June 1978
930
The Journal o f P E D I A T R I C S
Anemia in patients with juvenile rheumatoid arthritis Patients with juvenile rheumatoid arthritis may have an anemia attributable to the chronic disease, to iron deficiency, or to a combination of the two, The contribution of iron deficiency is often difficult to determine by routine laboratory studies. We studied 51 patients with pauciarticular and polyarticular juvenile rheumatoid arthritis with red blood cell counts, indices, free erythrocyte protoporphyrin, and serum ferrit!n. Fifteen of the 18 who were anemic were restudied after a 3 to 6-month period of iron therapy. Thirteen of the 15 responded by these criteria: a rise in hemoglobin of l.O gm/dl or more and an increase in mean corpuscular volume of 3 f l or more; in 11 of these 13, hemoglobin values returned to the normal range for age. These findings indicate that iron deficiency can be a major component of the anemia that is commonly found in patients with active juvenile rheumatoid arthritis.
Marion A. Koerper, M.D.,* David A. Sternpel, M.D., and Peter R. Dallman, M . D . , S a n F r a n c i s c o a n d P a l o A l t o , Calif.
PATIENTS with juvenile rheumatoid arthritis may have anemia that is due to the chronic disease, to iron deficiency, or both. It is hard to recognize iron deficiency in the presence of chronic disease because some of the laboratory characteristics of iron deficiency may be masked by chronic disease whereas other abnormalities are the same in both disorders? in both conditions the serum iron concentration may be depressed, while the free erythrocyte protoporphyrin is elevated. 1 Microcytosis can also be present in chronic disease, although not as consistently as in iron deficiency. The total iron binding capacity may be elevated ,by iron deficiency and depressed by chronic disease, but often is normal whether or not the two conditions coexist. The serum ferritin level From the Department of Pediatrics, University of California, San Francisco, and the Rheumatic Disease Service, Children~ Hospital at Stanford, Department of Pediatrics, Stanford University. Supported by National Institute of Health Grants No. HD 00828 and A M 13897, National Foundation-March of Dimes Grant No. 6-8~, and Children's Hospital at Stanford. *Reprint address: Departni'entof Pediatrics, M650, University of California Medical Center, San Francisco, CA 94143.,
VoL 92, No. 6, pp. 930-933
might yield ambiguous results 2 because an elevation produced by chronic disease can obliterate the depression associated with iron deficiency. In adults with rheumatoid arthritis, serum ferritin levels in the lower third of the normal range were correlated with absent bone marrow iron stores whereas higher values were associated with normal iron stores? Abbreviations used MCV: mean corpuscular volume FEP: free erythrocyte protoporphyrin ESR: erythrocyte sedimentation rate Hgb: hemoglobin P: femtoliters ng: nanograms In order to ascertain the contribution of iron deficiency to the anemia of chronic disease, we studied 51 patients, 2 to 20 years of age, with juvenile rheumatoid arthritis. Fifteen who were anemic were restudied after a therapeutic trial of iron.
METHODS The 51 individuals who had been followed for at least six months had either pauciarticular or p01yarticular
0022-3476/78/0692-0930500.40/0 9 1978 The C. V. Mosby Co.
Volume 92 Number 6
Anemia in juvenile arthritis
t
+~
C
[o
~-4
/.
"
... :
4.~-
9
9
9
:
93 1
E
!
z
l
;
9
'2o
3~,
s "30
2~
BE FORE I RON
AFTER I RON
BE FORE IRON
A F TER I RON J
Fig. 1. Values for hemoglobin and mean corpuscular volume before and after iron therapy of three to six months' duration. Values are expressed in gm/dl above or below normal mean hemoglobin for age and femtoliters above or below normal mean MCV for age. Shaded area represents normal range for each.
