Erythrocyte Volume Distribution Analysis and Hematologic Changes in Dogs with Iron Deficiency Anemia

Transcription

1 Vel. Pathol. 20: (1983) Erythrocyte Volume Distribution Analysis and Hematologic Changes in Dogs with Iron Deficiency Anemia G. WEISER and M. O'GRADY Department of Veterinary Pathobiology, and Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University, Columbus, Ohio Abstract. Hematologic features were characterized in 12 dogs with iron deficiency anemia attributable to chronic external blood loss. Consistent abnormalities in hemograms included moderate to marked reticulocytosis, decreased mean corpuscular volume, and decreased mean corpuscular hemoglobin concentration. Hypoproteinemia occurred in only four of 12 dogs. Consistent blood film findings included hypochromic cel1s, leptocytosis, and erythrocyte fragmentation. These dogs had significantly decreased serum iron values (p < 0.00I) and percent transferrin saturation values (p < 0.00I) compared with 33 clinically healthy adult dogs. The total iron binding capacity values of these dogs were not significantly different (p > 0.5) than those of the healthy dogs. Using erythrocyte volume distribution curves, the percentages ofmicrocytic cells (:5 45 fl) were determined to range from 20% to 82%. Sequential changes in erythrocyte subpopulations were evaluated in four dogs which received iron therapy. The hematologic response consisted of fairly rapid restoration of packed cell volume by production of normocytes followed by a more gradual replacement of residual microcytes by new normocytes. Iron deficiency anemia in dogs is regarded as a rare condition associated with blood sucking parasites or other causes of chronic external blood loss [17]. Hematologic features include anemia, microcytosis, hypochromic erythrocytes, and poikilocytosis [17]. The anemia generally is regarded as hypo proliferative or non-regenerative [II]. While recent sources indicate that decreased serum iron and increased total iron binding capacity are features of iron deficiency [II, 13], there are no reports of these values in iron deficient dogs. The increased use of automated cell counting systems, which provide accurate erythrocytic indices [10], may result in more frequent recognition of dogs with iron deficiency anemia. Furthermore, erythrocyte volume distribution curves (erythrograms) are more sensitive than mean corpuscular volume values in detecting disturbances of erythrocyte size [1-3, 15]. Erythrograms were used in this study to quantitate the degree of microcytosis and to estimate sequential changes in numbers of erythrocytes within microcytic and normocytic subpopulations following iron 230

