cell mass due to the erythropoietic effect of EPO. Se-EPO was not changed by the cal cium antagonist, isradipine.
In patients with severe chronic obstructive pulmo nary disease (COPD) polycythemia may develop as a result of an increase in erythropoietin (EPO) due to hypoxia. However, in 1963 Vanier et al. [1] found that hemoglobin and hematocrit were abnormally low in hypoxic emphysematous patients at sea level when compared to those of normal high altitude dwellers exposed to a similar degree of hypoxia. Only few em physematous patients suffered from polycythemia. Recently, scrum erythropoietin (se-EPO) has been analysed in selected polycythemic and nonpolycythemic patients with severe COPD and various degrees of hypoxia. The results are diverging and do not show an unequivocal trend towards a raised EPO [2-4], The first purpose of the present study was to characterize the level of se-EPO and to analyze the relation be tween se-EPO, blood hemoglobin and lung function in consecutive patients with severe COPD, who were not selected with respect to hemoglobin. A beneficial effect of calcium antagonists on the
pulmonary hemodynamics of patients with COPD has been indicated [5]. Thus, calcium antagonists may prove to be of importance in the treatment of these patients. The calcium antagonist verapamil has been found to enhance the EPO production in rats exposed to hypoxia [6]. The implication of this may be that es pecially anemic COPD patients may benefit from treatment with calcium antagonists and that calcium antagonists are relatively contraindicated in polycyt hemic COPD patients. The second purpose of the present study was to test the hypothesis, that se-EPO might increase after treatment with the calcium an tagonist, isradipine. The study was performed in a randomized, double-blind, placebo-controlled design.
Patients and Methods Patients below 70 years were included if forced expiratory vol ume in the first second (FEV,) was below 1.0 liter and FEV,/FVC was below 70% (FVC: forced vital capacity). Patients with asthma, serious diseases, ongoing treatment with calcium antagonists, or known allergy to calcium antagonists, were excluded. Eighteen
Downloaded by: King's College London 137.73.144.138 - 12/11/2017 1:56:05 PM
Introduction
Graudal/Galloe/Nielsen
142
Baseline
2h
3 months
Hemoglobin P
-0.63 0.024
-0.73 0.008
Hematocrit P
-0.63 0.024
-0.69 0.014
FEV, P
-0.19 n.s.
-0.26 n.s.
-0.37 n.s.
FVC P
-0.17 n.s.
-0.19 n.s.
-0.64 0.05
-
_
Table 2. Se-EPO in the isradipine and the placebo group at baseline (before ingestion of medicine), 2 h after ingestion of the first dose, and after 3 months of treatment with isradipine or pla cebo and change in se-EPO in relation to the baseline value 2 h af ter ingestion of the first dose and after 3 months Isradipine
Se-EPO, U/l Baseline (n = 18) 14.4 (10.6-21.8) 2 h (after first dose) 15.5 (9.1-22.3) (n = 18) 3 months (n = 12) 15.9 (13.9-22.0)
Placebo
Prob ability
22.0 (9.5-30.4)
0.27
18.8 (9.9-29.1) 12.3 (10.1-25.0)
0.29 0.60
Change in se-EPO, % 2 h/baseline 93.7 (74.8-132.8) 94.2 (68.3-118.9) 3 months/baseline 105.1 (71.3-157.6) 91.8 (42.4-112.6)
0.47 0.25
Median and 10-90 percentiles are shown in parentheses.
consecutive patients were included. All patients gave their in formed consent, and the study was approved by the Local Ethics Committee and the National Board of Health. The patients were assigned to isradipine or placebo by means of a single sequence of computer-generated random numbers. Coded drugs were prepared by Sandoz A/S. Isradipine is a calcium antagonist of the dihydropyridine group. Eighty percent of maxi mal efficacy on blood pressure is achieved by a dose of 2.5 mg. Study Procedure After allocation, the therapy was continued with placcbo/isradipine 2.5 mg twice daily for 12 weeks. After 4 and 8 weeks, the blood pressure was controlled. If systolic blood pressure was above 110 mm Hg, the dose was increased to 5 mg twice daily (all patients who terminated the 3-month study received 5 mg of isradipine).