disease, based on the number of joints involved at the t i m e of presentation according to the criteria of the American Rheumatism Association"; 22 had pauciarticular and 29 had polyarticular disease. At the time of study all patients were examined by a physician, and the state of the disease was judged either active or inactive based on the presence or absence of swollen or painful joints or an elevated erythrocyte sedimentation rate. All were receiving sufficient aspirin to maintain a blood level of 20 to 40 /,g/dl at the time of the study, and had been so treated for the preceding six months; none was on corticosteroids; two with active polyarticular disease were receiving intramuscular sodium aurothiomalate. None had received iron therapy in the previous three months. Fourteen patients with active polyarticular disease and one with active pauciarticular disease were restudied after a three-month trial of oral ferrous sulfate, 2 to 3 mg/kg, given once daily. Those individuals who still had subnormal hemoglobin values after three months of iron therapy were continued for an additional three months on the same dosage, Although this dosage is lower than that often used in pediatric practice, in adults it has been shown to produce a maximal rate of hemoglobin regeneration and reconstitution of iron stores after four months of treatment? We found our patients to be highly motivated and to be in compliance with aspirin therapy as monitored by blood levels. Consequently, we thought it sufficient to determine compliance to iron therapy by history. Informed consent was obtained in writing from all parents. All subjects had venous blood drawn for hemoglobin concentration and red blood cell indices using a Model S Coulter Counter. Erythrocyte sedimentation rate was
Table I. Normal values for hemoglobin, mean corpuscular volume, free erythrocyte protoporphyrin,* serum ferritin,t and erythrocyte sedimentation rate:~
Hemoglobin (gm/dl) Age (yr) 0.5-2 2-6 6-12 12-18 Female Male i8-49 Female Male
MCV (fl)
Mean
Lower(_2 sD)limit
Mean
Lower limit (--2 SD)
12.0 12.5 13.0
11.0 11.0 11.5
77 81 86
70 74 76
14.0 14.5
12.0 13.0
88 88
78 78
14.0 15.5
12.0 13,5
90 90
80 80
*Free erythrocyte protopbrphyfin ~< 3.0 t~g/gm hemoglobin. tSerum ferritin 10 to 200 ng/ml. :~Erythrocyte sedimentation rate _< 20 mm hr.
determined by the modified Westergren method? The FEP and serum ferritin levels were determined as previously described. ~ Normai values for each of these tests are taken from a recent review (Table I). 7 In the 6- to 12-year age group we substituted a hemoglobin value of 13.0 for 13.5 gm/dl because Our unpublished data have indicated that this is a more accurate value. Values for hemoglobin and MCV show an increase from 6 months to 18 years of age. In order to compare results in patients of different ages, we exPressed the values for hemoglobin and MCV in terms of grams and femtoliters above or below the normal mean for age. Normal values for ESR, FEP, and serum ferritin
932
Koerper, Stempel, and Dallman
200
The Journal of Pediatrics June 1978
$
100
c 5G
I
z" B
I'ee
na II.
9
_
_
_
|
m r
! RESPONSE TO IRON
NO RESPONSE
Fig. 2. Serum ferritin values prior to iron therapy in 11 patients with active juvenile rheumatoid arthritis and anemia who responded to oral iron therapy compared to 4 who did not. The dashed line represents the value for serum ferritin below which a response to iron therapy was uniformly obtained. Table 1I. Mean values before and after three to six months of oral ferrous sulfate in 14 patients with polyarticular and one with pauciarticular active juvenile rheumatoid arthritis and anemia Mean values before iron treatment Response [ No response No. of patients Hgb (gm below normal mean) MCV (fl below normal mean) FEP (~g/gm Hgb) Ferritin (ng/ml)
11 -3.4 -18 8.2 22
4 -3.3 -17 9.4 77
Mean values after iron treatment 15 --1.6" -12" 4.1"
22
*P < 0.0l compared to values for combined anemic groups before treatment. are not age dependent throughout childhood (Table I). A response to iron therapy was arbitrarily defined as an increase in Hgb of _> 1.0 gm/dl and an increase in MCV of___3 ft.
