SciELO - Scientific Electronic Library Online

vol.37 issue6Search for hemoglobinopathies in dogs with chronic anemia in the metropolitan region of Rio de JaneiroFipronil/(S)-methoprene spot-on to control fleas on cats in a field trial in Spain author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Pesquisa Veterinária Brasileira

Print version ISSN 0100-736XOn-line version ISSN 1678-5150

Pesq. Vet. Bras. vol.37 no.6 Rio de Janeiro June 2017 


Relation between anaemia and bone marrow features and serum erythropoietin in dogs with chronic kidney disease

Relação entre anemia e achados de medula óssea e eritropoetina sérica em cães com doença renal crônica

Mariana M. Torres2  * 

Felipe A.C.S. Cruz2 

Érica P. Silva2 

Daiane Poletto2 

Mayara A.A. Cayuela2 

Adriane J. Mendonça2 

Arleana B.P.F. Almeida2 

Valéria R.F. Sousa2 

2 Department of Veterinary Clinical Medicine, School of Veterinary Medicine, Federal University of Mato Grosso, Av. Fernando Corrêa da Costa 2367, Boa Esperança, Cuiabá, MT, 78068-600, Brazil.


Chronic anaemia is one of the most severe complications of chronic kidney disease, contributing to morbidity and mortality caused by the disease; therefore, bone marrow cytological evaluation is needed to monitor the progression of anaemia. This study aimed to correlate the anaemia in dogs at different stages of chronic kidney disease with their serum biochemistry, myelogram results and serum erythropoietin findings. Sixty-three dogs were grouped according to International Renal Interest Society (IRIS) classification in stages 1, 2, 3 and 4. Haematologic, serum and urinary biochemistry and serum erythropoietin were performed for comparison with the findings of bone marrow cytology obtained by aspiration of the manubrium. Cytological findings for erythroid hypoplasia were described in 93.65% of dogs, and the anaemia was observed in 84.1% of them. The haematological findings were correlated with azotaemia (p<0.05). It was concluded that the erythroid hypoplasia has correlation with persistent anaemia in dogs at all stages of chronic kidney disease, with iron deficiency in dogs in the early stages and with peripheral destruction of erythrocytes caused by azotaemia.

INDEX TERMS: Anaemia; bone marrow; serum erythropoietin; dogs; chronic kidney disease; cytology; erythrocytes; iron metabolism; myelogram; renal.


A anemia crônica é umas das complicações mais graves da doença renal crônica, contribuindo para a morbidade e mortalidade causada pela doença; Portanto, a avaliação citológica da medula óssea é necessária para monitorar a progressão da anemia. Assim, esse estudo objetivou correlacionar a anemia em cães em diferentes estágios da doença renal crônica aos achados de bioquímica sérica, mielograma e concentração sérica de eritropoietina. Sessenta e três cães foram agrupados de acordo com a classificação da International Renal Interest Society (IRIS) em estágios 1, 2, 3 e 4. Foram realizadas análises hematológicas, bioquímicas séricas e urinárias, e dosagem sérica de eritropoetina para comparação com os achados medulares obtidos por citologia aspirativa do manúbrio. Os achados citológicos de hipoplasia eritróide foram descritos em 93,65% dos cães, e a anemia foi observada em 84,1% dos cães. Os resultados hematológicos foram correlacionados com azotemia (p<0,05). Concluiu-se que a hipoplasia eritróide teve associação com a anemia persistente em cães em todas as fases de doença renal crônica, com deficiência de ferro em cães em fases iniciais e com a destruição periférica dos eritrócitos causada pela azotemia.

TERMOS DE INDEXAÇÃO: Anemia; medula óssea; eritropoetina sérica; caninos; doença renal crônica; citologia; eritrócitos; metabolismo do ferro; mielograma; renal.


Chronic anaemia, usually of the normocytic normochromic nonregenerative type, is a common and invariable complication of chronic kidney disease (CKD), resulting from decreased erythrocyte half-life, iron and folic acid deficiency, the action of uremic toxins and, especially, erythropoietin deficiency (Waki et al. 2010, Macdougal 2011, Polzin 2011, Zadrazil & Horak 2014).