2 Canine Iron Deficiency Anemia 231 supplementation. In addition, serum iron values and hematologic changes were characterized in dogs with iron deficiency anemia. Materials and Methods This study included 12 dogs hospitalized at the Ohio State University Veterinary Teaching Hospital during a one-year period. The dogs were selected on the basis of having anemia, microcytosis, hypochromic erythrocytes on the blood film, and clinical evidence of prolonged external blood loss. Four of the dogs (dogs 8-11) had been used as blood donors. Dogs 9 and 10 were bled at a rate of 20 mlkgweek over a five-week period. Dogs 8 and II were bled at an average rate of 10 nilkgweek over a 12- and 18-month period, respectively. At least one hemogram, erythrogram, serum iron determination, total iron binding capacity determination, and percent transferrin saturation calculation were done on each dog prior to any treatment. Hemograms were done on a multi-channel, automated blood cell counting system (Coulter Counter Model S-Senior, Coulter Electronics, Hialeah, Fla.). Wright-Giemsa-stained blood films were prepared for evaluation of erythrocyte morphology, and new methylene bluestained blood films were prepared for reticulocyte counts [23]. Erythrograms were done as previously described for adult dogs [24]. Serum iron, total iron binding capacity, and percent transferrin saturation values were determined by the ferrozine spectrophotometric method [21] (Ferro-Chek II kit, Hyland Diagnostics, Deerfield, Ill.). The procedure was modified by reducing all sample and reagent volumes to! those suggested by the manufacturer. A, commercial control serum (Validate, General Diagnostics, Morris Plains, N.J.) was included in each analysis. The iron values also were determined on 33 apparently healthy adult dogs. Using a logarithmic transformation, reference values for serum iron and percent transferrin saturation were established from the mean ± 2 standard deviations (14]. Reference values for total iron binding capacity were established from the mean ± 2 standard deviations. The Student's r-test was used to determine if significant differences in iron values existed between normal dogs and dogs with iron deficiency anemia. By integration of erythrograms, the percentage oferythrocytes having volume less than or equal to 45 femtoliters (11) (microcytes) and erythrocytes having greater than 4511 (normocytes) were calculated for each dog. The percentage of microcytes in apparently healthy adult dogs was determined from previously reported erythrograms [24]. Serial hemograms and erythrograms were done on four dogs (dogs 9-12) receiving iron therapy. Dogs 9 through II were treated with about 30 mgkg of intramuscular iron dextran once weekly for three weeks (Ferrextran, Fort Dodge Laboratories, Fort Dodge, Iowa). Thereafter, they each received about 20 mgkg daily of oral ferrous gluconate (Fergon, Breon Laboratories, Inc., New York, N.Y.). These three dogs (blood donors) were not used as donors for about 60 days after starting iron therapy. After this time about 20 mlkg of blood was removed from each dog every four to six weeks. Dog 12 received about 60 mgkg of ferrous gluconate daily for the duration of the observation period. The beginning ofiron therapy was established as day 0 for comparison of data collected on subsequent days. Hemograms and erythrograms were done at various intervals for 32, 203, 203, and 280 days in dogs 12, 9, 10, and II respectively. The absolute numbers ofrnicrocytesul were determined on each sampling day by multiplying the percentage of each cell type on erythrograms by the respective erythrocyte count. Results Pertinent hematologic data are presented in table I. Eleven of the dogs had mild to severe anemia, and dog 9 had a packed cell volume in the normal range. The II anemic dogs had moderate or marked reticulocytosis. All dogs had low mean

3 tv Wtv Table I. Hematologic data for 12 dogs with iron deficiency anemia Dog Age Sex Breed Cause of iron deficiency PCV Reticulo- MCV % Micro- MCHC cytespl (fl) cytes (gmdl) I 5 mo. M German shepherd Severe hookworm infestation , yr. M German shepherd Hyperadrenocorticism with II 214, severe hookworm infestation 3 3 yr. F Mixed breed Large hemorrhagic transmis , sible venereal tumor 4 II yr. M Mixed breed Intestinal leiomyosarcoma , yr. M Doberman Gastrointestinal lymphosar , ".., Plasma protein (gmdl) coma with chronic hemor- I" ::s c, rhagic diarrhea yr. M Mixed breed Intestinal lymphosarcoma , d with hemorrhage 7 6 yr. F Mixed breed Hepatic cirrhosis with gas , '< trointestinal hemorrhage 8 Adult F Mixed breed Excessive use as blood donor , yr. M Mixed breed Excessive use as blood donor 41 ND Adult F Boxer Excessive use as blood donor , II Adult M Mixed breed Excessive use as blood donor , mo. M Boxer Undetermined, fecal negative , Reference values' <60, pev = Packed cell volume; MCV = Mean corpuscular volume; MCHC = Mean corpuscular hemoglobin concentration; ND = Not determined. I Values established at O.S.U. Veterinary Hospital Hematology Laboratory.., I" P.