subjects were treated with" oral ferrous sulfate for three months, after which the studies were repeated in 15. Two patients (one with polyarticular and one with pauciarticular disease) had moved away and were lost to follow-up. One patient with polyarticular disease did not receive the iron and was not included in the analysis. Of the 15 patients who had received iron and were available for follow-up, 11 had responded by our criteria; however, two of them did not yet have hemoglobin values within the normal range. The four individuals who did not respond and the two who did not correct the anemia were continued on the same dose of iron for an additional three months, after which hemoglobin and MCV values were measured again. Four of the six showed continued improvement in hemoglobin and MCV values, and two of these four had values for Hgb within the normal range. Hemoglobin and MCV values in the 15 patients before and after three to six months of treatment are depicted in Fig. 1. The patients who were anemic also had elevated FEP levels which tended to return toward normal values with treatment. Almost all serum ferritin values were within the noi-mal range, and there was little change with therapy. Individual serum ferritin values prior to treatment are shown in Fig. 2 in patients who responded to iron therapy compared to those who did not. The six whose serum ferritin concentrations were below 25 ng/ml all responded tO iron therapy; five of the nine with serum ferritin values above 25 ng/ml also responded after three months and an additional two responded after six months of therapy. Mean values for Hgb, MCV, FEP, and serum ferritin in the 11 patients who responded to iron therapy are compared to the four who did not (Table II). Only the serum ferritin concentration distinguished the two groups initially. There was no significant difference in the values for average Hgb concentration, MCV, or FEP. The lack of significant difference in ESR suggested no difference in degree of activity of disease. Following iron therapy the mean values for Hgb, MCV, and FEP were all significantly closer to normal than before therapy. The mean serum ferritin value did not change appreciably, and neither did the mean ESR. In the nonanemic patients, values for MCV, FEP, and serum ferritin were within the normal range for age.
RESULTS The incidence of anemia varied widely among the groups. None of the nine patients with inactive disease, whether pauciarticular or polyarticular, was anemic. Of the 16 with active pauciarticular disease, two were anemic. The highest incidence of anemia occurred in the patients with active polyarticular disease'(16 of 26). The 18 anemic
DISCUSSION Anemia is frequently associated with rheumatoid arthritis; the more severe the disease, the more profound the anemia? The anemia may be secondary to gastrointestinal blood loss due to salicylate therapy. 9 Another explanation may be decreased absorption ~ or utilizationTM
Volume 92 Number 6
related to the underlying inflammatory disease. In this study, 13 of 15 anemic patients with active juvenile rheumatoid arthritis raised their hemoglobin concentration by at least 1.0 gm/dl, and 11 of the 13 corrected their anemia after three to six months of iron therapy. This indicates that in many individuals the anemia associated with juvenile rheumatoid arthritis and prolonged aspirin ingestion is at least partially due to correctible iron deficiency. A serum ferritin concentration below 25 n g / m l was useful in predicting a response to oral iron therapy, but a value above 25 n g / m l did not preclude a response. This value is close to the value of 20 n g / m l below which a group of adults with rheumatoid arthritis had no stainable iron in bone marrow. ~ A high incidence of iron deficiency in our patients with juvenile rheumatoid arthritis can be explained in two ways. First, large doses of aspirin may result in iron losses through occult intestinal bleeding. N o n e of the patients had melena; unfortunately, stool guaiacs for occult blood were not consistently checked until after the completion of this study. Conceivably, menstrual losses in females taking aspirin might also be greater than normal. A second possibility is a decrease in the absorption of iron that can be associated with fever, 11 infection, or inflammatory disease. 12 It is noteworthy that our patients, who might be expected to have such an absorption defect, did respond to oral iron therapy in modest doses. It is difficult to make the diagnosis of iron deficiency in the presence of juvenile rheumatoid arthritis. In the majority of tests of iron status, including Hgb, MCV, FEP, serum iron concentration, and transferrin saturation, the results can be interpreted as either iron deficiency or chronic anemia of juvenile rheumatoid arthritis. The total iron binding capacity and serum ferritin concentration may be useful in some cases but not in all. Even a larger battery of laboratory tests may not supply a conclusive answer. Consequently, a therapeutic trial of iron seems warranted in patients with a low hemoglobin concentra-
A n e m i a in juvenile arthritis
933
