British Journal of Haematology, 1976,32, 167.
Erythropoiesis and Carbon Monoxide Production in Hodgkin’s Disease EVA CAVALLIN-STAHL, C. MERCJCE AND B. LUNDH Department of Medicine and Department of Radiotherapy, University Hospital, Lund, Sweden (Received 28 M a y 1975 ; acceptedfor publication 30Iufy 1975) SUMMARY.Endogenous production of carbon monoxide (Vco), red cell survival and iron kinetics were studied in 15 subjects with Hodgkin’s disease. The subjects were divided into two groups, namely: eight patients with anaemia (group A, haemoglobin (Hb) concentration less than 11.5 gldl) and seven patients without anaemia (group B, Hb concentration greater than 11.5 g/dl). Red cell survival was not significantly different in the two groups being 91 & 40 days (mean? I SD) in group A and I I I f 54 days in group B. Relative Vco (pmol/mmol total body haem (TBH)/d) was, however, significantly higher (0.01 > P> 0.001) in group A (20.7k4.7) compared to group B (12.0+ 3.8). When absolute VCO(pmolld) was compared to the daily turnover of circulating red cell haemoglobinhaem (Vhaem-c), the Vco/Vliaem-c quotient was 2.1fo.9 in group A and 1.2f0.3 in group B. Erythron turnover of iron (ET, pmol Felmmol TBH/d) was calculated through subtraction of the nm-erythron turnover (NET) from the total plasma iron turnover (PIT). ET was significantly higher (0.05 >P>o.oI) in group A (39) 21) than in group B (20f 8). The conclusion drawn froin the finding of sigiiificant increases in Vco and ET without any coiiconiitant significant decrease in red cell survival iii the anaemia group is that ineffective erythropoiesis, i.e. bone marrow haemolysis, seems to play an important role in the anaemia of Hodgkin’s disease. The anaemia secondary to malignancy and to chronic inflammatory states has been the subject of extensive investigation. The pathogenesis of this type of anaemia has been found to involve many factors such as a slightly shortened red cell lifespan (Hyman & Harvey, I955), impaired release of iron from the reticuloendothelial system (Haurani et al, 1965) and a deficiency in the erythropoietin mechanism (Zucker et af, 1974). Ineffective erythropoiesis has been reported by some workers (Cline & Berlin, 1963 ; Banerjee & Narang, 1967) while others report normal or increased red cell ’Fe utilization (Hyman & Harvey, 1955 ; Haurani et al, 1965). To elucidate the importance of ineffective erythropoiesis a study was delineated, where the endogenous production of carbon monoxide (Vco), as a measure of the total haem catabolism, was compared to red cell haemoglobin haem turnover and iron turnover in patients with Hodgkin’s disease. Correspondcncc: Dr Eva Cavallin-Stihl, Department of Medicine, University Hospital, Fack, S-221 85 L u d , Sweden.
167
Eva Cauallin-Stihl, C. Mercke and B. Lundh
168
MATERIALS AND METHODS
Patients Fifteen patients were studied. One patient (Case 2) was investigated twice. The diagnosis of Hodgkin’s disease was based on a lymph node biopsy in all cases. Histological classification was performed according to Lukes & Butler (1966) and clinical staging according to the Ann Arbor system (1971). The patients were divided into two groups. Group A consisted of eight patients with anaemia (haemoglobin (Hb) concentration less than 11.5 g/dl) and group B of seven patients without anaemia (Hb concentration greater than I 1.5 g/dl). All patients were clinically stable and thought to be in a steady state haematologically at the time of the study. Pertinent clinical and laboratory data are given in Table I. Carbon Monoxide Metabolism Endogenous carbon monoxide production (Vco) was measured with a closed rebreathing system according to Coburn et a1 (1966). After an equilibration period of20 min blood samples for the determination of carboxyhaemoglobinper cent saturation (COHb)were drawn every fifteenth minute for 2 h (nine samples) through an indwelling venous catheter using EDTA as anticoagulant. Then 20.0 ml of 99.9% pure carbon monoxide (CO) (AB Alfax, Malmo, Sweden) were injected into the rebreathing system whereafter a final blood sample was drawn after another 25 min. The CO content of the blood was determined according to Rodkey & Collison (1970) modified as described by Lundh et al (1972). 99.9% pure CO injected as described by Celegin et a1 (1971) was used for calibration. The haemoglobin concentration was determined according to van Kampen & Zijlstra (1965). The absolute (ml/h STPD) Vco was calculated by the formula:
Vco, nd/h = ACOHb/hx dilution factor
In this formula ACOHb is the increase in COHb, determined from the COHb of the first nine blood samples, using the least squares criterion to compute a best fit for the relation between time and COHb. The ‘dilution factor’ was determined through the measurement of the increase in COHb (ACOHbinj)after the injection of a known amount of C O (COinj) into the rebreathing system. The total C O binding capacity (CO B.C.), which is a prerequisite for the calculation of total body haem (TBH), was determined from the dilution factor: C O B.C., ml = COinjx Ioo/ACOHbinj After converting C O B.C. into g haemoglobin by dividing by 1.39, TBH (mmol) was obtained by dividing this figure by 16.1. In two patients (Cases 11 and 12) the determination of C O B.C. failed so TBH was calculated from total circulating haemoglobin (TBHb-c; see Red Cell Survival Studies) according to the formula : TBH =
(1.02
x TBHb-c, mmol+ 4.35) mmol (Lundh et a!, 1975)
The results of the Vco studies will be given absolutely and in relative values, pmol/mmol TBH/day.
A
74
57 27 44 58
28 24
M
M
M
8 M e a n f ~ SD
B
M
M
M
M
zs 66
62
65
68 70
73
58
66
14.4 13.2 14.6 13.4f0.9
13.9 12.9
12.5
12.2
10.3 9.711.2
8.6
I.5fo.6
1.5
2.0
2.6
1.0
0.7
1.3 1.3
r.gko.9
2.0
-
0.s
1.1
9.4
11.0
2.7
3.0
8.8
10.0
-
1.7 2.0
11.0
7.7
(%I
Reticulocyte count
10.5
Haemoglobin (gY)
-
6.8 8.6 8.6 8.8f1.6
8.6
12.0
166&26
IS0
140
190
170 130 I 80
200
8.6
8.6
165 188143
IS0
165
190 290
204.3
I 80 200 150
Ceruloplasmin (% of normal mean)
6.8 8.1+1.8
10.3
6.8 10.3 8.6
8.6 8.6
5.1
Serum bilirubin (amolil.)