4 Canine Iron Deficiency Anemia 233 Table II. Serum iron, total iron binding capacity, and percent transferrin saturation values for normal dogs and dogs with iron deficiency anemia Normal dogs n = 33 Serum iron (ugdl) TlBC (J.tgdl) % Tr Sat ean Standard deviation Reference value range Dogs with microcytosis n = 12 Mean Standard deviation Observed range TlBC = total iron binding capacity; % Tr Sat = percent transferrin saturation. corpuscular volume values and an increased percentage of microcytic cells on erythrograms. Low mean corpuscular hemoglobin concentration values «34.4 gm dl) were observed in 11 of 12 dogs. Hypoproteinemia was present in only four dogs. Values for serum iron, total iron binding capacity, and percent transferrin saturation for normal dogs and dogs with iron deficiency anemia are presented in table II. Significantly lower (p < 0.00I) serum iron and percent transferrin saturation values occurred in dogs with iron deficiency anemia. The serum iron value was decreased in 11 of 12 dogs and was low normal in dog 8. The percent transferrin saturation value was decreased in 10 of 12dogs and was low normal in dogs I and 6. Total iron binding capacity values for the two groups of dogs were not significantly different (p > 0.5). Consistent morphologic features on blood films included increased polychromasia, hypochromic erythrocytes, leptocytosis, and fragmented erythrocytes (fig. 1). While the degree of these alterations varied amongst the dogs, they were observed easily on all blood films. Dogs 9 and 10 had a small proportion of hypochromic cells. Erythrocyte fragmentation appeared to be most prominent in dogs with the most severe microcytosis. Two types of erythrograms were observed in dogs with iron deficiency anemia (fig. 2). One type, observed in eight of 12 dogs, represented complete blood repopulation with a relatively homogeneous population of microcytes distinctly separate from the normocytic size range. The other type, observed in the remaining four dogs (dogs 2, 8-10), represented a more heterogeneous population overlapping with the normal volume range. Representative changes in serial erythrograms following iron therapy are indicated in fig. 3. Serial changes in erythrograms of dogs 11 and 12 showed simultaneous disappearance of the microcyte population and formation of a separate normocyte population. In dogs 9 and 10, two separate populations were not observed. Instead, an initial heterogeneous population gradually shifted toward larger size and became more homogeneous.

5 234 Weiser and O'Grady Fig. I: Erythrocyte morphologic abnormalities in dogs with iron deficiency anemia. Hypochromic cells (large arrows) and various defects interpreted as fragmentation (small arrows). Bar = 10 ILm. The rates of net microcyte disappearance and normocyte formation during iron therapy in dogs 9 through 12 are indicated in fig. 4. Dog 12 showed the most rapid rate of recovery although this dog was monitored for only 32 days. In dogs 9 through 11, the numbers of microcytes increased during parenteral iron therapy (first 2-3

6 Canine Iron Deficiency Anemia f.., II I I\ DOG I \\ o z w o W0::: I.J \ \\ \ W > <t -J W 0::: f.., J I J I DOG II I I I I I \ J I I \ I I I \ I \ I \ I \ \ \ ""\,....."..-- "' CELL VOLUME (fl) Fig. 2: Two representative erythrograms ofdogs with iron deficiency anemia. Erythrogram from a normal dog with a mean corpuscular volume of 64 fl is indicated by broken line. weeks) and then gradually declined thereafter. Dogs 10 and II continued to have elevated numbers of microcytes after 200 or more days. All four dogs had a gradual increase in numbers of normocytes. Serial changes in packed cell volume and mean corpuscular volume are presented in fig. 5. The packed cell volume returned to the normal range within 25 to 35 days and appeared to plateau at about 50 days in dogs 10 through 12. Dog 9 had a similar response but initially had a packed cell volume in the normal range. Mean corpuscular volume values returned toward normal much more slowly, taking between 80 and 200 days to reach low normal in dogs 9 through II. Dog 12 had not achieved a normal mean corpuscular volume by the end of the 32-day observation period. Discussion Chronic external blood loss appeared to be the mechanism resulting in the development of iron deficiency anemia in II of 12 dogs. The cause of anemia could