In humans, CKD anaemia is correlated with cardiovascular disturbances and other comorbidities (Palmer et al. 2010). Similarly, in dogs, the production of erythropoietin is markedly reduced in CKD, leading to anaemia that is difficult to treat. With any of the several treatments proposed to treat anaemia associated with CKD, regardless of the adopted therapy, the mensuration of erythropoietin serum concentration as a part of therapeutic monitoring is fundamental (Giampaoli et al. 2003).

The cytological evaluation of bone marrow allows for the analysis of a patient’s response to the progression of the anaemia, acting as another important tool for monitoring the development of the CKD. The anaemia becomes more evident in advanced stages of CKD, compromising the quality of life and contributing to the high morbidity and mortality of the dogs (IRIS 2013, Zadrazil & Horak 2014, Kogika et al. 2015).

Therefore, this study aimed to correlate the anaemia in dogs at different stages of CKD with their serum biochemistry, myelogram results and erythropoietin serum concentration.

Materials and Methods

Ethics statement. This study was approved and conducted in accordance with the recommendations of the Animal Care and Use Ethics Committee of the University of Mato Grosso (protocol # 23108.043331/12-9 CEUA).

Animals. The prospective study included dogs of different breeds and both sexes with CKD. As inclusion criteria, were selected dogs with or without anaemia, presenting azotaemia and hyposthenuria and which have not undergone to any treatment for this condition, as well as for high blood pressure. They were grouped according to the International Renal Interest Society (IRIS 2013) criteria, based on the creatinine serum concentration in: stage 1 (<1.4mg/dl), stage 2 (1.4-2.0mg/dl), stage 3 (2.1-5mg/dl) and stage 4 (≥5mg/dl). The substaging was determined through the assessment of the systolic blood pressure, measured in triplicate in all dogs by oscillometry (Pet Map®, Rio de Janeiro, Brazil), and of the presence of proteinuria by urine protein-to-creatinine (UPC) ratio.

Blood, urine and bone marrow samples were obtained from dogs for the following diagnostic tests: Complete blood count (CBC), serum and urinary biochemistry, urinalysis, dosage of serum erythropoietin and myelogram.

Another group comprising healthy dogs of different breeds and both sexes was used to obtain reference parameters, with those dogs presenting haematological, biochemical and urinary disorders being excluded.

Complete blood count and biochemistry. For the CBC, the haematocrit, haemoglobin, red blood count, mean corpuscular haemoglobin concentration (MCHC), mean corpuscular volume (MCV) and platelet count were assessed with an automatic analyser (Sysmex pocH-100iV, São Paulo, Brazil).

The serum biochemistry were assessed with a semiautomatic analyser (Celm SB-190®, São Paulo, Brazil) and commercial kits (Gold Analisa®, Belo Horizonte, Brazil) to determine levels of urea, creatinine, iron, total iron binding capacity (TIBC), calcium and phosphorus. The UPC ratio was obtained by calculating the ratio of the amount of urinary protein and urinary creatinine determined with a semiautomatic analyser (Celm SB-190®, São Paulo, Brazil) and commercial kits (Gold Analisa®, Belo Horizonte, Brazil), according to Castro et al. (2009).

Myelogram. Marrow aspirates were obtained from the manubrium of each dog, with EDTA (3%) being used as the anticoagulant. Then, the smears containing bone spicules were submitted to Romanowsky staining.

The evaluation was based on cellular identification according to the morpho-tinctorial features of the cells. A total of 500 cells were counted and the percentages of erythroid and myeloid cells and the myeloid/erythroid ratio were calculated, and the cytological findings were described. The cellularity of bone marrow was estimated by examining the proportion of cells versus fat, considering as a hypocellular marrow when the particles were composed of more than 75% of fat (Harvey 2001).

Reference values were those previously described (Harvey 2001, Weiss & Smith 2002).