111 A I1 A I1 A
I11 B IVA IA I11 A
IVB
111 B
111 B IV B IV B 111 A IV B IVB I1 B
Stage at time of the study
None None None
M.c.
M.c. L.d.
Splenectomy None None None
None
Chemoth.
None Radioth. Radioth. Radioth. Radioth. None None
Present
=
I
I
I
I
2
:
I
17
I
I
34 36 84
24
I
-
Duration of disease after diagnosis (months)
radiotherapy; Chemoth. =
None None None Radioth. since 5 days None None None
None None None MOPP since 6 days MOPP since 4 months None
None None -8
Treatment Previous
L.d. N.s. N.s. L.p.
M.c.
L.d.
N.s. M.c. M.c. M.c. N.s. L.p. N.s.
Histology
Abbreviations: L.P. = lymphocytic predominance. M.c. = mixed cellularity; N.s. = nodular sclerosing type; L.d. = Iymphocytic depletion; Radioth. chemotherapy; MOPP = quadruple chemotherapy &ith nitrogen mustard, vinca alkaloids, procarbazine and prednisolone (Devita et al, 1970). Group A (1-8), haemoglobin concentration < 11.5 g/dl; Group B @-IS), haemoglobin concentration > 11.5 g/dI. Vinblastin since 6 months and androgens in pharmacoIogica1doses.
Meanftr SD
1s
13 14
12
I1
10
9
M
79
s
M
7
65
32
i3
33
21
28
41
3 4
2:I 2:II
65 59 46 67
57 58
16
(kd
Weight
ZS
(ur)
Age
F M M M F F M
I
Sex
Patients
TABLE I. Clinical and laboratory data in patients with Hodgkin's disease
2
G !2!
%2"
3
-. s
3
0
B
2
3
kJ
170
Eva Cavallin-Stahl, C. Mercke and B. Lundh
Normal values. Vco measured in 20 young healthy males fasting in the morning after at least 36 h free from C O exposition, was found to be 10.8 f2.8 pmol/mmol TBH/d. The variation in Vco fiom month to month within the same individual has been calculated in our laboratory from 20 Vco studies in seven healthy males with a mean VCOof 12.3 pmol/mmol TBH and day. The standard deviation from month to month for each individual was 2.9 pmol corresponding to a coefficient of variation of 23%. The mean coefficient of variation for the regression coefficient used for calculation of Vco was 14% in this latter group.
Red Cell Survival Studies Determination of red cell survival was performed by labelling the patients) own red cells with "Cr according to method C from ICSH Panel on Diagnostic Application of Radioisotopes in Haematology (1971). After correction for elution of chromium according to these recommendations, the data were plotted in a linear diagram and red cell survival was calculated from the slope of a straight line fitted to all the data points. TBHb-c was obtained from the dilution of the 5 1 Cr-labelled red cells. The rate of daily catabolism of circulating haemoglobin haem (Vhaem-c, pmolld) was calculated by dividing TBHb-c, mmol, by red cell survival in days.
Iron Kinetic Studies The ferrokinetic studies were performed in the morning with a dose of 5 pCi of 59Fe citrate (sp.a. more than 10 mCi/mg) (AB Atomenergi, Studsvik, Sweden). The tracer was incubated for 20 min with donor plasma, with a total iron binding capacity of at least 10 times the amount of added iron. Heparinized blood samples were drawn at 5 , 15, 30, 40, so, 60,90 and 120 min after the injection. Radioactivity in 2 ml of plasma was measured in a well-type scintillation detector. The data obtained were plotted on semilogarithmic coordinates. A regression line was then fitted to the logarithm of these activities by the method of least squares and the zero time activity calculated as the antilogarithm of the y intercept. Plasma iron turnover (PIT) (pmol ironll. whole bloodld) was calculated according to the formula given by Cook et a1 (1970) : PIT = PIXPCT,,/T+ where PI is the plasma iron (pmol/l.), PCT,, is the whole body plasmatocrit (I-PCVx 0.0092) and the T+is the half-time of the initial clearance in minutes. Non-erythron turnover of iron (NET) (pmol/l. whole bloodld) was calculated according to the formula:
NET
=
PI x PCTwbx 0.0035
and erythron turnover of iron (ET) (pmol/l. whole bloodld) was obtained by the subtraction of NET from PIT (Cook et al, 1970).
Other Methods Bone marrow smears were obtained by aspiration from the sternum or the iliac crest and stained with May-Griinwald-Giemsa. Serum iron was determined according to Young & Hicks (1965), serum bilirubin according to Gainbino & Schreiber (1964) and ceruloplasmin according to Laurel1 (I 972).
CO Production in Hodgkin’s Disease
171
TABLE 11. Carbon monoxide production and haem catabolism data in patients with Hodgkin’s disease
vco TBHb-c
Patient
(mmol)
A
I
29.4 19.7 32.7 23.3
2:1 2:II
TBH (mmol)
Red cell survival
Vhaem-c
(,umol/d)
pmol/d
pmol/mniol TBHId
Vco/Vhaern-c
(4
28.4
92 50 82 62 I44 68 83 170 64 91 k 4 0
321 3 90 399 378 I44 416 414 258 466 354k99
696 740 1232 545 406 502 569 942 672
24.3 23.1 28.3 20.6 15.8 16.3 14.8 24.8 18.0
2.2
700k253
20.7k4.7
Z.IkO.9
I33 65 62 84 68 168 I94 111k54
264 675 704 416 72 1 269 344 485f208
342 678 626 685 61 I
1.3 0.9 1.7 0.9
563
9.4 14.5 13.2 17.6 11.7 5.6 11.9
533+178
12.0+3.5
1.2fk0.3
32.0
28.5 34.2 36.3 28.4
43-5 26.5 25.7 30.9 38.5 38.0 37.4
M e a n k I SD
28.1k6.0
33.4k6.2
B
35.2 43.8 43.2 34.7 49.4 37.9 48.5 41.826.0
36.6 46.7 46.7 39.1
3 4 5 6 7 8
20.5
g I0 I1
I2
13 14 IS
M e a n k I SD
52.1
40.2 47.2 44.1k5.5
255
quotient
1-9 3.1 1.4 2.8 I .2
1.4 3.7 1.7
1.0
1.0
1.6
____
Abbreviations: TBHb-c = total circulating haemoglobin haem; TBH = total body haem; Vhaem-c = daily catabolism of circulating haemoglobin haem; Vco = endogenous production of carbon monoxide. Group A (I-8), haemoglobin concentration < 11.5 g/dl; Group B (9-IS), haenioglobin concentration > 11.5 g/dl.