Se-EPO. A blood sample for se-EPO analysis was obtained be fore randomization (‘baseline’ value), 2 h and 12 weeks after in gestion of isradipine or placebo (‘2-hour’ and ‘3-month’ values). Samples were obtained, stored and analyzed as described previ ously (7). Pulmonary Function Tests. Lung function indices (FEV, and FVC) were obtained immediately before the blood samples for EPO analysis. They were determined as the best of three consec utive graphic recordings from a Vitalograph. Hemoglobin and Hematocrit. These were obtained with the first EPO sample. Statistical Methods Data are reported as medians and ranges. For each patient the ratios 2-hour valuc/baseline value and 3-month value/baseline val ue were calculated. These ratios reflect the changes of se-EPO in the observation period. The ratios of the isradipine group were compared to the ratios of the placebo group by means of the Mann-Whitney test as were all unpaired comparisons. Multiplepaired comparisons were performed by means of the Friedman test. Correlation coefficients were calculated as Spearman’s p. The level of significance was 5%.
Results
Ten of the 18 patients in the acute study received placebo and 8 isradipine. These patients participated in the 2-hour study of the first day. None of the pa tients had side-effects (tachycardia, peripheral ede ma, flushing) after the first dose. Seven patients on placebo and 5 on isradipine participated in the 3month study. Three patients on placebo left the study, two due to exhaustion and one because of dizziness. Three patients on isradipine left the study, one be cause of a rash, one due to peripheral edema and one due to exhaustion. The baseline level was 19.2 mli ml-1 (range 8.7-30.4), which was generally in the range of 99 normal persons who had a se-EPO of 24.8 mU ml-1 (range 12^18). The population was characterized by the following parameters (median and ranges): age = 65.5 years (47-69), FEV, = 0.6 I (0.3-0.93), FVC = 1.26 1(0.8-2.77), FEV,/FVC = 42% (29-68), and blood hemoglobin = 9.1 mmol l-1 (7.7-10.2). No significant differences in age, FEV,, FVC, hemoglobin, hemato crit and baseline se-EPO values between the isradi pine group and the placebo group were found. None of the patients were on long-term oxygen treatment. The correlation coefficients between se-EPO and blood hemoglobin, hematocrit, FEV, and FVC were all negative (table 1). The correlation coefficients be tween FEV, and blood hemoglobin at baseline (p = 0.32) and after 2 h (p = 0.37), and between FVC and
Downloaded by: King's College London 137.73.144.138 - 12/11/2017 1:56:05 PM
Table 1. Correlation coefficients (Spearman’s p) between seEPO and hemoglobin, hematocrit, FEV, and FVC
Erythropoietin in COPD
blood hemoglobin at baseline (p = 0.31) and after 2 h (p = 0.28) were all positive, but not statistically signif icant. The EPO level at 2 h was 16.6 U/l (0.8-24.1) and at 3 months it was 15.2 U/I (10.2-24.2). A multiple com parison of the 3 EPO levels (baseline, 2 h, 3 months) showed no statistically significant difference (p = 0.58). Significant differences in changes in se-EPO be tween the isradipine group and the placebo group were not found, neither after 2 h (p = 0.47) nor after 3 months (p = 0.25; table 2).
143
Pa02 FEV I
Discussion
Fig. 1. Curve 1 shows decreasing lung function and Pa02 with time. Curve 2 shows that the basal level of EPO is normal for a long time, hut, in the period of intermittent hypoxia. EPO spikes can be registered (grating area). In the period of continuous hyp oxia, the basal level of EPO rises, but in phase III it decreases again due to a feedback mechanism caused by polycythemia. Spikes can still be registered in case of intermittent deteriorations. Curve 3 shows that hemoglobin is maintained in the normal area by EPO-spikes and, in the late phase of continuous hypoxia, it may rise secondary to a rise in EPO. Curve 4 shows the hypothetical decrease in hemoglobin with time in a progressive chronic disease. Curve 5 shows the net effect of EPO on hemoglobin, i.e. the dif ference between curve 3 and 4.
nal phase of the disease, the hypoxic drive may be so strong that the effect of EPO may result in polycythe mia. When polycythemia has developed, a feedback mechanism results in a relative decrease in EPO. The hypothesis is illustrated in figure 1. It may explain the divergent results of the different studies. Our patients and the patients of Miller et al. [2] and Wedzicha et al. [3], who had a normal hemoglobin and a normal se-EPO, could belong to the phase of intermittent hypoxia or to phase I of a continuous hypoxia. The patients of Guidet et al. [4] with a normal hemoglobin and a raised EPO could belong to phase II of a con tinuous hypoxia, whereas the polycythemic patients with a raised EPO of Wedzicha et al. [3] and Guidet et al. [4] could belong to phase III of a continuous hy poxia.