Serum erythropoietin. The serum erythropoietin in dogs was determined with an ELISA sandwich approach using a canine-specific commercial kit (Cloud-Clone Corp.®, Houston, TX, USA).

Statistical analysis. The normality of the data was determined using the Kolmogorov-Smirnov test. An analysis of variance was employed for comparisons among the CKD stages, and the Pearson’s correlation coefficient was applied to the data on the bone marrow findings. The statistical analyses were performed with GraphPad Prism 4 software.


Sixty-three dogs, including 29 females and 34 males, with a mean age of 6.7 years (ranging from one to 16 years) and of varied breeds, composed the CKD group. Nine healthy dogs of different breeds, six males and three females with an average age of three years (one to five years), comprised the control group.

Among the CKD dogs, 11 were classified as being in stage 1, eight were classified as being in stage 2, 22 were classified as being in stage 3 and 22 were classified as being in stage 4. Sub-staging was determined by systolic blood pressure and by the UPC ratio, with all dogs showing proteinuria (>0.5) and hyposthenuria, along with a specific urinary gravity ranging from 1008 to 1018.

A high systolic blood pressure (≥180 mmHg) was observed in 34% of the dogs, with a predominance occurring in stage 4 in comparison with the other stages (p=0.02).

Clinical signs observed in CKD dogs ranged in severity from occasional vomiting (47.6%) and inappetence (93.6%) to severe anorexia (19%), frequent vomiting (52.4%), tongue necrosis (1.6%) and pale mucosa (62.26%).

Regarding the erythrogram, all the assessed parameters, except MCV and MCHC, declined as the CKD progressed, which characterizes the normocytic normochromic anaemia observed in 84.1% of the dogs. Of these, 37.5% had severe anaemia (haematocrit ≤ 20%) with reticulocytosis indexes ranging from zero to 13.2% (mean 2.7%), and 11.1% presented normocytic hypochromic anaemia.

The red blood count, haematocrit and haemoglobin levels showed statistically significant differences only compared to those of the control dogs (p≤ 0.05). Thrombocytopenia occurred in 39.6% of the anaemic dogs; however, there was no significant difference (p=0.81) among the CKD stages.

No significant difference (p= 0.37) between stages was observed for the myeloid/erythroid ratio, as determined from the myelogram. The marrow findings (Figure 1) of the CKD dogs are described in Table 1.

Fig.1. Bone marrow aspirate in a dog with chronic kidney disease in stage 4. Romanowsky stain. (A) The normal marrow cellularity. Obj.10x. (B) Myeloid cells in higher proportion than erythroid, presence of several plasmocytes. Obj.40x. (C) Plasmocyte (arrow head). Obj.100x. (D) Dyserythropoiesis (arrow head) and prorubricytes. Obj.100x. (E) Two band cells and metamyelocyte. Obj.100x. (F) Myeloblasts and band cells. Obj.100x. 

Table 1. Cytologic features of the bone marrow in dogs with chronic kidney disease 

Parameters %
Normal cellularity 38 60.31
Myeloid hyperplasia 20 31.74
Megakaryocytic hyperplasia 05 7.93
Erythroid hypoplasia 59 93.65
Myeloid hypoplasia 04 6.35
Lymphocytosis 06 9.52
Plasmocytic hyperplasia 07 11.11
Dysgranulopoesis 05 7.93
Myelodysplastic syndrome 02 3.17

In the serum biochemistry, the calcium, phosphorus, total iron binding capacity and iron concentrations did not show significant differences among the CKD stages, except for the serum phosphorus, which showed a significant difference in dogs at stage 4 in comparison to the control group (Table 2).