tion and low M C V for age. If there is no response after six months, it is unlikely that iron deficiency plays a role in the anemia. We thank John J. Miller III, M.D., Ph.D., for his help in obtaining the patient samples and reviewing the manuscript. We would also like to acknowledge the skilled technical assistance of Evelyn Manies and Robert Spirito. REFERENCES
1. Cartwright GE, and Lee GR: The anaemia of chronic disorders, Br J Haematol 21:147, 1971. 2. Bentley DP, and Williams P: Serum ferritin concentration as an index of storage iron in rheumatoid arthritis, J Clin Pathol 27:786, 1974. 3. Brewer E J, Bass J, Baum J, Cassidy JT, Fink C, Jacobs J, Hanson V, Levinson JE, Schaller J, and Stillman JS: Current proposed revision of JRA criteria, Arthritis Rheum 20:195, 1977. 4. Norrby A: Iron absorption studies in iron deficiency, Scand J Haematol (Suppl) 20:1, 1974. 5. Boroviczeny K-Gv, B6ttiger LE, Chattfis A, Dawson JB, Fukutake K, Gunz FW, Lewis SM, Rewald E, Roubicek M, Ruhenstroth-Bauer G, and Westergren A: Reference method for the erythrocyte sedimentation rate (ESR) test on human blood, J Clin Pathol 26:301, 1973. 6. Koerper MA, and Dallman PR: Serum iron concentration and transferrin saturation in the diagnosis of iron deficiency in children: Normal developmental changes, J PEDIATR 91:870, 1977. 7. Dallman PR: New approaches to screening for iron deficiency, J PEDIATR90:678, 1977. 8. Jeffrey MR: Some observations on anemia in rheumatoid arthritis, Blood 8:502, 1953. 9. Roberts FD, Hagedorn AB, Slocumb CH, and Owen CA: Evaluation of the anemia of rheumatoid arthritis, Blood 21:470, 1963. 10. Weinstein IM: A correlative study of the erythrokinetics and disturbances in iron metabolism associated with the anemia of rheumatoid arthritis, Blood 14:950, 1959. 11. Beresford CH, Neale R J, and Brooks OG: Iron absorption and pyrexia, Lancet 1:568, 1971. 12. Raymond FD, Bowie MA, and Dugan A: Iron metabolism in rheumatoid arthritis, Arthritis Rheum 8:233, 1965.
930
The Journal o f P E D I A T R I C S
Anemia in patients with juvenile rheumatoid arthritis Patients with juvenile rheumatoid arthritis may have an anemia attributable to the chronic disease, to iron deficiency, or to a combination of the two, The contribution of iron deficiency is often difficult to determine by routine laboratory studies. We studied 51 patients with pauciarticular and polyarticular juvenile rheumatoid arthritis with red blood cell counts, indices, free erythrocyte protoporphyrin, and serum ferrit!n. Fifteen of the 18 who were anemic were restudied after a 3 to 6-month period of iron therapy. Thirteen of the 15 responded by these criteria: a rise in hemoglobin of l.O gm/dl or more and an increase in mean corpuscular volume of 3 f l or more; in 11 of these 13, hemoglobin values returned to the normal range for age. These findings indicate that iron deficiency can be a major component of the anemia that is commonly found in patients with active juvenile rheumatoid arthritis.
Marion A. Koerper, M.D.,* David A. Sternpel, M.D., and Peter R. Dallman, M . D . , S a n F r a n c i s c o a n d P a l o A l t o , Calif.
PATIENTS with juvenile rheumatoid arthritis may have anemia that is due to the chronic disease, to iron deficiency, or both. It is hard to recognize iron deficiency in the presence of chronic disease because some of the laboratory characteristics of iron deficiency may be masked by chronic disease whereas other abnormalities are the same in both disorders? in both conditions the serum iron concentration may be depressed, while the free erythrocyte protoporphyrin is elevated. 1 Microcytosis can also be present in chronic disease, although not as consistently as in iron deficiency. The total iron binding capacity may be elevated ,by iron deficiency and depressed by chronic disease, but often is normal whether or not the two conditions coexist. The serum ferritin level From the Department of Pediatrics, University of California, San Francisco, and the Rheumatic Disease Service, Children~ Hospital at Stanford, Department of Pediatrics, Stanford University. Supported by National Institute of Health Grants No. HD 00828 and A M 13897, National Foundation-March of Dimes Grant No. 6-8~, and Children's Hospital at Stanford. *Reprint address: Departni'entof Pediatrics, M650, University of California Medical Center, San Francisco, CA 94143.,
VoL 92, No. 6, pp. 930-933
might yield ambiguous results 2 because an elevation produced by chronic disease can obliterate the depression associated with iron deficiency. In adults with rheumatoid arthritis, serum ferritin levels in the lower third of the normal range were correlated with absent bone marrow iron stores whereas higher values were associated with normal iron stores? Abbreviations used MCV: mean corpuscular volume FEP: free erythrocyte protoporphyrin ESR: erythrocyte sedimentation rate Hgb: hemoglobin P: femtoliters ng: nanograms In order to ascertain the contribution of iron deficiency to the anemia of chronic disease, we studied 51 patients, 2 to 20 years of age, with juvenile rheumatoid arthritis. Fifteen who were anemic were restudied after a therapeutic trial of iron.
