Scand J Haematol(l979) 22,289-295
Erythropoiesis in Sickle Cell Anaemia during Acute Infection and Crisis THERESA B. W D Y , JEANNE M.LUSHER, SUSAN HENDRICKS & BEVERLY K A Y TROSKO Department of Human Development, College of Human Medicine, Michigan State University, East Lansing, Michigan and Department of Pediatrics, Wayne State Medical School, Detroit, Michigan, USA
Erythropoiesis was evaluated in 37 patients with sickle cell anaemia, 26 of them children under 12 years of age. Mean haemoglobin, haematocrit, reticulocyte, and erythropoietin levels were similar for 11 who were asymptomatic, 11 with infections, and 12 in vaso-occlusive crisis. Mean haemoglobin, haematocrit, and reticulocyte values were significantly lower and the mean erythropoietin level significantly higher for three patients in aplastic crisis. Reticulocyte counts reflected erythropoietic activity during the asymptomatic state but were variable during infection and crisis. No erythropoietic inhibitory activity was found in any of the four clinical states. It has been suggested that erythropoietin production decreases during infection. Patients in this study responded appropriately t o stress, showing no decrease in erythropdetic activity during acute infection or crisis. Key words: aplastic crisis - erythropoiesis - erythropoietin - infection - painful crisis sickle cell anaemia - thrombotic crisis - vaseocclusive crisis
-
Accepted for publication January 11, 1979 Correspondence to: Theresa B. Haddy, M.D., Building 31, Room 4A-05, Blood Diseases Branch, Division of Blood Diseases and Resources, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20014, USA
Our study aims were to find out if a relative failure of erythropoiesis occurs during acute infection, as it does in chronic infection, or during sickle cell crisis. The anaemia of chronic infection is due to impaired erythropoietin production (Ward et
a1 1971, Lukens 1973, Zucker & Lysik 1974), with erythropoietin levels failing to rise in response to the degree of anaemia that is present. Erythropoietic activity and reticulocyte production are characteristically elevated above normal in sickle cell anae-
Supported in part by funds from the Bureau of Personal Health Services, Michigan Department of Public Health, Lansing, Michigan, USA. 003&553W79/0402894 $02.50/0 @ 1979 Munksgaard, Copenhagen
T. B. HADDY, J. M. LUSHER, S. HENDRICKS & B. K. TROSKO
290
TABLE 1 Types of injection encountered in I I patients with sickle cell anaemia Disease Pneumonia Cholecystitis Osteomyelitis Gastro-enteritis Total
I
Number of patients 8 1 1 1
11
mia (Hammond 1968, Alexanian 1973), and the inability to maintain these levels in the presence of acute infection would indicate a decrease in erythropoietic hormone production. MATERIALS AND METHODS Patients 37 patients ranging in age from 22 months t o 22 years, all of whom had homozygous sickle cell anaemia, were evaluated. 26 were children under 12 years of age; 7 of those remaining were girls and 4 were boys. Pre-pubertal children and females can be considered t o be free of androgenic steroid effect upon erythropoiesis (Jepson et a1 1973). 11 patients were asymptomatic but had chronic haemolysis; 11 had acute infections, listed by type in Table 1; and 12 had acute vaso-occlusive (thrombotic or painful) crisis with moderate to markedly severe symptoms. All patients in the latter two groups were hospitalized and were carefully selected to fit into either one category or the other. Those with infection had n o symptoms or signs of crisis, and those in crisis revealed no evidence of infection as evaluated by physical examination, cultures, and x-rays. 3 patients had acute aplastic crisis. 2 had low Hb, haematocrit, and reticulocyte values, and the third had low Hb and haematocrit values but an elevated reticulocyte count, signifying that the aplastic crisis was relenting (Schwartz 1974). Methods Our study compared Hb, haematocrit, reticulocyte, and erythropoietin levels of sickle cell anaemia patients in the asymptomatic (non-crisis or 'resting') state with levels found during acute
infections and in vaso-occlusive and aplastic crisis. H b and haematocrit values were determined with an electronic (Coulter Model S) counter. Reticulocyte enumerations were done by counting lo00 red blood cells after staining with brilliant cresyl blue. Blood samples for routine blood counts and plasma erythropoietin assays were all drawn on the same day. Informed consent, according to the Declaration of Helsinki, was obtained from parents of the children and from the adults included in the study after the purpose of the research was explained. The exhypoxic mouse bioassay for erythropoietin was carried out using Swiss-Webster mice, according to a modification of the method of Erslev (Erslev 1978). The mice were exposed to low atmospheric pressure for 20 h daily for 3 weeks. 2 ml of plasma were used for each assay. Radioactive iron was administered as 0.25 pCi of 59Fecitrate in 0.25 ml of isotonic saline. Results from mice with microhaematocrit readings of less than 55 % at the time of sacrifice were discarded. A mouse red blood cell S9Fe uptake dose-response curve, using doses ranging from 0.01 to 2.0 u of commercial erythropoietin ob-
2
'
of1012
'
' 0.'5
1.0
' ' ' I
2.I0 '
DOSE UNITS OF ERYTHROPOIETIN (CONNAUGHT STEP 1)
Figure 1. Log dose-response curve for commercial erythropoietin.
