Welcome to the all-new Vetlearn

  • What’s new on Vetlearn?
  • The latest issues of Compendium and
    Veterinary Technician
  • New CE articles for veterinarians and technicians
  • Expert advice on practice management
  • Care guides on more than 400 subjects
    to give to your clients
  • And more!

To access Vetlearn, you must first sign in or register.


Become a Member

Compendium June 2010 (Vol 32, No 6)

Practical Hematology in Birds — Prospects and Limits

by H. Pendl, Dr.med.vet.


    Source: Tierärztl Prax 2008; 36 (K):290-298. Translated by and reprinted with permission of the publisher.

    A blood panel represents the actual status of circulating cells in the peripheral bloodstream. Many influencing factors cause conspicuous variations, which results in hematological analysis being a very sensitive diagnostic tool with low specificity. With the exception of the detection of hemoparasites and leukemic neoplasias, etiologic diagnoses in hematology are rare. However, the general health status of the patient can be assessed. Serial blood samplings can yield particularly important information regarding the progress of the disease, the prognosis, and the efficacy of a treatment. Avian hematological results demonstrate a high intra- and interspecific variability under physiological conditions, which is caused by the richness in species and the high diversity of metabolic activity due to gender, age, reproductive status, and season.

    In comparison to mammals, avian blood cells show unique morphological characteristics, e.g., erythrocytes and thrombocytes contain a nucleus. The granula-bearing heterophil granulocytes correspond to mammalian neutrophils. Due to these differences, several cell types in the avian blood film are difficult to correctly characterize. This is particularly the case for the differentiation of monocytes versus large lymphocytes, the three types of granulocytes, and the triplet of small lymphocytes versus thrombocytes and juvenile erythrocytes. Associated difficulties are then exacerbated in diseased individuals with altered cell morphology. Therefore, an automatic method could not be previously established that delivered reliable results for a complete blood cell count, which would be comprehensive across all avian species. Time-consuming and labor-intensive manual methods still represent the gold standard for most blood parameters in birds. However, the avian patient requires a rapid and precise laboratory diagnostic workup, as clinical findings often do not correlate with the severity of the disease. Furthermore, the disease length plays a vital role in the success of the treatment and potential for a positive outcome. The methodology presented herein utilizes the measurement of the hematocrit and the evaluation of a stained blood film, which allows a quick first evaluation of the hematological status of the patient. This technique enables the practitioner to initiate therapeutic steps, while awaiting more detailed laboratory results. The following consists of a description introducing the methodology, with special emphasis on possible technical artifacts. Common results in the clinically-healthy bird are described and completed by a list of references and comments on the applications of those results in daily practice. Pathological changes and the associated interpretation are the subjects of a subsequent publication (24).

    Practical Approach

    The hematocrit and a stained blood film are related using a standardized evaluation protocol (Fig. 1). The protocol design follows a strict order, as the assessment of the technical quality and the cytomorphological characteristics of the blood film precede the performance of the differential count and the estimation of total cell numbers. A high quality microscope with a Köhler's transmission light source, including a halide lamp of 6V, 30W minimum and a continuous brightness control, is essential for correct evaluation. The microscope should be equipped with at least three objectives, 10×, 40×, and 100×, resulting in a total magnification of 100×, 400×, and 1,000×. The optical features of the 100× oil immersion objective require special attention. An apparent field of view of 22, with a resolution of 0.27 µm and a total depth of field of 0.69 µm is sufficient to achieve satisfactory results. In case of photographic documentation an apparent field of view of 26.5, a resolution of 0.24 µm and a total depth of field of 0.59 µm will produce good quality images. The photographic system should be installed firmly to the microscope, and the optic features of the camera should correspond to those of the microscope. In digital systems, the use of digital instead of analog cables minimizes data loss due to transfer. The resolution of the computer screen has to be of comparable power to the microscopic capacities to ensure optimal final adjustment before exposure. Computer assisted archiving and analysis of the results is highly recommended, as the generation of a complete blood cell count is considerably accelerated and rapid data access facilitates retrospective studies and evaluation of serial samplings.