METHODS The 51 individuals who had been followed for at least six months had either pauciarticular or p01yarticular
0022-3476/78/0692-0930500.40/0 9 1978 The C. V. Mosby Co.
Volume 92 Number 6
Anemia in juvenile arthritis
t
+~
C
[o
~-4
/.
"
... :
4.~-
9
9
9
:
93 1
E
!
z
l
;
9
'2o
3~,
s "30
2~
BE FORE I RON
AFTER I RON
BE FORE IRON
A F TER I RON J
Fig. 1. Values for hemoglobin and mean corpuscular volume before and after iron therapy of three to six months' duration. Values are expressed in gm/dl above or below normal mean hemoglobin for age and femtoliters above or below normal mean MCV for age. Shaded area represents normal range for each.
disease, based on the number of joints involved at the t i m e of presentation according to the criteria of the American Rheumatism Association"; 22 had pauciarticular and 29 had polyarticular disease. At the time of study all patients were examined by a physician, and the state of the disease was judged either active or inactive based on the presence or absence of swollen or painful joints or an elevated erythrocyte sedimentation rate. All were receiving sufficient aspirin to maintain a blood level of 20 to 40 /,g/dl at the time of the study, and had been so treated for the preceding six months; none was on corticosteroids; two with active polyarticular disease were receiving intramuscular sodium aurothiomalate. None had received iron therapy in the previous three months. Fourteen patients with active polyarticular disease and one with active pauciarticular disease were restudied after a three-month trial of oral ferrous sulfate, 2 to 3 mg/kg, given once daily. Those individuals who still had subnormal hemoglobin values after three months of iron therapy were continued for an additional three months on the same dosage, Although this dosage is lower than that often used in pediatric practice, in adults it has been shown to produce a maximal rate of hemoglobin regeneration and reconstitution of iron stores after four months of treatment? We found our patients to be highly motivated and to be in compliance with aspirin therapy as monitored by blood levels. Consequently, we thought it sufficient to determine compliance to iron therapy by history. Informed consent was obtained in writing from all parents. All subjects had venous blood drawn for hemoglobin concentration and red blood cell indices using a Model S Coulter Counter. Erythrocyte sedimentation rate was
Table I. Normal values for hemoglobin, mean corpuscular volume, free erythrocyte protoporphyrin,* serum ferritin,t and erythrocyte sedimentation rate:~
Hemoglobin (gm/dl) Age (yr) 0.5-2 2-6 6-12 12-18 Female Male i8-49 Female Male
MCV (fl)
Mean
Lower(_2 sD)limit
Mean
Lower limit (--2 SD)
12.0 12.5 13.0
11.0 11.0 11.5
77 81 86
70 74 76
14.0 14.5
12.0 13.0
88 88
78 78
14.0 15.5
12.0 13,5
90 90
80 80
*Free erythrocyte protopbrphyfin ~< 3.0 t~g/gm hemoglobin. tSerum ferritin 10 to 200 ng/ml. :~Erythrocyte sedimentation rate _< 20 mm hr.
determined by the modified Westergren method? The FEP and serum ferritin levels were determined as previously described. ~ Normai values for each of these tests are taken from a recent review (Table I). 7 In the 6- to 12-year age group we substituted a hemoglobin value of 13.0 for 13.5 gm/dl because Our unpublished data have indicated that this is a more accurate value. Values for hemoglobin and MCV show an increase from 6 months to 18 years of age. In order to compare results in patients of different ages, we exPressed the values for hemoglobin and MCV in terms of grams and femtoliters above or below the normal mean for age. Normal values for ESR, FEP, and serum ferritin
932
Koerper, Stempel, and Dallman
200
The Journal of Pediatrics June 1978
$
100
c 5G
I
z" B
I'ee
na II.