ERYTHROPOIESIS I N SICKLE CELL ANAEMIA
291
TABLE 2 Mean' Hb, haematocrit, reticulocyte, and mouse red blood cell "Fe uptake values for sickle cell anaemia patients who were asymptomatic, had infection, or were in vaso-occlusive or aplastic crisis Clinical state
Asymptomatic Infection Vaso-occlusive crisis Aplastic crisis
* **
Number of patients
Hb (g/dl)
Haematocrit (vol %)
11 11 12 3
7.9 f 0.3 8.6 f 0.3 9.2 f 0.4 3.2 f 0.6
24f 1 25+1 27f2 10+2
Reticulocytes
Mouse reduptake blood cell 59Fe
(%I
(%I
16.8 f 2.4 15.7 f 1.4 14.6f 1.6" 4.1 f 3.0
7 . 6 f 1.2 11.2f 1.9 10.2f 1.5 26.9 f 10.7
M e a n f SEM. Based on values from 11 patients.
tained from anaemic sheep (Connaught Laboratories, Toronto), is illustrated in Figure 1. 59Fe uptake for the control mice was 1.8 f 0.6 %. The exhypoxic mouse bioassay for erythropoietic inhibitor was carried out in the same way except that 0.2 u of commercial erythropoietin was mixed with the first dose of test plasma. A control group of mice was given an identical dose of commercial erythropoietin along with the first dose of normal plasma. The presence of an inhibitor would de demonstrated in
the test mice by significantly lower 59Fered blood cell incorporation. RESULTS
Mean Hb, haematocrit, reticulocyte, and erythropoietin values for the asymptomatic, infection, and vaso-occlusive crisis groups did not differ significantly from each other. These results appear in Table 2. Erythropoietic activity in the asymptomatic
0 I-
22 25-
#
under 12 yrs
0
12 yrs and older #
K 0
# #
Figure 2. Erythropoietic activity compared t o H b values of 11 asymptomatic patients (correlation coefficient .67), 11 patients with infection, and 12 patients with vaso-occlusive crisis and sickle cell anaemia.
T. B. HADDY, J. M. LUSHER, S. HENDRICKS & B. K. TROSKO
292
TABLE 3 Age, sex, Hb, haematocrit, reticulocyte, and mouse red blood cell 59Fe uptake values for three sickle cell anaemia patients who were in aplastic crisis Haematocrit (vol
Hb
Patient number
Sex
Age (years)
(ddl)
Reticub cyte count
Mouse red blood cell s9Fe uptake
(%)
(%I
Comment
1
M
9
4.1
11
2.0
5.8
Vaso-occlusive crisis and aplastic crisis
2
F
5
2.0
6
0.4
39.8
Vaso-occlusive crisis and aplastic crisis
3
F
16
3.4
12
10.0
35.3
Aplastic crisis, relenting
group was indirectly proportional to Hb and haematocrit levels (Figures 2 and 3), with least squares linear regression curves differing significantly from zero. Erythropoietic activity in the infection and vaso-occlusive
z 0 F 30.-
d
2
crisis groups was also indirectly proportional as shown in Figures 2 and 3, with curves that were similar to those of the asymptomatic group. The slopes did not differ significantly from zero.
under 12 yrs 0 12
yrs and older
25-
0
1 U
2ot
0
HEMATOCRIT
N O L %)
Figure 3. Erythropoietic activity compared to haematocrit values of 11 asymptomatic patients (correlation coefficient .68), 11 patients with infection, and 12 patients with vaso-occlusive crisis and sickle cell anaemia.