N
a
0
‘f a
a
a
FIG I. Endogenous production of carbon monoxide (vco)in Hodgkin’s disease. A = eight patients (one studied twice) with haemoglobin concentration < 11.5 g/dl; B = seven patients with haemoglobin concentration 11.5 g/dl; N = 20 healthy young males. The bars to the right indicate mean +I
SD.
Eva Cavallin-Stahl, C. Mercke and B. Lundh
172
RESULTS
Carbon Mo,ioxide Production and Haem Catabolism (Table 11) In group A relative Pco (pmol/mmol TBH/d) was z0.7l. 4.7 (meanf I SD) compared to 12.0k 3.8 in group B. The difference is significant (P< 0.01). Vco in group B is similar to Vco in a control group consisting of 20 healthy young men (Fig I). TABLE 111. Ferrokmetic data in patients with Hodgkin's disease
Patients
PI (pcmolil.)
A
I 2:T
2:II 3 4 5 6 7 8 Mean* I SD B
9 I0
I1 I2
I3 I4 IS Meankr SD
8.1 8.1 17.9 7.2 11.6 8.1 8.1 43.9 9.0 13.6f11.9
PI T+ (rnin)
PIT
NET
ET
(pumol/l. bloodld)
(pmalll. bloodld)
(pmol/l.
19.7 21.5 43.0 17.9 28.6 19.7 (17.9) 116.4 23.3
IS9
32 SO
34 I9 23 25
(76) 59 37 35+14
8.1 6.3 12.5 18.8 18.8 31.3 7.2 14.71 9.0
66 35 63 36 48 k I S
17912.7
85f12
55 28 50
ET (pmol/d)
bloodld)
I02
79s 459 1638 1227 1192
315 261 331 213 (5s) 448 IS0
1150 (308) 3136 735
36.3f33.4
247+114
1292k829
17.9 14.3 26.9 43.0 39.4 69.8 16.1 32.2 f 19.7
88 I34 131 141 277 249 I09 161 & 72
403 831 721 682 1696 1178 709 8895424
26.9f5.4
I07f9
ET (pmollmmol TBHId) 28.4 14.3 37.7 48.3 46.4 37.2 (8.0)
82.5
19.7 39.3 f21.2 11.0
17.8 15.2
17.4 32.6 29-3 15.2 19.8+ 8.0
Normal values +I
SD
13419
Normal values were taken from Cook et a2 (1970) and recalculated according to the Systkme International #Unites. Abbreviations: PI = plasma iron; PI T+= plasma iron initial T,; PIT = plasma iron turnover; NET = nonerythron turnover of iron; ET = erythron turnover of iron. Group A ( I - ~ ) ,haemoglobin concentration < 11.5 g/dl; Group B (9-IS), haemoglobin concentration > 11.5 g/dl.
Red cell survival in group A was 91 k 40 days with a range between 50 and 170 days corresponding to a mean Vhaem-c of 354k 99 pmol/d. Red cell lifespan in group B was 111k 54 d, range 62-194 and mean Vhaem-c was 485 k 208 pmol/d. The difference between red cell survival in the two groups was not significant. For all patients absolute Vco (pmol/d) was compared to Vhaem-c in the VcolVhaem-c quotient. In group A this ratio was 2.1ko.g and in group B 1.2k0.3. The difference is probably significant (P< 0.05).
Iron Kinetics (Table 111) In both groups the initial plasma half-time of injected iron isotope was short, 35rt 15 min
CO Production in Hodgkin’s Disease
I73
in group A and 48 f 15 min in group B. Plasma iron turnover (PIT) was markedly increased in group A (284f 142pmol/l. whole bloodld) and moderately increased in group B (1g4f 88) compared to the normal value (134f 9) given by Cook et a! (1970). The difference in halftime and PIT between groups A and B is not significant. However, a probably significant difference was found in erythron turnover of iron (ET) (P< o.os),which, calculated per mmol TBH/d, was found to be 39.3rf:zr.z pmol in group A and 19.828.0 pmol in group B. One patient in group A (Case 6) was excluded from the calculations since the results markedly deviated from the other patients in the group. This patient had received his first quadruple chemotherapy treatment (Devita et al, 1970) 6 d before the study of iron kinetics, and probably therapy caused a temporary bone marrow inhibition. DISCUSSION Recent advances in histological classification and clinical staging of Hodgkin’s disease have made patients with this disease suitable for comparative clinical research. This was one of the reasons why such patients were chosen for the present study on the mechanism of anaemia secondary to malignant diseases. Even though the methods used have many shortcomings which have been discussed recently (Lundh et al, 1975), they may be of value especially when reasonably large groups of similar patients are compared. The emphasis has to be laid on the relation between the haem and iron kinetic studies within the same patient as well as on a comparison between the mean values of patients with and without anaemia. The mean duration of the disease in the anaemia group was 21 months (range 1-84 months) against 1.4 months (range 1-3 months) in the patients without anaemia and the groups differed in the amount of treatment received and in the extension of the disease, respectively (for details, see Table I). However, this comparison is important, since we have no reference values of our own due to ethical reasons. Red cell survival in both groups was normal or slightly shortened and although the mean survival in the anaemia group was somewhat shorter, the difference was not significant. Some survival values in both groups were extremely long, over 150 d, even though the patients were in steady state. These results may be due to the error of the method, which is large, especially in the normal or near-normal range. Of course it cannot be excluded that red cell survival might be slightly shortened in anaemic patients with Hodgkin’s disease, as was recently described by Beamish et al (1972) in their study of untreated cases. Anyhow this fact does not invalidate the following discussion. Hitherto, endogenous production of carbon monoxide (Vco) has not been studied in patients with anaemia of chronic disease. White et al (1967) have, however, reported that Vco greatly exceeds Vhaem-c in patients with refractory anaemia and hypercellular bone marrow and they interpreted this high Vco/Vhaem-c quotient as a sign on ineffective erythropoiesis. Their findings were recently confirmed (Lundh et al, 1975). The result of the present study with markedly increased Vco and high Vco/Vhaem-c quotients in patients with Hodgkin’s disease and anaemia compared to patients without anaemia, suggest that ineffective erythropoiesis may play an important role in the mechanism of anaemia, especially since there was a close correlation between the VcolVhaem-c quotient and the degree of anaemia (r = 0.84, n = 16; P< 0.001) (Fig 2). There are no reasons to believe that the increase in Vco was due to an increase in turnover of liver haem or other haem pools.
