Introduction

Ineffective erythropoiesis is increasingly recognised as a cause of severe idiopathic non-regenerative anaemia (INRA) in the dog. Bone marrow abnormalities range from erythroid hypoplasia due to red cell aplasia through to erythroid hyperplasia with erythrocyte maturation arrest. Based on response to immunosuppressive therapy in a proportion of these cases, an immune-mediated pathogenesis has been proposed.

Objectives

To evaluate outcome in dogs with INRA and to identify objective criteria that could be used to predict response to therapy and long-term outcome.

Study Design

A retrospective observational study of 28 dogs with INRA.

Methods

Dogs with INRA were identified by searching The University of Edinburgh databases between January 2000 and August 2011. Inclusion criteria were PCV < 25%, clinical signs of ≥ 5 days duration, an inappropriate regenerative response (reticulocytes < 120 x 10³/μl) and absence of identifiable underlying disease; all cases had bone marrow evaluation performed. Signalment, duration of clinical signs, prior treatment with glucocorticoids (> 5 days), haematology and serum biochemical parameters, Coombs' status, bone marrow findings, immunosuppressive therapy, and survival at 1, 6 and 12 months were recorded.

Results

Median age was 7 years (range 1–12 years). There was no breed or sex association.

Five of 6 dogs treated with prednisolone alone were alive at 1 month. Of those cases that failed to respond to prednisolone monotherapy, 36% received azathioprine as adjunctive therapy, and 50% received cyclosporine; 75% and 82% respectively of these dogs were alive at 1 month. Three dogs received a combination of prednisolone, azathioprine and cyclosporine with 67% surviving to 1 month.

Overall mortality was 21% at 1 month and 36% at 6 months. All dogs that survived to 6 months were alive at 12-month follow-up.

There was no significant difference in duration of clinical signs, corticosteroid pre treatment, haematology and biochemical parameters, presence of auto-agglutination or spherocytosis, Coombs status, or presence of erythroid hyperplasia between non survivors and dogs that were alive at 1, 6 and 12 months.

Conclusions

Our findings are in broad agreement with previous studies. Prognosis appears to be fair. Further studies are needed to characterise the pathogenesis of this disorder and to determine optimal therapy.