7 236 Weiser and O'Grady DAY > (.) z 80 DAY ạ ;:)... a:: 60 I<.. \ \... \ > 40 \ \ Cl: \...J... \ a:: 20 ae " " "... " CELL VOL (FL) Fig,3: Serial erythrograms during iron therapy, dog 12. Day 0 erythrogram is indicated by broken line. not be established in dog 12. The variability in degree of anemia can not be explained fully. It may be related to the average daily rate of blood loss. The demand on erythropoiesis imposed by growth in young dogs may contribute to severity of anemia. The moderate to marked reticulocytosis was interpreted as evidence of a functional bone marrow. This is in contrast to the statement that iron deficiency

8 Canine Iron Deficiency Anemia en w I 7200 o > U o a: :I: I > a: LIJ v 3 \ I I DOG 10,, <f',, DOG TIME - DAYS Fig. 4: Serial changes in numbers of normocytes (e e) and microcytes (e--- e) in four dogs receiving iron therapy. anemia usually is poorly regenerative [11]. While we have experienced non-regenerative iron deficiency anemia in puppies with severe hookworm infestation, dogs in this study showed that a regenerative response to blood loss usually can be maintained. Hypoproteinemia occurred in only four of the dogs. It appeared that most of the dogs were able to replace lost plasma proteins. Since the severity of hypoproteinemia did not correlate with the severity ofanemia, it is likely that factors other than blood loss contributed to hypoproteinemia. For example, dog 7 had cirrhosis, severe hypoproteinemia, and only mild anemia. Dogs 1,5, and 6 may have had generalized malabsorption, associated with their gastrointestinal disease, contributing to hypoproteinemia. Dog 3, with the transmissible venereal tumor, and the blood donors were able to maintain plasma proteins in the mid-normal range. Low mean corpuscular hemoglobin concentration values were present in 11 of 12 dogs when compared to reference values determined on the automated cell counting system. Low values were attributed to a combination of iron deficiency and reticu-

9 238 Weiser and O'Grady > ua > o " '---0--_.-0-'-"-; " ' -' 0---<) Dog 9 [)-.-(] Dog Dog II --. Dog TIME (DAYS) Fig. 5: Serial changes in packed cell volume and mean corpuscular volume in four dogs receiving iron therapy. locytosis. The most widely accepted canine mean corpuscular hemoglobin concentration reference values of 32.0 to 36.0 gmdl were derived from the microhematocrit method (17]. If these values were used, only four of 12dogs in this study would have low mean corpuscular hemoglobin concentration values. In our experience, mean corpuscular hemoglobin concentration values calculated from microhematocrit data tend to be lower than those determined on the automated counting system. Furthermore, in humans with iron deficiency anemia there is considerable plasma trapping in spun hematocrits resulting in spuriously low mean corpuscular hemoglobin concentration values [9]. In the dog, mean corpuscular hemoglobin concentration values must be interpreted with respect to reference values determined for the method or system being used. Microcytosis was a consistent finding in these dogs. In fact, a low mean corpuscular volume value usually initiated the clinician'sconsideration of iron deficiency anemia. As defined in this study, the percentage of microcytic cells on erythrograms was altered more clearly than mean corpuscular volume values. For example, dogs 2, 8, and 9 had borderline low mean corpuscular volume values but had a large proportion of microcytes compared with normal dogs. Thus the erythrogram could be expected to detect the development of microcytosis before the mean corpuscular volume would necessarily be abnormally low. Since counting systems which measure cell volume