Eva Cauallin-Stihl, C. Mercke and B. Lundh
I74
3
I
7
e aJ
;2 u
\0 O O
\ c
0
0
-
0
I -
\
0
I 5
I 10 Hb(g/dl)
I
15
FIG 2. Relation between vco/Vhaem-c quotient and haemoglobin concentration in 15 patients (one studied twice) with Hodgkin’s disease.
One patient (Case 2) was studied twice. At first Vco was high and the disease active with marked anaemia. Later Vco was still high, which was surprising since the patient then was thought to be free from active disease. However, only a week after the second study he relapsed with rapidly developing anaemia and died within a few months. Even though the validity of the formula for calculation of ET has not been tested in anaemia of malignancy, the formula was shown to be valid in more than 40 patients with various types of anaemia (Cook et al, 1970). W e have not been able to find anything contradicting their conclusion that the formula should be valid for all types of anaemia. There was a significant correlation between lico/Vhaem-c quotient and ET calculated per mmol TBH and day (r = 0.68, n = 15; Pco.01). When ET was calculated in pmol/d it was found to exceed the observed absolute Vco by almost 100% in the anaemia group, while the difference was moderate in patients without anaemia. This might reflect the increased amount of ‘wastage’ iron in this type of anaemia; i.e. iron taken up by the bone marrow erythroblasts but not incorporated into haem. Another explanation for the difference between ET and Vco might be that the PIT values obtained were too high due to a rapid clearance of unbound iron. However, calculations were always made so that the iron binding capacity during the incubation was well above the amount of added iron. Furthermore, PIT, even though higher than the values obtained by Beamish et al (1972), was quite comparable to that reported by other groups (Giannopoulos & Bergsagel, 1959; Cline & Berlin, 1963) in Hodgkin’s disease. The difference between our results and those of Beamish et a1 might be due to the fact that almost none of the patients in the latter study was anaemic. ACKNOWLEDGMENTS
This investigation was supported by grants from the Swedish Medical Research Council
CO Production in Hodskin’s Disease
I75
(Projects No. 19 X-3523 and 61P-3611), from the Medical Faculty, University of Lund, from John and Augusta Persson’s Foundation for Scientific Medical Research and from Lotten Bohman’s Foundation. REFERENCES BANERJEE, R.N. & NARANG, R.M. (1967) Haematological changes in malignancy. British Journal of Haematology, 13,829. BEAMISH,M.R., ASHLEYJONES,P., TREVETT,D., HOWELL EVANS,I. &JACOBS, A. (1972) Iron metabolism in Hodgkin’s discase. British Journal of Cancer, 26, 444. CELEGIN, M., HANSSON, R. & SUNDSTROM, G. (1971) A sub-microliter sampling device for quantitative collection of gases. Application to ‘gas-liquid chromatographic’ analysis. Scandinavian Journal of Clinical, and Laboratory Ifivestigation, 27, 367. CLINE,M.J. & BERLIN,N.I. (1963) Anemia in Hodgkin’s disease. Cancer, 16,526. COBURN, R.F., WILLIAMS, W.J. & KAHN,S.B. (1966) Endogenous carbon monoxide production in patients with hemolytic anemia. Journal of Clinical Investigation, 45, 460. COOK,J.D., MARSAGLIA, G., ESCHBACH, J.W., FUNK, D.D. & FINCH,C.A. (1970) Ferrokinetics: a biologic model for plasma iron exchange in man. Journal of Clinical Investigation, 49, 197. DEVITA,V.T., SERPICK, A.A. & CARBONE, P.P. (1970) Combination chemotherapy in the treatment of advanced Hodgkin’s disease. Anrials of Infernal Medicine, 73, 881. GAMBINO, S.R. & SCHREIBER, H. (1964) The measurcnient of bilirubin on the autoanalyzer by the method of Jendrassik and Grof. Technicon Symposion, paper 54. GIANNOPOULOS, P.P. & BERGSAGEL, D.E. (1959) The mechanism of the anemia associated with Hodgkin’s disease. Blood, 14,856. E.J. (1965) HAURANI, F.I., BURKE,W. & MARTINEZ, Defective reutilization of iron in the anemia of inflammation. Journal of Laboratory and Clinical Medicine, 6 5 560. HYMAN, G.A. &HARVEY, J.E. (1955) Thc pathogenesis of anaemia in patients with carcinoma. American Journal of Medicine, 19, 350. INTERNATIONAL COMMITTEE FOR STANDARDIZATION IN
C
HEMATOLOGY (1971) Recommended methods for radioisotopc red cell survival studies. Blood, 38, 378. LAURELL, C.-B. (1972) Electroimmuno assay. Scandinavian Jourual of Clinical and Laboratory Investigation, Suppl. 124.21. LUKES, R.J. & BUTLER, J.J. (1966) The pathology and nomenclature of Hodgkin’s disease. Cancer Research, 26, 1310. LUNDH,B.,CAVALLIN-STAHL, E. & MERCKE, C. (1975) Haem catabolism, carbon monoxide production and red cell survival in anaemia. A d a Medica Scandinavica, 197, 161. LUNDH, B., JOHANSSON, M.-B., MERCKE,C. & CAVALLIN-ST~L, E. (1972) Enhancement of hemecatabolism by caloric restriction in man. Scandinavian Journal of Clinical and Laboratory Investigation, 30, 421. RODKEY, F.L. & COLLISON, H.A. (1970) An artifact in the analysis of oxygenated blood for its low carbon monoxide content. Clinical Chemistry, 16,896. ON STAGING IN HODGKIN’S DISEASE (1971) SYMPOSIUM Cancer Research, 31, 1707. VAN KAMPEN, E.J. & ZIJLSTRA, W.G. (1965) Determination of hemoglobin and its derivatives. Advances in Clinical Chemistry, 8, 141. WEBB,D.I., UBOGY,G. & SILVER, R.T. (1970) Importance of bone marrow biopsy in the clinical staging of Hodgkin’s disease. Cancer, 26, 3 I 3. WHITE,P., COBURN, R.F., WILLIAMS, W.J., GOLDB.C. (1967) WEN,M.I., ROTHER,M.L. & SHAFER, Carbon monoxide production associated with ineffective erythropoiesis. Journal of Clinical Investigation, 46, 1986. YOUNG,D.S. & HICKS,J.M. (1965) Method for the automatic determination of serum iron. Journal of Clinical Pathology, 18, 98. ZUCKER,S., FRIEDMAN, S. & LYSIK, R.M. (1974) Bone marrow erythropoiesis in the anemia of infection, inflammation and malignancy. Journal of Clinical Investigation, 53, 1132.