    Sampling and Preparation

    Blood sampling is most effectively performed with a combination of a small lumen syringe armed with large lumen needle to minimize suction power, as the thin-walled avian vessels will collapse if the pull is too great. The anticoagulant ethylene diamine tretra acetate (EDTA) causes artificial hemolysis (Fig. 5a) in certain avian species, such as ravens (Corvidae), kingfishers (Alcedinidae), Cranes (Gruidae), and ostriches (Struthionidae). Heparin should be used alternatively. In general, fresh blood can be stored for a short period in EDTA or Heparin without significant losses. Long transport or storage should be avoided, since in addition to cell lysis (Fig. 5c), an increased bacterial growth will occur due to the non-sterile nature of avian blood. Both the associated appearance of bacterial nests (Fig. 5b) and cell lysis in the blood film cannot be differentiated from pathological conditions intra vitam.


    Methodically, the determination of the hematocrit is the same for mammals and birds. The use of a microhematocrit centrifuge is advantageous with small sample sizes. A centrifugation of 5 min at 12,000g is applicable for most centrifuges. However, the timespan for maximum cell compression is device-dependent and should be adjusted according to the individual machine. For most of the avian species, a hematocrit value between 35 - 55% is considered physiologically relevant (3).

    Blood Film Preparation

    Ideally, the blood film is produced without anticoagulant from the last drop of the hub of the needle. In case of storage in EDTA or heparin, mixture after sampling and prior to preparation is vital to avoid separation of blood components and clustering of blood cells. Methodically, the wedge smear technique is a significant component of this procedure (Fig. 2). The use of a bevel edged slide for streaking minimizes the amount of ruptured cells in the blood film. The drop of blood should be completely distributed on the slide resulting in a film with a thinning vane at the end (Fig. 3). Only these samples will contain a monolayer of cells located approximately between the second and last third of the film, which is crucial for a proper examination. The monolayer is defined as the area in the blood film, where 50% of the cells in the view field touch adjacent cells (33). The maximum flattening of the cells in this area enables the examiner to also evaluate subcellular structures (Fig. 4a), which becomes difficult to impossible in denser parts of the film (Fig. 4b). Native blood films need to be protected immediately from environmental dust, since dust will appear as dark precipitate in the stained sample (Fig. 5h).


    The best results are achieved with native, air-dried blood films; alcohol fixation alters the staining quality. Most frequently, these alterations affect the granula in heterophils and basophils. Changes range from slight aberrations (Fig. 5m-n) to optically empty, colorless structures (Fig. 5d). A Wright-Giemsa-Staining (WGS) protocol, which was established by Samour (30), has been demonstrated to be of superior value for avian blood films (Tab. 1). The use of high-quality (acetone-free) methanol in the colorant is vital to achieve good staining results. The solution is stable for several weeks and provides excellent results with distinct delineation of subcellular structures. The widely-used quick stain kits cause a more or less severe loss of the granular structure in granules of heterophils, and therefore, subtle pathological changes in cell morphology cannot be diagnosed (Fig. 5i-l). Basophils exhibit signs of disintegration in the majority of cases, even in WGS stains (Fig. 6j-l), and sometimes disappear completely in quick stained samples (15).

    The optical quality of the stained blood film is increased when mounted with a coverslip and a mounting medium. This is particularly advantageous for evaluation with the 40× objective (Fig. 5e-f). Understaining of cell nuclei in the WGS stain are either due to an aged Giemsa component or an increased pH of the buffer (Fig. 5g). In case of overstaining or the presence of precipitates, the colorant should be filtered and the staining time should be reduced. The grade of staining of heterophilic granules is variable. Occasionally, only the central bodies of the granules are stained with the usual staining protocol and appear as eosinophilic, brick-red dots in the cytoplasm (Fig. 6a). A longer staining time (5 min colorant, 10 min colorant and buffer) results in complete staining of the granules (Fig. 6b). In contrast, eosinophils display an identical appearance irrespective of staining time (Fig. 6c)