293
ERYTHROPOIESIS IN SICKLE CELL ANAEMIA
Erythropoietic activity appeared to be directly proportional, as expected, to reticulocyte counts in the asymptomatic patients (Figure 4). The pattern was scattered for those in the infection group (Figure 5 ) and indirectly proportional for those in vasoocclusive crisis (Figure 6). None of the relationships was statistically significant. Plasma samples from the 3 patients with aplastic crisis had elevated mouse red blood cell 59Fe uptakes. Two were extremely high, while the third did not differ significantly from the controls. These results appear in Tables 2 and 3. 8 additional plasma samples, 2 each from asymptomatic, infection, vaso-occlusive crisis, and aplastic crisis patients, were evaluated for an erythropoietic inhibitor. No inhibitor was identified.
.under 12 yrs 0 1 2 yrs and oldei
n
infection
*
*
to
reticulocyte counts of 11 patients with infection and sickle cell anaemia.
-*
DISCUSSION
sk 30
Previous investigators have attributed the anaemia of infection to low erythropoietin under 12 yrs a levels (Ward et a1 1971, Lukens 1973, Zuco 12 yrs and older ker & Lysik 1974). Our patients demonstrated appropriate erythropoietic activity 8 15 . z 1 20 during the stress of acute infection, and no I inhibitor was identified. No inferences can be drawn from this work concerning the multifactorial anaemia of chronic infection. Our study shows, however, that sickle cell anaemia patients do not have suppression of erythropoietin production in response to acute infection. Higher reticulocyte counts reflected higher 2 SD erythropoietin levels for asymptomatic per_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _--_ - - - - - - - - - w v) sons with sickle cell anaemia. The pattern 1 I for patients who were undergoing or had undergone an acute alteration in clinical status was less predictable. Reticulocyte counts in infection did not demonstrate a I
T. B. HADDY, J. M. LUSHER, S. HENDRICKS & B. K. TROSKO
294
consistent relationship to plasma erythropoietic activity, but reticulocyte counts appeared to be inversely related to erythropoietin levels in vaso-occlusive crisis. Individuals with high reticulocyte counts had low erythropoietin levels, and the relationship seemed to be almost linear. A possible explanation may be that the erythropoietic activity peak preceded the reticulocytosis by a few days, as it is known to do in hypoxia (Abbrecht & Littel 1972). Erythropoietic activity was highest for 2 patients who were in aplastic crisis. High levels are not unexpected in view of the severe degree of anaemia and are consistent with previously obtained results (Hammond et a1 1968). The third patient with aplastic crisis also had an elevated erythropoietin level, but not to the extent that would be expected. Failure of erythropoietin production can reflect renal damage. Since the pa-
under 12 yrs
z
2
0 12 yrs and older
2 30a
tient had no abnormal clinical or laboratory findings except for hyposthenuria and was later observed to have an appropriate reticulocyte response, this result was probably due to variability of the bioassay. Possible etiologies of relative failure of the bone marrow to increase erythropoiesis in response to anaemia in aplastic crisis include decreased production of erythropoietin, the presence of an erythropoietic inhibitor, and impaired bone marrow response to erythropoietin. Our study appears to rule out the first two causes and suggest temporary impairment of bone marrow response. It is possible, however, that the entire episode may be preceded by a fall in erythropoietin production with resultant reduction in erythropoiesis. This could only be ruled out by measurements taken before an acute episode became clinically evident. Although we were unable to identify the presence of an inhibitor by the method described, it may have been possible to do so using a different assay method. In addition, an inhibitor may have been present prior to the time the crisis was diagnosed.