9
_
_
_
|
m r
! RESPONSE TO IRON
NO RESPONSE
Fig. 2. Serum ferritin values prior to iron therapy in 11 patients with active juvenile rheumatoid arthritis and anemia who responded to oral iron therapy compared to 4 who did not. The dashed line represents the value for serum ferritin below which a response to iron therapy was uniformly obtained. Table 1I. Mean values before and after three to six months of oral ferrous sulfate in 14 patients with polyarticular and one with pauciarticular active juvenile rheumatoid arthritis and anemia Mean values before iron treatment Response [ No response No. of patients Hgb (gm below normal mean) MCV (fl below normal mean) FEP (~g/gm Hgb) Ferritin (ng/ml)
11 -3.4 -18 8.2 22
4 -3.3 -17 9.4 77
Mean values after iron treatment 15 --1.6" -12" 4.1"
22
*P < 0.0l compared to values for combined anemic groups before treatment. are not age dependent throughout childhood (Table I). A response to iron therapy was arbitrarily defined as an increase in Hgb of _> 1.0 gm/dl and an increase in MCV of___3 ft.
subjects were treated with" oral ferrous sulfate for three months, after which the studies were repeated in 15. Two patients (one with polyarticular and one with pauciarticular disease) had moved away and were lost to follow-up. One patient with polyarticular disease did not receive the iron and was not included in the analysis. Of the 15 patients who had received iron and were available for follow-up, 11 had responded by our criteria; however, two of them did not yet have hemoglobin values within the normal range. The four individuals who did not respond and the two who did not correct the anemia were continued on the same dose of iron for an additional three months, after which hemoglobin and MCV values were measured again. Four of the six showed continued improvement in hemoglobin and MCV values, and two of these four had values for Hgb within the normal range. Hemoglobin and MCV values in the 15 patients before and after three to six months of treatment are depicted in Fig. 1. The patients who were anemic also had elevated FEP levels which tended to return toward normal values with treatment. Almost all serum ferritin values were within the noi-mal range, and there was little change with therapy. Individual serum ferritin values prior to treatment are shown in Fig. 2 in patients who responded to iron therapy compared to those who did not. The six whose serum ferritin concentrations were below 25 ng/ml all responded tO iron therapy; five of the nine with serum ferritin values above 25 ng/ml also responded after three months and an additional two responded after six months of therapy. Mean values for Hgb, MCV, FEP, and serum ferritin in the 11 patients who responded to iron therapy are compared to the four who did not (Table II). Only the serum ferritin concentration distinguished the two groups initially. There was no significant difference in the values for average Hgb concentration, MCV, or FEP. The lack of significant difference in ESR suggested no difference in degree of activity of disease. Following iron therapy the mean values for Hgb, MCV, and FEP were all significantly closer to normal than before therapy. The mean serum ferritin value did not change appreciably, and neither did the mean ESR. In the nonanemic patients, values for MCV, FEP, and serum ferritin were within the normal range for age.