10 Canine Iron Deficiency Anemia 239 directly are not influenced by plasma trapping or erythrocyte concentration, they detect microcytosis more accurately [9, 10]. The increased use ofthis instrumentation may result in more frequent recognition of iron deficiency anemia in dogs. Erythrograms provided insight into the hematologic responses associated with iron deficiency anemia. The two types of erythrograms were interpreted to represent different stages ofthe disease. The heterogeneous curve including the normal volume range was considered to reflect relatively early development of microcytosis. Iron limited erythropoiesis was resulting in production of microcytes but normal cells were still present. The bleeding history in dogs 9 and 10 supports this explanation. The homogeneous curve consisting of predominantly microcytes indicated that iron limited erythropoiesis had occurred long enough to replace the normocyte population. It is likely that reticulocytes in these dogs were smaller than normal. It has been shown in a variety ofspecies that reticulocytes released in response to acute blood loss or hemolysis are large cells [4, 5, 12, 18-20,22,25]. After some initial size reduction, these cells remain large for a considerable period of time [18-20, 22, 25]. In our dogs however, no populations of large cells corresponding to the proportion of reticulocytes were observed on the erythrograms. This indicates that reticulocytes are small in iron deficiency anemia. This is in agreement with more direct measurement of reticulocyte size in iron deficiency anemia in man [7]. Serial erythrograms provided a means of assessing changes in erythrocyte subpopulations during iron therapy. The hematologic response to iron involved a fairly rapid production of normocytic cells to restore red cell mass followed by a more gradual replacement ofmicrocytes with normocytes. The serial changes in erythrocyte subpopulations are very similar to those observed during recovery in iron deficiency anemia in man [I]. The microcyte disappearance rate would depend on their survival time as well as any continued production. In dogs 9 through II microcytosis persisted considerably longer than a normal erythrocyte lifespan, suggesting that there was continued microcyte production. This was likely due to the blood donation resumed in these dogs after day 60. Bleeding resulted in an estimated loss of about 200 mg of iron every four to six weeks. From this it would appear that if the cause of blood loss is not corrected, microcytosis may persist indefinitely in the face of iron supplementation. It is also noteworthy that the numbers of microcytes continued to increase in these dogs after they were given injectable iron. Numbers of microcytes began to decline in dogs 9 through 12 when given oral iron. This observation suggests that a controlled trial comparing the efficacy of oral and injectable iron would be useful. Findings on stained blood films were consistent with previous descriptions [17]. We interpreted the previously described poikilocytosis [17] as erythrocyte fragmentation. Since microcytic cells in man with iron deficiency anemia have an altered thickness to diameter ratio [6], they may be more susceptible to normal intravascular trauma. Oxidation of the inner erythrocyte membrane may contribute to fragmentation. Decreased levels ofglutathione peroxidase occur in rabbits with iron deficiency anemia [16]. Localized membrane oxidation and associated increased membrane

11 240 Weiser and O'Grady rigidity could result in the observed fragmentation. Consistent with this hypothesis is a recent report that peroxidation of erythrocyte membrane components was associated with decreased erythrocyte deform ability in iron deficient rats and man (27). Similar prominent fragmentation has been reported in kittens with iron deficiency anemia (26). Reference iron values obtained in this study were quite comparable to those in man [8]. In man, percent transferrin saturation values below 20% are compatible with iron deficiency anemia. In dogs, a transferrin saturation of 15% or below would be compatible with iron deficiency anemia and values between 15 and 20% would be suggestive of the same. Iron values alone would not be helpful for distinguishing iron deficiency anemia from anemia ofinflammatory disease in dogs. Total iron binding capacity values are reported to be increased in man [8] and dogs [II, 13] with iron deficiency anemia. In contrast, total iron binding capacity values were normal in these dogs. References BESSMAN, D.: Erythropoiesis during recovery from iron deficiency: normocytes and macrocytes. Blood 50: , BESSMAN, D.: Red cell volume heterogeneity. Johns Hopkins Med J 146: , BESSMAN, J.D.; JOHNSON, R.K.: Erythrocyte volume distribution in normal and abnormal subjects. Blood 46: , BRECHER, G.; PRENANT, M.; HALEY, J.; BESSIS, M.: Origin ofstress macroreticulocytes from macronormoblasts. Nouv Rev Fr Hematol15: 13-18, BRECHER, G.; STOHLMAN, F.: Reticulocyte size and erythropoietic stimulation. Proc Soc Exp Bioi Med 107: , CHILD, J.A.; BOWRY, W.M.; KNOWLES, J.P.: Abnormality of red-cell diameterthickness ratio: Findings in iron-deficiency anemia. Br J Haematoll9: , CLARKSON, D.R.; MORE, E.M.: Reticulocyte size in nutritional anemias. Blood 48: , DAVIDSOHN, I.; HENRY, J.B.: Clinical Diagnosis by Laboratory Methods, pp , 1379, 15th ed. W. B. Saunders, Philadelphia, ENGLAND, J.M.; WALFORD, D.M.; WATERS, D.A.W.: Re-assessment ofthe reliability ofthe haematocrit. Br J Haematol 23: , FAIRBANKS, V.F.: Nonequivalence ofautomated and manual hematocrit and erythrocytic indices. Am J Clio Pathoi 73:55-62, 1980 II FELDMAN, B.F.: Hypoproliferative anemias and anemias caused by ineffective erythropoiesis. In: Veterinary Clinics of North America, ed. Jain and Zinkl, vol. II, pp W. B. Saunders, Philadelphia, HILLMAN, R.S.: Characteristics of marrow production and reticulocyte maturation in normal man in response to anemia. J Clin Invest 48: , KANEKO, J.1.: Clinical Biochemistry of Domestic Animals, p. 665, 3rd ed. Academic Press, New York, LUMSDEN, J.H.; MULLEN, K.: On establishing reference values. Can J Comp Med 42: , PROCTOR, S.J.; Cox, J.R.; SHERIDAN, T.J.: Anisocytosis and the C-I000 channelyzer in macrocytic anemia. J Clin Pathoi 29: , RODVIEN, R.; GILLUM, A.; WEINTRAUB, L.R.: Decreased glutathione peroxidase activity