2 250
0
z
201 U
ACKNOWLEDGEMENTS
We wish to thank R. Gerald Rice, M.D., and Thomas R. Kirk, M.D., for giving advice and encouragement; Burnell H. Selleck, Ph.D., and Raymond Johnston, D.V.M., Ph.D. (deceased), for supplying equipment; and Thomas E. Obremski, Ph.D., and Judy Stephenson, M.S., for assisting with the statistical analysis.
REFERENCES
R E T IC U LOCY TES ('% 1 Figure 6. Erythropoietic activity compared to reticulocyte counts of 11 patients with vasoocclusive crisis and sickle cell anaemia.
Abbrecht P H & Littel J K (1972) Plasma erythropoietin in men and mice during acclimatization to different altitudes. J A p p l Physiol 32, 54-58. Alexanian R (1973) Erythropoietin excretion in bone marrow failure and hemolytic anemia. J Lab CIin Med 82, 43845.
ERYTHROPOIESIS IN SICKLE CELL ANAEMIA
295
Erslev A J (1977) Erythropoietin assay. In W J rats with adjuvant-induced chronic inflammaWilliams, E Beutler, A J Erslev & R W Rundles tion. Blood 41, 37-44. (eds) Hematology, pp 1616-18. McGraw-Hill Schwartz E (1974) Aplastic and hypoplastic aneBook Co, New York. mias. In D G Nathan & F A Oski (eds) Hematology of infancy and childhood, p 175. W B Hammond D, Shore N & Movassaghi N (1968) Production, utilization and excretion of erythroSaunders Co, Philadelphia. poietin: I. Chronic anemias. 11. Aplastic crisis. Ward H P, Kurnick J E & Pisaczyk M J (1971) 111. Erythropoietic effects of normal plasma. Serum level of erythropoietin in anemias asAnn N Y Acad Sci 149, 516-27. sociated with chronic infection, malignancy, Jepson J H, Gardner F H, Gorshein D & Hait and primary hematopoietic disease. J Clin I n W M (1973) Current concepts of the action of vest 50, 332-35. androgenic steroids on erythropoiesis. J Pediatr Zucker S & Lysik R (1974) Bone marrow erythro83, 703-08. poiesis in anemia of inflammation. J Lab Clin Lukens J N (1973) Control of erythropoiesis in Med 84, 62&31.
Erythropoiesis in Sickle Cell Anaemia during Acute Infection and Crisis THERESA B. W D Y , JEANNE M.LUSHER, SUSAN HENDRICKS & BEVERLY K A Y TROSKO Department of Human Development, College of Human Medicine, Michigan State University, East Lansing, Michigan and Department of Pediatrics, Wayne State Medical School, Detroit, Michigan, USA
Erythropoiesis was evaluated in 37 patients with sickle cell anaemia, 26 of them children under 12 years of age. Mean haemoglobin, haematocrit, reticulocyte, and erythropoietin levels were similar for 11 who were asymptomatic, 11 with infections, and 12 in vaso-occlusive crisis. Mean haemoglobin, haematocrit, and reticulocyte values were significantly lower and the mean erythropoietin level significantly higher for three patients in aplastic crisis. Reticulocyte counts reflected erythropoietic activity during the asymptomatic state but were variable during infection and crisis. No erythropoietic inhibitory activity was found in any of the four clinical states. It has been suggested that erythropoietin production decreases during infection. Patients in this study responded appropriately t o stress, showing no decrease in erythropdetic activity during acute infection or crisis. Key words: aplastic crisis - erythropoiesis - erythropoietin - infection - painful crisis sickle cell anaemia - thrombotic crisis - vaseocclusive crisis
-
Accepted for publication January 11, 1979 Correspondence to: Theresa B. Haddy, M.D., Building 31, Room 4A-05, Blood Diseases Branch, Division of Blood Diseases and Resources, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20014, USA
Our study aims were to find out if a relative failure of erythropoiesis occurs during acute infection, as it does in chronic infection, or during sickle cell crisis. The anaemia of chronic infection is due to impaired erythropoietin production (Ward et
a1 1971, Lukens 1973, Zucker & Lysik 1974), with erythropoietin levels failing to rise in response to the degree of anaemia that is present. Erythropoietic activity and reticulocyte production are characteristically elevated above normal in sickle cell anae-