Table 2. Summary of the serum concentration statistics for calcium (Ca), phosphorus (P), iron, iron binding capacity (TIBC) and erythropoietin (EPO) in the diseased and control groups 

Parameters Stage 1 Stage 2 Stage 3 Stage 4 Control
n= 11 n= 08 n= 22 n= 22 n= 09
Ca (mg/dl)
Mean 11.44 10.45 11.18 11.78 12.52
Median 11.57 10.53 11.50 11.40 12.54
Standard deviation 1.66 0.96 3.33 2.38 0.51
p-value* 0.7
P (mg/dl)
Mean 2.08 1.75 2.97 3.58a 1.1a
Median 1.45 1.60 2.20 3.20 1.1
Standard deviation 1.75 0.88 1.95 1.95 0.12
p-value* 0.04
Iron (μg/dl)
Mean 195.8 145.8 146.4 109.0 233.8
Median 193.0 115.0 128.5 83.00 230.0
Standard deviation 89.93 85.48 108.1 112.3 113.0
p-value* 0.07
TIBC (μg/dl)
Mean 565.6 599.8 556.7 572.1 478.8
Median 584.0 590.5 487.0 528.0 506.0
Standard deviation 92.19 65.22 204.7 200.2 139.8
p-value* 0.83
Mean 1421 - 1474 1461 1499
Median 1469 - 1480 1484 1501
Standard deviation 112.1 - 51.61 80.23 18.09
p- value* 0.27

*Analysis of variance (Ca, P, Iron, TIBC). EPO = control vs stages). a Means followed by the same letter in the same row differ at the 5% level of significance. ** Evaluated in 44 dogs (39 CKD.

The concentration of erythropoietin was determined in 44 of the 63 dogs in the study. The values obtained were not correlated with the bone marrow or erythrogram findings.

Among the dogs diagnosed with erythroid hypoplasia, 19 presented with haematocrit ≤ 20%, and nine of them did not present reticulocytosis. Additionally, whereas megakaryocytic hyperplasia was present in five dogs; two of these also presented with thrombocytopenia.

Myelodysplastic syndrome was described in two dogs based on the presence of dysmyelopoiesis and rubriblasts observed in 7 to 18% of the nucleated cells.

Serum iron levels were correlated with erythroid precursors, with a significant correlation with the myeloid/erythroid ratio (r= -0.95) and the percentage of rubriblasts (r= -0.89), rubricytes (r= 0.89) and metarubricytes (r= 0.91) being observed in stage 2 dogs, whereas in stage 4 dogs, a correlation was observed only with the myeloid/erythroid ratio (r= 0.66).

The haematocrit of all dogs was correlated with the serum biochemistry, serum levels of erythropoietin and the myeloid/erythroid ratio, as shown in Table 3.

Table 3. Spearman’s rank correlation between the haematocrit and the myeloid/erythroid ratio (MER), serum iron, total iron binding capacity (TIBC), calcium, phosphorus, urea, creatinine and erythropoietin (EPO) in diseased dogs 

Variables r p-value
Haematocrit and MER 0.13 0.49
Haematocrit and Iron 0.10 0.39
Haematocrit and TIBC -0.13 0.35
Haematocrit and Calcium -0.02 0.86
Haematocrit and Phosphorus 0.09 0.45
Haematocrit and Urea -0.33 0.003*
Haematocrit and Creatinine -0.27 0.01*
Haematocrit and EPO 0.0006 0.99

* Significant difference.


According to the IRIS (2013) classification for CKD, the initial stages of 1 and 2 tend to be subclinical, explaining why most of the dogs in this study belong to stages 3 and 4. These dogs were also classified as proteinurics and hypertensives in all CKD stages, with systolic blood pressure higher than 180 mmHg observed predominantly in stage 4. Proteinuric and/or hypertensive patients can present clinical manifestations in the early stages of CKD; therefore, the classification and sub-staging proposed by IRIS (2013) allows the assessment of renal damage and associated complications, which can serve as a prognostic factor (Waki et al. 2010, Polzin 2011).

Although there is no breed or age predilection (Polzin 2011), the morbidity and mortality occur predominately in elderly patients (Lees 2004), as demonstrated by the dogs in this study.

In humans with kidney disease, a strong correlation exists between the degree of anaemia and a risk of morbidity and mortality from cardiovascular diseases (Locatelli et al. 2003). Although this correlation was not significant in our study, the fact that 34% of the CKD dogs were hypertensive (systolic blood pressure ≥ 180 mmHg), exhibited increased azotaemia, and were in the advanced stages classifies these dogs as critically ill.