12 Canine Iron Deficiency Anemia 241 secondary to severe iron deficiency: A possible mechanism responsible for the shortened life span of the iron-deficient red cell. Blood 43: , SCHALM, O.W.; JAIN, N.C.; CARROLL, E.J.: Veterinary Hematology, pp. 422, , 3rd ed. Lea and Febiger, Philadelphia, SCHNAPPAUF, H.; STEIN, H.B.; SIPE, C.R.; CRONKITE, E.P.: Erythropoietic response in calves following blood loss. Am J Vet Res 28: , SENO, S.; MIYAHARA, M.; ASAKURA, H.; OCHI, 0.; MATSUOKA, K.; TOYAMA, T.: Macrocytosis resulting from early denucleation of erythroid precursors. Blood 24: , SHATTIL, S.1.; COOPER, R.A.: Maturation of macroreticulocyte membranes in vivo. J Lab Clin Med 79: , STOOKEY, L.L.: Ferrozine-A new spectrophotometric reagent for iron. Anal Chern 42: , VAN DILLA, M.A.; SPALDING, J.F.: Erythrocyte volume distribution during recovery from bone marrow arrest. Nature 213: , WEISER, M.G.: Hematologic techniques. In: Veterinary Clinics of North America, ed. Jain and Zinkl, vol. II, pp W. B. Saunders, Philadelphia, WEISER, M.G.: Erythrocyte volume distribution analysis in healthy dogs, cats, horses, and dairy cows. Am J Vet Res 43: , WEISER, M.G.; KOCIDA, G.J.: Persistent macrocytosis assessed by erythrocytesubpopulation analysis following erythrocyte regeneration in cats. Blood 60: , WEISER, M.G.; KOCIBA, G.J.: Sequential changes in erythrocyte volume distribution and microcytosis associated with iron deficiency in kittens. Vet Pat hoi 20: 1-12, YIP, R.; MOHANDAS, N.; JAIN, S.K.; CLARK, M.R.; SHOHET, S.B.; DALLMAN, P.R.: Reduced red cell (RBC) deformability in iron deficiency evidence of increased peroxidation (Abstract). Blood 58(Supplement I): 38a, 1981 Request reprints from M. G. Weiser, Department of Veterinary Pathobiology, 1925 Coffey Road, Columbus, OH (USA).