Supported in part by funds from the Bureau of Personal Health Services, Michigan Department of Public Health, Lansing, Michigan, USA. 003&553W79/0402894 $02.50/0 @ 1979 Munksgaard, Copenhagen
T. B. HADDY, J. M. LUSHER, S. HENDRICKS & B. K. TROSKO
290
TABLE 1 Types of injection encountered in I I patients with sickle cell anaemia Disease Pneumonia Cholecystitis Osteomyelitis Gastro-enteritis Total
I
Number of patients 8 1 1 1
11
mia (Hammond 1968, Alexanian 1973), and the inability to maintain these levels in the presence of acute infection would indicate a decrease in erythropoietic hormone production. MATERIALS AND METHODS Patients 37 patients ranging in age from 22 months t o 22 years, all of whom had homozygous sickle cell anaemia, were evaluated. 26 were children under 12 years of age; 7 of those remaining were girls and 4 were boys. Pre-pubertal children and females can be considered t o be free of androgenic steroid effect upon erythropoiesis (Jepson et a1 1973). 11 patients were asymptomatic but had chronic haemolysis; 11 had acute infections, listed by type in Table 1; and 12 had acute vaso-occlusive (thrombotic or painful) crisis with moderate to markedly severe symptoms. All patients in the latter two groups were hospitalized and were carefully selected to fit into either one category or the other. Those with infection had n o symptoms or signs of crisis, and those in crisis revealed no evidence of infection as evaluated by physical examination, cultures, and x-rays. 3 patients had acute aplastic crisis. 2 had low Hb, haematocrit, and reticulocyte values, and the third had low Hb and haematocrit values but an elevated reticulocyte count, signifying that the aplastic crisis was relenting (Schwartz 1974). Methods Our study compared Hb, haematocrit, reticulocyte, and erythropoietin levels of sickle cell anaemia patients in the asymptomatic (non-crisis or 'resting') state with levels found during acute
infections and in vaso-occlusive and aplastic crisis. H b and haematocrit values were determined with an electronic (Coulter Model S) counter. Reticulocyte enumerations were done by counting lo00 red blood cells after staining with brilliant cresyl blue. Blood samples for routine blood counts and plasma erythropoietin assays were all drawn on the same day. Informed consent, according to the Declaration of Helsinki, was obtained from parents of the children and from the adults included in the study after the purpose of the research was explained. The exhypoxic mouse bioassay for erythropoietin was carried out using Swiss-Webster mice, according to a modification of the method of Erslev (Erslev 1978). The mice were exposed to low atmospheric pressure for 20 h daily for 3 weeks. 2 ml of plasma were used for each assay. Radioactive iron was administered as 0.25 pCi of 59Fecitrate in 0.25 ml of isotonic saline. Results from mice with microhaematocrit readings of less than 55 % at the time of sacrifice were discarded. A mouse red blood cell S9Fe uptake dose-response curve, using doses ranging from 0.01 to 2.0 u of commercial erythropoietin ob-
2
'
of1012
'
' 0.'5
1.0
' ' ' I
2.I0 '
DOSE UNITS OF ERYTHROPOIETIN (CONNAUGHT STEP 1)
Figure 1. Log dose-response curve for commercial erythropoietin.
ERYTHROPOIESIS I N SICKLE CELL ANAEMIA
291
TABLE 2 Mean' Hb, haematocrit, reticulocyte, and mouse red blood cell "Fe uptake values for sickle cell anaemia patients who were asymptomatic, had infection, or were in vaso-occlusive or aplastic crisis Clinical state
Asymptomatic Infection Vaso-occlusive crisis Aplastic crisis
* **
Number of patients
Hb (g/dl)
Haematocrit (vol %)
11 11 12 3
7.9 f 0.3 8.6 f 0.3 9.2 f 0.4 3.2 f 0.6
24f 1 25+1 27f2 10+2
Reticulocytes
Mouse reduptake blood cell 59Fe
(%I
(%I
16.8 f 2.4 15.7 f 1.4 14.6f 1.6" 4.1 f 3.0
7 . 6 f 1.2 11.2f 1.9 10.2f 1.5 26.9 f 10.7
M e a n f SEM. Based on values from 11 patients.