In CKD dogs, the azotaemia, acting mainly on the haematocrit, proved to be a major factor in the persistence of anaemia (p<0.05) due to the progressive deterioration in renal function, culminating with decreased haemoglobin concentration, which becomes evident when the glomerular filtration rate declines (Macdougall 2011).

The effects of uremia on the oxidative stress in erythrocytes in anaemic dogs with CKD was previously studied, and although no correlation was demonstrated, anaemia is associated with reduced erythrocyte lifespan (Kogika et al. 2015). Although the oxidative stress assessment has not been performed in our study, the uremia described could be one of the factors involved in chronic anaemia and a possible cause for the described peripheral alterations due to the significant correlation among the haematocrit and azotaemia observed in our study.

Furthermore, hyperphosphatemia contributes to this process, mainly in the advanced stages of CKD, as observed in the dogs in stage 4. Serum phosphorus level elevation, in association with normal calcium levels, were also reported in a majority of dogs in a previous study (Kogika et al. 2006). In CKD, such an imbalance can be observed from the earliest stages of the disease (Cortadellas et al. 2010) and may result in renal secondary hyperparathyroidism, which implies a lower survival for these dogs (Geddes et al. 2013).

Regarding the serum iron concentrations, although no differences were observed among the stages, the effect of these concentrations on erythroid cells, primarily in stages 2 and 4, show that iron acts on mature cell lines in one way and, curiously, in an inverse way on immature cell lines. This mild anaemia may be the result of inflammatory disease and, additionally, it reduces the erythrocyte lifespan and its responsiveness to the anaemia due to low erythropoietin synthesis and iron deficiency, thereby culminating in limited erythropoiesis (Almeida et al. 2009).

Iron serum dosage is one of the confirmation methods of this deficiency when the haematological findings suggest hypochromic microcytic anaemia, this diagnosis can be affirmed based on our results which prove that in CKD, the occurrence of hyposideremia has lower relation with anaemia. Furthermore, microcytosis is a good indicator of iron deficiency only when MCV values are extremely low, and this situation, which was not reported in our study, is related to blood loss (Paltrinieri et al. 2010).

As previously shown, differences between the means of the serum levels of erythropoietin and iron in the different CKD stages were not observed and did not correlate with the myelogram and CBC findings; however, the normal values of the myeloid/ erythroid ratio in anaemic dogs and the low reticulocytosis indexes in 47.3% of the dogs with erythroid hypoplasia indicate erythropoiesis suppression. Macdougall (2011) suggests that in addition to the observed decrease in erythropoietin production, decreased erythropoietin activity on bone marrow may also occur. This mechanism could explain our erythropoietin findings.

The most frequent anaemia in CKD is normocytic normochromic anaemia (Macdougall 2011), as reported in 84.1% of the dogs in this study. This form of anaemia is described as frequent and recurrent (Weiss 2006, Raskin & Messik 2012).


The erythroid hypoplasia observed in CKD dogs is preponderant in the persistence of anaemia from the early stages of this disease on.

The erythroid hypoplasia was associated with iron deficiency in the initial stages and with peripheral erythrocyte destruction caused by progression of the azotaemia.

Cytological examination of the bone marrow enabled to determinate the erythropoiesis suppression occurring in the CKD dogs, primarily against mature erythroid progenitors, which emphasizes the usefulness of this method as a prognostic tool in anaemic CKD patients.


The authors are grateful to Capes (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) for the scholarship during the study.