Erythropoiesis and Carbon Monoxide Production in Hodgkin’s Disease EVA CAVALLIN-STAHL, C. MERCJCE AND B. LUNDH Department of Medicine and Department of Radiotherapy, University Hospital, Lund, Sweden (Received 28 M a y 1975 ; acceptedfor publication 30Iufy 1975) SUMMARY.Endogenous production of carbon monoxide (Vco), red cell survival and iron kinetics were studied in 15 subjects with Hodgkin’s disease. The subjects were divided into two groups, namely: eight patients with anaemia (group A, haemoglobin (Hb) concentration less than 11.5 gldl) and seven patients without anaemia (group B, Hb concentration greater than 11.5 g/dl). Red cell survival was not significantly different in the two groups being 91 & 40 days (mean? I SD) in group A and I I I f 54 days in group B. Relative Vco (pmol/mmol total body haem (TBH)/d) was, however, significantly higher (0.01 > P> 0.001) in group A (20.7k4.7) compared to group B (12.0+ 3.8). When absolute VCO(pmolld) was compared to the daily turnover of circulating red cell haemoglobinhaem (Vhaem-c), the Vco/Vliaem-c quotient was 2.1fo.9 in group A and 1.2f0.3 in group B. Erythron turnover of iron (ET, pmol Felmmol TBH/d) was calculated through subtraction of the nm-erythron turnover (NET) from the total plasma iron turnover (PIT). ET was significantly higher (0.05 >P>o.oI) in group A (39) 21) than in group B (20f 8). The conclusion drawn froin the finding of sigiiificant increases in Vco and ET without any coiiconiitant significant decrease in red cell survival iii the anaemia group is that ineffective erythropoiesis, i.e. bone marrow haemolysis, seems to play an important role in the anaemia of Hodgkin’s disease. The anaemia secondary to malignancy and to chronic inflammatory states has been the subject of extensive investigation. The pathogenesis of this type of anaemia has been found to involve many factors such as a slightly shortened red cell lifespan (Hyman & Harvey, I955), impaired release of iron from the reticuloendothelial system (Haurani et al, 1965) and a deficiency in the erythropoietin mechanism (Zucker et af, 1974). Ineffective erythropoiesis has been reported by some workers (Cline & Berlin, 1963 ; Banerjee & Narang, 1967) while others report normal or increased red cell ’Fe utilization (Hyman & Harvey, 1955 ; Haurani et al, 1965). To elucidate the importance of ineffective erythropoiesis a study was delineated, where the endogenous production of carbon monoxide (Vco), as a measure of the total haem catabolism, was compared to red cell haemoglobin haem turnover and iron turnover in patients with Hodgkin’s disease. Correspondcncc: Dr Eva Cavallin-Stihl, Department of Medicine, University Hospital, Fack, S-221 85 L u d , Sweden.
167
Eva Cauallin-Stihl, C. Mercke and B. Lundh
168
MATERIALS AND METHODS
Patients Fifteen patients were studied. One patient (Case 2) was investigated twice. The diagnosis of Hodgkin’s disease was based on a lymph node biopsy in all cases. Histological classification was performed according to Lukes & Butler (1966) and clinical staging according to the Ann Arbor system (1971). The patients were divided into two groups. Group A consisted of eight patients with anaemia (haemoglobin (Hb) concentration less than 11.5 g/dl) and group B of seven patients without anaemia (Hb concentration greater than I 1.5 g/dl). All patients were clinically stable and thought to be in a steady state haematologically at the time of the study. Pertinent clinical and laboratory data are given in Table I. Carbon Monoxide Metabolism Endogenous carbon monoxide production (Vco) was measured with a closed rebreathing system according to Coburn et a1 (1966). After an equilibration period of20 min blood samples for the determination of carboxyhaemoglobinper cent saturation (COHb)were drawn every fifteenth minute for 2 h (nine samples) through an indwelling venous catheter using EDTA as anticoagulant. Then 20.0 ml of 99.9% pure carbon monoxide (CO) (AB Alfax, Malmo, Sweden) were injected into the rebreathing system whereafter a final blood sample was drawn after another 25 min. The CO content of the blood was determined according to Rodkey & Collison (1970) modified as described by Lundh et al (1972). 99.9% pure CO injected as described by Celegin et a1 (1971) was used for calibration. The haemoglobin concentration was determined according to van Kampen & Zijlstra (1965). The absolute (ml/h STPD) Vco was calculated by the formula:
Vco, nd/h = ACOHb/hx dilution factor
In this formula ACOHb is the increase in COHb, determined from the COHb of the first nine blood samples, using the least squares criterion to compute a best fit for the relation between time and COHb. The ‘dilution factor’ was determined through the measurement of the increase in COHb (ACOHbinj)after the injection of a known amount of C O (COinj) into the rebreathing system. The total C O binding capacity (CO B.C.), which is a prerequisite for the calculation of total body haem (TBH), was determined from the dilution factor: C O B.C., ml = COinjx Ioo/ACOHbinj After converting C O B.C. into g haemoglobin by dividing by 1.39, TBH (mmol) was obtained by dividing this figure by 16.1. In two patients (Cases 11 and 12) the determination of C O B.C. failed so TBH was calculated from total circulating haemoglobin (TBHb-c; see Red Cell Survival Studies) according to the formula : TBH =
(1.02
x TBHb-c, mmol+ 4.35) mmol (Lundh et a!, 1975)
The results of the Vco studies will be given absolutely and in relative values, pmol/mmol TBH/day.
A
74
57 27 44 58
28 24
M
M
M
8 M e a n f ~ SD
B
M
M
M
M
zs 66
62
65
68 70
73
58
66
14.4 13.2 14.6 13.4f0.9
13.9 12.9
12.5
12.2
10.3 9.711.2
8.6
I.5fo.6
1.5
2.0
2.6
1.0
0.7
1.3 1.3
r.gko.9
2.0
-
0.s
1.1
9.4
11.0
2.7
3.0
8.8
10.0
-
1.7 2.0
11.0
7.7
(%I
Reticulocyte count
10.5
Haemoglobin (gY)
-
6.8 8.6 8.6 8.8f1.6
8.6
12.0
166&26
IS0
140
190
170 130 I 80
200
8.6
8.6
165 188143
IS0
165
190 290
204.3
I 80 200 150
Ceruloplasmin (% of normal mean)
6.8 8.1+1.8
10.3
6.8 10.3 8.6
8.6 8.6
5.1
Serum bilirubin (amolil.)