tained from anaemic sheep (Connaught Laboratories, Toronto), is illustrated in Figure 1. 59Fe uptake for the control mice was 1.8 f 0.6 %. The exhypoxic mouse bioassay for erythropoietic inhibitor was carried out in the same way except that 0.2 u of commercial erythropoietin was mixed with the first dose of test plasma. A control group of mice was given an identical dose of commercial erythropoietin along with the first dose of normal plasma. The presence of an inhibitor would de demonstrated in
the test mice by significantly lower 59Fered blood cell incorporation. RESULTS
Mean Hb, haematocrit, reticulocyte, and erythropoietin values for the asymptomatic, infection, and vaso-occlusive crisis groups did not differ significantly from each other. These results appear in Table 2. Erythropoietic activity in the asymptomatic
0 I-
22 25-
#
under 12 yrs
0
12 yrs and older #
K 0
# #
Figure 2. Erythropoietic activity compared t o H b values of 11 asymptomatic patients (correlation coefficient .67), 11 patients with infection, and 12 patients with vaso-occlusive crisis and sickle cell anaemia.
T. B. HADDY, J. M. LUSHER, S. HENDRICKS & B. K. TROSKO
292
TABLE 3 Age, sex, Hb, haematocrit, reticulocyte, and mouse red blood cell 59Fe uptake values for three sickle cell anaemia patients who were in aplastic crisis Haematocrit (vol
Hb
Patient number
Sex
Age (years)
(ddl)
Reticub cyte count
Mouse red blood cell s9Fe uptake
(%)
(%I
Comment
1
M
9
4.1
11
2.0
5.8
Vaso-occlusive crisis and aplastic crisis
2
F
5
2.0
6
0.4
39.8
Vaso-occlusive crisis and aplastic crisis
3
F
16
3.4
12
10.0
35.3
Aplastic crisis, relenting
group was indirectly proportional to Hb and haematocrit levels (Figures 2 and 3), with least squares linear regression curves differing significantly from zero. Erythropoietic activity in the infection and vaso-occlusive
z 0 F 30.-
d
2
crisis groups was also indirectly proportional as shown in Figures 2 and 3, with curves that were similar to those of the asymptomatic group. The slopes did not differ significantly from zero.
under 12 yrs 0 12
yrs and older
25-
0
1 U
2ot
0
HEMATOCRIT
N O L %)
Figure 3. Erythropoietic activity compared to haematocrit values of 11 asymptomatic patients (correlation coefficient .68), 11 patients with infection, and 12 patients with vaso-occlusive crisis and sickle cell anaemia.
293
ERYTHROPOIESIS IN SICKLE CELL ANAEMIA
Erythropoietic activity appeared to be directly proportional, as expected, to reticulocyte counts in the asymptomatic patients (Figure 4). The pattern was scattered for those in the infection group (Figure 5 ) and indirectly proportional for those in vasoocclusive crisis (Figure 6). None of the relationships was statistically significant. Plasma samples from the 3 patients with aplastic crisis had elevated mouse red blood cell 59Fe uptakes. Two were extremely high, while the third did not differ significantly from the controls. These results appear in Tables 2 and 3. 8 additional plasma samples, 2 each from asymptomatic, infection, vaso-occlusive crisis, and aplastic crisis patients, were evaluated for an erythropoietic inhibitor. No inhibitor was identified.
.under 12 yrs 0 1 2 yrs and oldei
n
infection
*
*
to
reticulocyte counts of 11 patients with infection and sickle cell anaemia.