RESULTS The incidence of anemia varied widely among the groups. None of the nine patients with inactive disease, whether pauciarticular or polyarticular, was anemic. Of the 16 with active pauciarticular disease, two were anemic. The highest incidence of anemia occurred in the patients with active polyarticular disease'(16 of 26). The 18 anemic
DISCUSSION Anemia is frequently associated with rheumatoid arthritis; the more severe the disease, the more profound the anemia? The anemia may be secondary to gastrointestinal blood loss due to salicylate therapy. 9 Another explanation may be decreased absorption ~ or utilizationTM
Volume 92 Number 6
related to the underlying inflammatory disease. In this study, 13 of 15 anemic patients with active juvenile rheumatoid arthritis raised their hemoglobin concentration by at least 1.0 gm/dl, and 11 of the 13 corrected their anemia after three to six months of iron therapy. This indicates that in many individuals the anemia associated with juvenile rheumatoid arthritis and prolonged aspirin ingestion is at least partially due to correctible iron deficiency. A serum ferritin concentration below 25 n g / m l was useful in predicting a response to oral iron therapy, but a value above 25 n g / m l did not preclude a response. This value is close to the value of 20 n g / m l below which a group of adults with rheumatoid arthritis had no stainable iron in bone marrow. ~ A high incidence of iron deficiency in our patients with juvenile rheumatoid arthritis can be explained in two ways. First, large doses of aspirin may result in iron losses through occult intestinal bleeding. N o n e of the patients had melena; unfortunately, stool guaiacs for occult blood were not consistently checked until after the completion of this study. Conceivably, menstrual losses in females taking aspirin might also be greater than normal. A second possibility is a decrease in the absorption of iron that can be associated with fever, 11 infection, or inflammatory disease. 12 It is noteworthy that our patients, who might be expected to have such an absorption defect, did respond to oral iron therapy in modest doses. It is difficult to make the diagnosis of iron deficiency in the presence of juvenile rheumatoid arthritis. In the majority of tests of iron status, including Hgb, MCV, FEP, serum iron concentration, and transferrin saturation, the results can be interpreted as either iron deficiency or chronic anemia of juvenile rheumatoid arthritis. The total iron binding capacity and serum ferritin concentration may be useful in some cases but not in all. Even a larger battery of laboratory tests may not supply a conclusive answer. Consequently, a therapeutic trial of iron seems warranted in patients with a low hemoglobin concentra-
A n e m i a in juvenile arthritis
933
tion and low M C V for age. If there is no response after six months, it is unlikely that iron deficiency plays a role in the anemia. We thank John J. Miller III, M.D., Ph.D., for his help in obtaining the patient samples and reviewing the manuscript. We would also like to acknowledge the skilled technical assistance of Evelyn Manies and Robert Spirito. REFERENCES
1. Cartwright GE, and Lee GR: The anaemia of chronic disorders, Br J Haematol 21:147, 1971. 2. Bentley DP, and Williams P: Serum ferritin concentration as an index of storage iron in rheumatoid arthritis, J Clin Pathol 27:786, 1974. 3. Brewer E J, Bass J, Baum J, Cassidy JT, Fink C, Jacobs J, Hanson V, Levinson JE, Schaller J, and Stillman JS: Current proposed revision of JRA criteria, Arthritis Rheum 20:195, 1977. 4. Norrby A: Iron absorption studies in iron deficiency, Scand J Haematol (Suppl) 20:1, 1974. 5. Boroviczeny K-Gv, B6ttiger LE, Chattfis A, Dawson JB, Fukutake K, Gunz FW, Lewis SM, Rewald E, Roubicek M, Ruhenstroth-Bauer G, and Westergren A: Reference method for the erythrocyte sedimentation rate (ESR) test on human blood, J Clin Pathol 26:301, 1973. 6. Koerper MA, and Dallman PR: Serum iron concentration and transferrin saturation in the diagnosis of iron deficiency in children: Normal developmental changes, J PEDIATR 91:870, 1977. 7. Dallman PR: New approaches to screening for iron deficiency, J PEDIATR90:678, 1977. 8. Jeffrey MR: Some observations on anemia in rheumatoid arthritis, Blood 8:502, 1953. 9. Roberts FD, Hagedorn AB, Slocumb CH, and Owen CA: Evaluation of the anemia of rheumatoid arthritis, Blood 21:470, 1963. 10. Weinstein IM: A correlative study of the erythrokinetics and disturbances in iron metabolism associated with the anemia of rheumatoid arthritis, Blood 14:950, 1959. 11. Beresford CH, Neale R J, and Brooks OG: Iron absorption and pyrexia, Lancet 1:568, 1971. 12. Raymond FD, Bowie MA, and Dugan A: Iron metabolism in rheumatoid arthritis, Arthritis Rheum 8:233, 1965.