111 A I1 A I1 A
I11 B IVA IA I11 A
IVB
111 B
111 B IV B IV B 111 A IV B IVB I1 B
Stage at time of the study
None None None
M.c.
M.c. L.d.
Splenectomy None None None
None
Chemoth.
None Radioth. Radioth. Radioth. Radioth. None None
Present
=
I
I
I
I
2
:
I
17
I
I
34 36 84
24
I
-
Duration of disease after diagnosis (months)
radiotherapy; Chemoth. =
None None None Radioth. since 5 days None None None
None None None MOPP since 6 days MOPP since 4 months None
None None -8
Treatment Previous
L.d. N.s. N.s. L.p.
M.c.
L.d.
N.s. M.c. M.c. M.c. N.s. L.p. N.s.
Histology
Abbreviations: L.P. = lymphocytic predominance. M.c. = mixed cellularity; N.s. = nodular sclerosing type; L.d. = Iymphocytic depletion; Radioth. chemotherapy; MOPP = quadruple chemotherapy &ith nitrogen mustard, vinca alkaloids, procarbazine and prednisolone (Devita et al, 1970). Group A (1-8), haemoglobin concentration < 11.5 g/dl; Group B @-IS), haemoglobin concentration > 11.5 g/dI. Vinblastin since 6 months and androgens in pharmacoIogica1doses.
Meanftr SD
1s
13 14
12
I1
10
9
M
79
s
M
7
65
32
i3
33
21
28
41
3 4
2:I 2:II
65 59 46 67
57 58
16
(kd
Weight
ZS
(ur)
Age
F M M M F F M
I
Sex
Patients
TABLE I. Clinical and laboratory data in patients with Hodgkin's disease
2
G !2!
%2"
3
-. s
3
0
B
2
3
kJ
170
Eva Cavallin-Stahl, C. Mercke and B. Lundh
Normal values. Vco measured in 20 young healthy males fasting in the morning after at least 36 h free from C O exposition, was found to be 10.8 f2.8 pmol/mmol TBH/d. The variation in Vco fiom month to month within the same individual has been calculated in our laboratory from 20 Vco studies in seven healthy males with a mean VCOof 12.3 pmol/mmol TBH and day. The standard deviation from month to month for each individual was 2.9 pmol corresponding to a coefficient of variation of 23%. The mean coefficient of variation for the regression coefficient used for calculation of Vco was 14% in this latter group.
Red Cell Survival Studies Determination of red cell survival was performed by labelling the patients) own red cells with "Cr according to method C from ICSH Panel on Diagnostic Application of Radioisotopes in Haematology (1971). After correction for elution of chromium according to these recommendations, the data were plotted in a linear diagram and red cell survival was calculated from the slope of a straight line fitted to all the data points. TBHb-c was obtained from the dilution of the 5 1 Cr-labelled red cells. The rate of daily catabolism of circulating haemoglobin haem (Vhaem-c, pmolld) was calculated by dividing TBHb-c, mmol, by red cell survival in days.
Iron Kinetic Studies The ferrokinetic studies were performed in the morning with a dose of 5 pCi of 59Fe citrate (sp.a. more than 10 mCi/mg) (AB Atomenergi, Studsvik, Sweden). The tracer was incubated for 20 min with donor plasma, with a total iron binding capacity of at least 10 times the amount of added iron. Heparinized blood samples were drawn at 5 , 15, 30, 40, so, 60,90 and 120 min after the injection. Radioactivity in 2 ml of plasma was measured in a well-type scintillation detector. The data obtained were plotted on semilogarithmic coordinates. A regression line was then fitted to the logarithm of these activities by the method of least squares and the zero time activity calculated as the antilogarithm of the y intercept. Plasma iron turnover (PIT) (pmol ironll. whole bloodld) was calculated according to the formula given by Cook et a1 (1970) : PIT = PIXPCT,,/T+ where PI is the plasma iron (pmol/l.), PCT,, is the whole body plasmatocrit (I-PCVx 0.0092) and the T+is the half-time of the initial clearance in minutes. Non-erythron turnover of iron (NET) (pmol/l. whole bloodld) was calculated according to the formula:
NET
=
PI x PCTwbx 0.0035
and erythron turnover of iron (ET) (pmol/l. whole bloodld) was obtained by the subtraction of NET from PIT (Cook et al, 1970).
Other Methods Bone marrow smears were obtained by aspiration from the sternum or the iliac crest and stained with May-Griinwald-Giemsa. Serum iron was determined according to Young & Hicks (1965), serum bilirubin according to Gainbino & Schreiber (1964) and ceruloplasmin according to Laurel1 (I 972).
CO Production in Hodgkin’s Disease
171
TABLE 11. Carbon monoxide production and haem catabolism data in patients with Hodgkin’s disease
vco TBHb-c
Patient
(mmol)
A
I
29.4 19.7 32.7 23.3
2:1 2:II
TBH (mmol)
Red cell survival
Vhaem-c
(,umol/d)
pmol/d
pmol/mniol TBHId
Vco/Vhaern-c
(4
28.4
92 50 82 62 I44 68 83 170 64 91 k 4 0
321 3 90 399 378 I44 416 414 258 466 354k99
696 740 1232 545 406 502 569 942 672
24.3 23.1 28.3 20.6 15.8 16.3 14.8 24.8 18.0
2.2
700k253
20.7k4.7
Z.IkO.9
I33 65 62 84 68 168 I94 111k54
264 675 704 416 72 1 269 344 485f208
342 678 626 685 61 I
1.3 0.9 1.7 0.9
563
9.4 14.5 13.2 17.6 11.7 5.6 11.9
533+178
12.0+3.5
1.2fk0.3
32.0
28.5 34.2 36.3 28.4
43-5 26.5 25.7 30.9 38.5 38.0 37.4
M e a n k I SD
28.1k6.0
33.4k6.2
B
35.2 43.8 43.2 34.7 49.4 37.9 48.5 41.826.0
36.6 46.7 46.7 39.1
3 4 5 6 7 8
20.5
g I0 I1
I2
13 14 IS
M e a n k I SD
52.1
40.2 47.2 44.1k5.5
255
quotient
1-9 3.1 1.4 2.8 I .2
1.4 3.7 1.7
1.0
1.0
1.6
____
Abbreviations: TBHb-c = total circulating haemoglobin haem; TBH = total body haem; Vhaem-c = daily catabolism of circulating haemoglobin haem; Vco = endogenous production of carbon monoxide. Group A (I-8), haemoglobin concentration < 11.5 g/dl; Group B (9-IS), haenioglobin concentration > 11.5 g/dl.