-*
DISCUSSION
sk 30
Previous investigators have attributed the anaemia of infection to low erythropoietin under 12 yrs a levels (Ward et a1 1971, Lukens 1973, Zuco 12 yrs and older ker & Lysik 1974). Our patients demonstrated appropriate erythropoietic activity 8 15 . z 1 20 during the stress of acute infection, and no I inhibitor was identified. No inferences can be drawn from this work concerning the multifactorial anaemia of chronic infection. Our study shows, however, that sickle cell anaemia patients do not have suppression of erythropoietin production in response to acute infection. Higher reticulocyte counts reflected higher 2 SD erythropoietin levels for asymptomatic per_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _--_ - - - - - - - - - w v) sons with sickle cell anaemia. The pattern 1 I for patients who were undergoing or had undergone an acute alteration in clinical status was less predictable. Reticulocyte counts in infection did not demonstrate a I
T. B. HADDY, J. M. LUSHER, S. HENDRICKS & B. K. TROSKO
294
consistent relationship to plasma erythropoietic activity, but reticulocyte counts appeared to be inversely related to erythropoietin levels in vaso-occlusive crisis. Individuals with high reticulocyte counts had low erythropoietin levels, and the relationship seemed to be almost linear. A possible explanation may be that the erythropoietic activity peak preceded the reticulocytosis by a few days, as it is known to do in hypoxia (Abbrecht & Littel 1972). Erythropoietic activity was highest for 2 patients who were in aplastic crisis. High levels are not unexpected in view of the severe degree of anaemia and are consistent with previously obtained results (Hammond et a1 1968). The third patient with aplastic crisis also had an elevated erythropoietin level, but not to the extent that would be expected. Failure of erythropoietin production can reflect renal damage. Since the pa-
under 12 yrs
z
2
0 12 yrs and older
2 30a
tient had no abnormal clinical or laboratory findings except for hyposthenuria and was later observed to have an appropriate reticulocyte response, this result was probably due to variability of the bioassay. Possible etiologies of relative failure of the bone marrow to increase erythropoiesis in response to anaemia in aplastic crisis include decreased production of erythropoietin, the presence of an erythropoietic inhibitor, and impaired bone marrow response to erythropoietin. Our study appears to rule out the first two causes and suggest temporary impairment of bone marrow response. It is possible, however, that the entire episode may be preceded by a fall in erythropoietin production with resultant reduction in erythropoiesis. This could only be ruled out by measurements taken before an acute episode became clinically evident. Although we were unable to identify the presence of an inhibitor by the method described, it may have been possible to do so using a different assay method. In addition, an inhibitor may have been present prior to the time the crisis was diagnosed.
2 250
0
z
201 U
ACKNOWLEDGEMENTS
We wish to thank R. Gerald Rice, M.D., and Thomas R. Kirk, M.D., for giving advice and encouragement; Burnell H. Selleck, Ph.D., and Raymond Johnston, D.V.M., Ph.D. (deceased), for supplying equipment; and Thomas E. Obremski, Ph.D., and Judy Stephenson, M.S., for assisting with the statistical analysis.
REFERENCES
R E T IC U LOCY TES ('% 1 Figure 6. Erythropoietic activity compared to reticulocyte counts of 11 patients with vasoocclusive crisis and sickle cell anaemia.
Abbrecht P H & Littel J K (1972) Plasma erythropoietin in men and mice during acclimatization to different altitudes. J A p p l Physiol 32, 54-58. Alexanian R (1973) Erythropoietin excretion in bone marrow failure and hemolytic anemia. J Lab CIin Med 82, 43845.
ERYTHROPOIESIS IN SICKLE CELL ANAEMIA
295
Erslev A J (1977) Erythropoietin assay. In W J rats with adjuvant-induced chronic inflammaWilliams, E Beutler, A J Erslev & R W Rundles tion. Blood 41, 37-44. (eds) Hematology, pp 1616-18. McGraw-Hill Schwartz E (1974) Aplastic and hypoplastic aneBook Co, New York. mias. In D G Nathan & F A Oski (eds) Hematology of infancy and childhood, p 175. W B Hammond D, Shore N & Movassaghi N (1968) Production, utilization and excretion of erythroSaunders Co, Philadelphia. poietin: I. Chronic anemias. 11. Aplastic crisis. Ward H P, Kurnick J E & Pisaczyk M J (1971) 111. Erythropoietic effects of normal plasma. Serum level of erythropoietin in anemias asAnn N Y Acad Sci 149, 516-27. sociated with chronic infection, malignancy, Jepson J H, Gardner F H, Gorshein D & Hait and primary hematopoietic disease. J Clin I n W M (1973) Current concepts of the action of vest 50, 332-35. androgenic steroids on erythropoiesis. J Pediatr Zucker S & Lysik R (1974) Bone marrow erythro83, 703-08. poiesis in anemia of inflammation. J Lab Clin Lukens J N (1973) Control of erythropoiesis in Med 84, 62&31.