Almeida R.K., Vasconcelos A.C., Carneiro R.A., Paes P.R.O. & Moro L. 2009. Alterações citológicas do sangue periférico e da medula óssea de cães com cinomose. Arq. Bras. Med. Vet. Zootec. 61:1255-1260. [ Links ]

Castro M.C.N., Marcello G.C.G., Alencar N.X. & Ferreira A.M.R. 2009. Avaliação da relação proteína-creatinina urinária em gatos com doença renal crônica. Pesq. Vet. Bras. 29:605-609. [ Links ]

Cortadellas O., Fernández del Palacio M.J., Talavera J. & Bayón A. 2010. Calcium and phosphorus homeostasis in dogs with spontaneous chronic kidney disease at different stages of severity. J. Vet. Intern. Med. 24:73-79. [ Links ]

Geddes R.F., Finch N.C., Syme H.M. & Elliot J. 2013. The role of phosphorus in the pathophysiology of chronic kidney disease. J. Vet. Emerg. Crit. Care 23:122-133. [ Links ]

Giampaoli S., Facciabene A. & Mennuni C. 2003. A protocol to guide development of a sensitive ELISA for canine erythropoietin. Vet. Clin. Pathol. 32:199-201. [ Links ]

Harvey J.W. 2001. Atlas of veterinary hematology: blood and bone marrow of domestic animals. W.B. Saunders, Philadelphia, p.92-184. [ Links ]

IRIS 2013. IRIS staging of CKD < IRIS staging of CKD > Acessed on Aug. 14, 2014. [ Links ]

Kogika M.M., Lustoza M.D., Notomi N.K., Wirthl V.A.B.F., Mirandola R.M.S. & Hagiwara M.K. 2006. Serum ionized calcium in dogs with chronic renal failure and metabolic acidosis. Vet. Clin. Pathol. 35:441-445. [ Links ]

Kogika M.M., Lustoza M.D., Hagiwara M.K., Caragelasco D.S., Martorelli C.R. & Mori C.S. 2015. Evaluation of oxidative stress in the anemia of dogs with chronic kidney disease. Vet. Clin. Pathol. 44:70-78. [ Links ]

Lees G.E. 2004. Early diagnosis of renal disease and renal failure. Vet. Clin. Small Anim. Pract. 34:867-885. [ Links ]

Locatelli F., Canaud B., Eckardt K.-U., Stenvinkel P., Wanner C. & Zoccali C. 2003. Oxidative stress in endstage renal disease: an emerging threat to patient outcome. Nephrol. Dial Transplant. 18:1272-1280. [ Links ]

Macdougall I.C. 2011. Anemia and chronic renal failure. Medicine 39:425-429. [ Links ]

Palmer S.C., Navaneethan S.D., Craig J.C., Johnson D.W., Tonelli M., Garg A.X., Pellegrini F., Ravani P., Jardine M., Perkovic V., Graziano G., McGee R., Nicolucci A., Tognoni G. & Strippoli G.F.M. 2010. Meta-analysis: erythropoiesis-stimulating agents in patients with chronic kidney disease. Ann. Intern. Med. 153:23-33. [ Links ]

Paltrinieri S., Preatoni M. & Rossi S. 2010. Microcytosis does not predict serum iron concentrations in anaemic dogs. Vet. J. 185:341-343. [ Links ]

Polzin DJ. 2011. Chronic kidney disease in small animals. Vet. Clin. Small Anim. 41:15-30. [ Links ]

Raskin R.E. & Messick J.B. 2012. Bone marrow cytologic and histologic biopsies: indications, technique, and evaluation. Vet. Clin. Small Anim. 42:23-42. [ Links ]

Waki M.F., Martorelli C.R., Mosko P.E. & Kogika M.M. 2010. Classificação em estágios da doença renal crônica em cães e gatos - abordagem clínica, laboratorial e terapêutica. Ciência Rural 40:2226-2234. [ Links ]

Weiss D. & Smith S.A. 2002. Collection and assessment of canine bone marrow. Compendium 24:670-678. [ Links ]

Weiss D.J. 2006. A retrospective study of the incidence and the classification of bone marrow disorders in the dog at a veterinary teaching hospital (1996-2004). J. Vet. Intern. Med. 20:955-961. [ Links ]

Zadrazil J. & Horak P. 2014. Pathophysiology of anemia in chronic kidney diseases: a review. Biomed. Pap. Med. Fac. Univ. Palacky Olomouc Czech Republ. 159:197-202. [ Links ]

Received: January 08, 2016; Accepted: September 12, 2016

*Corresponding author:

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License