N
a
0
‘f a
a
a
FIG I. Endogenous production of carbon monoxide (vco)in Hodgkin’s disease. A = eight patients (one studied twice) with haemoglobin concentration < 11.5 g/dl; B = seven patients with haemoglobin concentration 11.5 g/dl; N = 20 healthy young males. The bars to the right indicate mean +I
SD.
Eva Cavallin-Stahl, C. Mercke and B. Lundh
172
RESULTS
Carbon Mo,ioxide Production and Haem Catabolism (Table 11) In group A relative Pco (pmol/mmol TBH/d) was z0.7l. 4.7 (meanf I SD) compared to 12.0k 3.8 in group B. The difference is significant (P< 0.01). Vco in group B is similar to Vco in a control group consisting of 20 healthy young men (Fig I). TABLE 111. Ferrokmetic data in patients with Hodgkin's disease
Patients
PI (pcmolil.)
A
I 2:T
2:II 3 4 5 6 7 8 Mean* I SD B
9 I0
I1 I2
I3 I4 IS Meankr SD
8.1 8.1 17.9 7.2 11.6 8.1 8.1 43.9 9.0 13.6f11.9
PI T+ (rnin)
PIT
NET
ET
(pumol/l. bloodld)
(pmalll. bloodld)
(pmol/l.
19.7 21.5 43.0 17.9 28.6 19.7 (17.9) 116.4 23.3
IS9
32 SO
34 I9 23 25
(76) 59 37 35+14
8.1 6.3 12.5 18.8 18.8 31.3 7.2 14.71 9.0
66 35 63 36 48 k I S
17912.7
85f12
55 28 50
ET (pmol/d)
bloodld)
I02
79s 459 1638 1227 1192
315 261 331 213 (5s) 448 IS0
1150 (308) 3136 735
36.3f33.4
247+114
1292k829
17.9 14.3 26.9 43.0 39.4 69.8 16.1 32.2 f 19.7
88 I34 131 141 277 249 I09 161 & 72
403 831 721 682 1696 1178 709 8895424
26.9f5.4
I07f9
ET (pmollmmol TBHId) 28.4 14.3 37.7 48.3 46.4 37.2 (8.0)
82.5
19.7 39.3 f21.2 11.0
17.8 15.2
17.4 32.6 29-3 15.2 19.8+ 8.0
Normal values +I
SD
13419
Normal values were taken from Cook et a2 (1970) and recalculated according to the Systkme International #Unites. Abbreviations: PI = plasma iron; PI T+= plasma iron initial T,; PIT = plasma iron turnover; NET = nonerythron turnover of iron; ET = erythron turnover of iron. Group A ( I - ~ ) ,haemoglobin concentration < 11.5 g/dl; Group B (9-IS), haemoglobin concentration > 11.5 g/dl.
Red cell survival in group A was 91 k 40 days with a range between 50 and 170 days corresponding to a mean Vhaem-c of 354k 99 pmol/d. Red cell lifespan in group B was 111k 54 d, range 62-194 and mean Vhaem-c was 485 k 208 pmol/d. The difference between red cell survival in the two groups was not significant. For all patients absolute Vco (pmol/d) was compared to Vhaem-c in the VcolVhaem-c quotient. In group A this ratio was 2.1ko.g and in group B 1.2k0.3. The difference is probably significant (P< 0.05).
Iron Kinetics (Table 111) In both groups the initial plasma half-time of injected iron isotope was short, 35rt 15 min
CO Production in Hodgkin’s Disease
I73
in group A and 48 f 15 min in group B. Plasma iron turnover (PIT) was markedly increased in group A (284f 142pmol/l. whole bloodld) and moderately increased in group B (1g4f 88) compared to the normal value (134f 9) given by Cook et a! (1970). The difference in halftime and PIT between groups A and B is not significant. However, a probably significant difference was found in erythron turnover of iron (ET) (P< o.os),which, calculated per mmol TBH/d, was found to be 39.3rf:zr.z pmol in group A and 19.828.0 pmol in group B. One patient in group A (Case 6) was excluded from the calculations since the results markedly deviated from the other patients in the group. This patient had received his first quadruple chemotherapy treatment (Devita et al, 1970) 6 d before the study of iron kinetics, and probably therapy caused a temporary bone marrow inhibition. DISCUSSION Recent advances in histological classification and clinical staging of Hodgkin’s disease have made patients with this disease suitable for comparative clinical research. This was one of the reasons why such patients were chosen for the present study on the mechanism of anaemia secondary to malignant diseases. Even though the methods used have many shortcomings which have been discussed recently (Lundh et al, 1975), they may be of value especially when reasonably large groups of similar patients are compared. The emphasis has to be laid on the relation between the haem and iron kinetic studies within the same patient as well as on a comparison between the mean values of patients with and without anaemia. The mean duration of the disease in the anaemia group was 21 months (range 1-84 months) against 1.4 months (range 1-3 months) in the patients without anaemia and the groups differed in the amount of treatment received and in the extension of the disease, respectively (for details, see Table I). However, this comparison is important, since we have no reference values of our own due to ethical reasons. Red cell survival in both groups was normal or slightly shortened and although the mean survival in the anaemia group was somewhat shorter, the difference was not significant. Some survival values in both groups were extremely long, over 150 d, even though the patients were in steady state. These results may be due to the error of the method, which is large, especially in the normal or near-normal range. Of course it cannot be excluded that red cell survival might be slightly shortened in anaemic patients with Hodgkin’s disease, as was recently described by Beamish et al (1972) in their study of untreated cases. Anyhow this fact does not invalidate the following discussion. Hitherto, endogenous production of carbon monoxide (Vco) has not been studied in patients with anaemia of chronic disease. White et al (1967) have, however, reported that Vco greatly exceeds Vhaem-c in patients with refractory anaemia and hypercellular bone marrow and they interpreted this high Vco/Vhaem-c quotient as a sign on ineffective erythropoiesis. Their findings were recently confirmed (Lundh et al, 1975). The result of the present study with markedly increased Vco and high Vco/Vhaem-c quotients in patients with Hodgkin’s disease and anaemia compared to patients without anaemia, suggest that ineffective erythropoiesis may play an important role in the mechanism of anaemia, especially since there was a close correlation between the VcolVhaem-c quotient and the degree of anaemia (r = 0.84, n = 16; P< 0.001) (Fig 2). There are no reasons to believe that the increase in Vco was due to an increase in turnover of liver haem or other haem pools.