Downloaded by: King's College London 137.73.144.138 - 12/11/2017 1:56:05 PM
Miller et al. [2] found a mean EPO level of 19.6 mU ml-1 in patients with COPD and no polycythemia and a mean EPO level of 18.5 ml) ml"1 in normal subjects. Wedzicha et al. [3] found a mean EPO level of 19 mU ml"1 in patients with COPD and no polycythemia, and a raised mean EPO level of 30 mU ml"1 in patients with COPD and polycythemia. Guidet et al. [4] found that the mean se-EPO level was 60 mU ml"1 in those who had a normal hemoglobin, which was higher than in those who were polycythemic (mean se-EPO = 40 mU ml"1). Our finding of a normal se-EPO in patients with a severely reduced lung function and a normal hemo globin corresponds to the results of Miller et al. [2] and Wedzicha et al. [3]. The negative correlation coef ficient between se-EPO and hemoglobin should have been expected to be positive if EPO had been raised due to hypoxia. The negative correlations between the lung function parameters and se-EPO indicate that EPO increases with decreasing lung function within the population. The positive correlation coefficients between lung function (FEV, and FVC) and hemo globin indicate that hemoglobin decreases with de creasing lung function. We propose the hypothesis that, in parallel to other chronic diseases, the hemo globin decreases in patients with COPD as the chron ic disease progresses, that is as the FEV1( FVC and Pa02 decrease. In the early phase of respiratory fail ure, intermittent hypoxia during sleep, exhaustion and acute exacerbations result in frequent spikes of EPO which maintain a normal level of hemoglobin. Later, during continuous hypoxia, the basal level of EPO ris es, but still it only maintains an otherwise decreased hemoglobin on the normal level. Rarely, in the termi-
Graudal/Galloe/Nielsen
144
The unchanged se-EPO in the isradipine group af ter 2 h and after 3 months compared to the placebo group may be the net effect of a direct stimulatory ef fect of the calcium antagonist and an indirect inhib itory effect of an improved oxygen uptake. Alterna tively, isradipine has no effect on the EPO-production or the oxygen uptake. At any rate polycythemia does not seem to be a contraindication to treatment with a calcium antagonist. Acknowledgements The authors are grateful to Ernst and Vibeke Husmans Foun dation and Sandoz A/S for financial support.
3 Wedzicha JA, Cotes PM, Empey DW. Newland AC, Royston JP, Tam RC: Serum immunoreactive erythopoictin in hypoxic lung disease with and without polycythaemia. Clin Sci 1985; 69:413-422. 4 Guidet B, Offcnstadt G, Boffa G, Najman A, Baiilou C. Hatzfeld C. Amstutz P: Polycythemia in chronic obstructive pulmo nary disease. A study of serum and urine erythropoietin and medullary erythroid progenitors. Chest 1987;92:867-870. 5 Bratel T, Hedenstierna G. Nyquist O, Ripe E: The use of a vasodilator, felodipine, as an adjuvant to long-term oxygen treatment in COLD patients. Eur Respir J 1990:3:46-54. 6 McGonicle RJS. Brookins J, Pcgram PL. Fisher JW: Enhanced erythropoietin production by calcium entry blockers in rats ex posed to hypoxia. J Pharmacol Exp Thcr 1987;241:428-432. 7 Nielsen OJ: Determination of human erythropoietin by ra dioimmunoassay. Method and clinical data. Clin Chim Acta 1988:176:303-313.
R eferen ces
Received: January 17, 1991 Accepted after revision: May 2, 1991 Niels Graudal, MD Wiedeweltsgade 56 DK-2100 Copenhagen (Denmark)
Downloaded by: King's College London 137.73.144.138 - 12/11/2017 1:56:05 PM
1 Vanier T, Dulfano MJ, Wu C. Desforges JF: Emphysema, hy poxia and the polycythemic response. N Engl J Med 1963; 269:169-178. 2 Miller ME. Garcia JF, Cohen RA, Cronkite EP. Moccia G. Acevedo J: Diurnal levels of immunoreactive erythropoietin in normal subjects and subjects with chronic lung disease. Br J Haematol 1981;49:189-200.