Eva Cauallin-Stihl, C. Mercke and B. Lundh
I74
3
I
7
e aJ
;2 u
\0 O O
\ c
0
0
-
0
I -
\
0
I 5
I 10 Hb(g/dl)
I
15
FIG 2. Relation between vco/Vhaem-c quotient and haemoglobin concentration in 15 patients (one studied twice) with Hodgkin’s disease.
One patient (Case 2) was studied twice. At first Vco was high and the disease active with marked anaemia. Later Vco was still high, which was surprising since the patient then was thought to be free from active disease. However, only a week after the second study he relapsed with rapidly developing anaemia and died within a few months. Even though the validity of the formula for calculation of ET has not been tested in anaemia of malignancy, the formula was shown to be valid in more than 40 patients with various types of anaemia (Cook et al, 1970). W e have not been able to find anything contradicting their conclusion that the formula should be valid for all types of anaemia. There was a significant correlation between lico/Vhaem-c quotient and ET calculated per mmol TBH and day (r = 0.68, n = 15; Pco.01). When ET was calculated in pmol/d it was found to exceed the observed absolute Vco by almost 100% in the anaemia group, while the difference was moderate in patients without anaemia. This might reflect the increased amount of ‘wastage’ iron in this type of anaemia; i.e. iron taken up by the bone marrow erythroblasts but not incorporated into haem. Another explanation for the difference between ET and Vco might be that the PIT values obtained were too high due to a rapid clearance of unbound iron. However, calculations were always made so that the iron binding capacity during the incubation was well above the amount of added iron. Furthermore, PIT, even though higher than the values obtained by Beamish et al (1972), was quite comparable to that reported by other groups (Giannopoulos & Bergsagel, 1959; Cline & Berlin, 1963) in Hodgkin’s disease. The difference between our results and those of Beamish et a1 might be due to the fact that almost none of the patients in the latter study was anaemic. ACKNOWLEDGMENTS
This investigation was supported by grants from the Swedish Medical Research Council
CO Production in Hodskin’s Disease
I75
(Projects No. 19 X-3523 and 61P-3611), from the Medical Faculty, University of Lund, from John and Augusta Persson’s Foundation for Scientific Medical Research and from Lotten Bohman’s Foundation. REFERENCES BANERJEE, R.N. & NARANG, R.M. (1967) Haematological changes in malignancy. British Journal of Haematology, 13,829. BEAMISH,M.R., ASHLEYJONES,P., TREVETT,D., HOWELL EVANS,I. &JACOBS, A. (1972) Iron metabolism in Hodgkin’s discase. British Journal of Cancer, 26, 444. CELEGIN, M., HANSSON, R. & SUNDSTROM, G. (1971) A sub-microliter sampling device for quantitative collection of gases. Application to ‘gas-liquid chromatographic’ analysis. Scandinavian Journal of Clinical, and Laboratory Ifivestigation, 27, 367. CLINE,M.J. & BERLIN,N.I. (1963) Anemia in Hodgkin’s disease. Cancer, 16,526. COBURN, R.F., WILLIAMS, W.J. & KAHN,S.B. (1966) Endogenous carbon monoxide production in patients with hemolytic anemia. Journal of Clinical Investigation, 45, 460. COOK,J.D., MARSAGLIA, G., ESCHBACH, J.W., FUNK, D.D. & FINCH,C.A. (1970) Ferrokinetics: a biologic model for plasma iron exchange in man. Journal of Clinical Investigation, 49, 197. DEVITA,V.T., SERPICK, A.A. & CARBONE, P.P. (1970) Combination chemotherapy in the treatment of advanced Hodgkin’s disease. Anrials of Infernal Medicine, 73, 881. GAMBINO, S.R. & SCHREIBER, H. (1964) The measurcnient of bilirubin on the autoanalyzer by the method of Jendrassik and Grof. Technicon Symposion, paper 54. GIANNOPOULOS, P.P. & BERGSAGEL, D.E. (1959) The mechanism of the anemia associated with Hodgkin’s disease. Blood, 14,856. E.J. (1965) HAURANI, F.I., BURKE,W. & MARTINEZ, Defective reutilization of iron in the anemia of inflammation. Journal of Laboratory and Clinical Medicine, 6 5 560. HYMAN, G.A. &HARVEY, J.E. (1955) Thc pathogenesis of anaemia in patients with carcinoma. American Journal of Medicine, 19, 350. INTERNATIONAL COMMITTEE FOR STANDARDIZATION IN
C
HEMATOLOGY (1971) Recommended methods for radioisotopc red cell survival studies. Blood, 38, 378. LAURELL, C.-B. (1972) Electroimmuno assay. Scandinavian Jourual of Clinical and Laboratory Investigation, Suppl. 124.21. LUKES, R.J. & BUTLER, J.J. (1966) The pathology and nomenclature of Hodgkin’s disease. Cancer Research, 26, 1310. LUNDH,B.,CAVALLIN-STAHL, E. & MERCKE, C. (1975) Haem catabolism, carbon monoxide production and red cell survival in anaemia. A d a Medica Scandinavica, 197, 161. LUNDH, B., JOHANSSON, M.-B., MERCKE,C. & CAVALLIN-ST~L, E. (1972) Enhancement of hemecatabolism by caloric restriction in man. Scandinavian Journal of Clinical and Laboratory Investigation, 30, 421. RODKEY, F.L. & COLLISON, H.A. (1970) An artifact in the analysis of oxygenated blood for its low carbon monoxide content. Clinical Chemistry, 16,896. ON STAGING IN HODGKIN’S DISEASE (1971) SYMPOSIUM Cancer Research, 31, 1707. VAN KAMPEN, E.J. & ZIJLSTRA, W.G. (1965) Determination of hemoglobin and its derivatives. Advances in Clinical Chemistry, 8, 141. WEBB,D.I., UBOGY,G. & SILVER, R.T. (1970) Importance of bone marrow biopsy in the clinical staging of Hodgkin’s disease. Cancer, 26, 3 I 3. WHITE,P., COBURN, R.F., WILLIAMS, W.J., GOLDB.C. (1967) WEN,M.I., ROTHER,M.L. & SHAFER, Carbon monoxide production associated with ineffective erythropoiesis. Journal of Clinical Investigation, 46, 1986. YOUNG,D.S. & HICKS,J.M. (1965) Method for the automatic determination of serum iron. Journal of Clinical Pathology, 18, 98. ZUCKER,S., FRIEDMAN, S. & LYSIK, R.M. (1974) Bone marrow erythropoiesis in the anemia of infection, inflammation and malignancy. Journal of Clinical Investigation, 53, 1132.