    Technical and Morphological Evaluation of the Blood Film

    Examination follows the evaluation protocol outlined in Fig. 1 , beginning with a search for thrombocytic aggregates. In case of occurrence, an estimation of total thrombocyte counts is not possible. Occasionally, aggregations of monocytes and heterophils occur at the borders of the smear. These aggregates can interfere with the results of the differential count and compromise the technical quality of the film. For quality evaluation, a semi-quantitative scale from 1=not acceptable to 5=excellent is useful. Any observed artifacts due to sampling, processing, transport, storage, fixation and staining must be considered.

    To avoid counting errors, an impression of the cell morphology is recommended before performing numerical counts. Erythrocytes represent the majority of cells seen in the film and are elliptic cells with an elliptic nucleus, and a homogeneous eosinophilic cytoplasm. Immature red blood cells exhibit a less elliptic, paler, and more basophilic appearance with a less condensed nucleus. In the adult healthy bird, up to 10% of all erythrocytes in the peripheral blood stream are immature, which correlates well with the amount of polychromatic cells in blood films stained with Romanowsky-type colorants, such as WGS or commonly used quick staining kits. A portion of 1-5% of polychromatic cells is considered physiologically relevant for adult healthy birds. According to Dein (7, 24), these values are consistent with a polychromatic index (PI) of 2 and approximately 2-10 cells per oil immersion field in the monolayer. The ability to differentiate leukocytes and thrombocytes requires some experience. Blood films with differential cell morphology, such as heterophils versus eosinophils, monocytes versus large lymphocytes, and thrombocytes versus small lymphocytes, can be problematic for identification by beginners (Tab. 2 and  Tab. 3). The artifactual variability of heterophil and basophil morphology serves as a distinctive feature for the differentiation from eosinophils. To a large extent, eosinophils are resistant to artifactual alterations and are characterized by species-specific variations in appearance (Fig. 5 j-l), Fig. 6 c-i).

    Even for the experienced examiner, the differentiation of large lymphocytes and monocytes is not always easy using light microscopy and general stainings of the Romanowsy-type (Fig. 6n). Thrombocytes can be recognized by a more pyknotic nucleus (Fig. 5p) and Fig. 6 p), the occurrence of intracytoplasmic pole bodies close to the nucleus (Fig. 6p), and tendencies to aggregate (Fig. 5p). Occasionally, vacuoles can be observed in the cytoplasm. In combination with an irregular, undulated cell membrane, the vacuoles are considered a sign of activation. Thromobocytes are typically easy to identify in blood films with physiological cell morphology, but these cells can be difficult to differentiate from activated lymphocytes and immature erythrocytes in cases with pathologic alterations (24).

    Numerical Evaluation of the Blood Film

    The numerical evaluation of the blood film (Fig. 1) consists of the differential count and estimation of the total thrombocyte and white blood cell count (ttc and twbc). The blood film is examined by means of a meandering search from border to border (Fig. 3) at 1000× oil magnification. For the differential count, at least 100 leukocytes have to be defined as heterophils (H), lymphocytes (L), monocytes (M), eosinophils (E), basophils (B), or non -definable blood cells (NDB). The latter category includes all white blood cells, which cannot be further differentiated due to immature appearance. Depending on the predominant cell type, birds can display a physiologically lymphocytic differential count, such as in Galliformes, Phasianiformes, Anseriformes, and Columbiformes, or a granulocytic differential count like in Psittaciformes, Falconiformes, and Strigiformes.

    The value ranges for total cell count estimations in Table 4 are based on a modified technique, according to Campbell (3). The total number of leukocytes (thrombocytes, respectively) of 20 consecutive oil immersion-view fields (1,000× magnification) is multiplied by a factor of 875 to obtain a total estimated count. The factor consists of constants, such as the average number of erythrocytes per monolayer oil immersion view field, at a physiological hematocrit and physiological total red blood cell count (trbc). In case the hematocrit value is outside of the physiological range between 35-55%, the result has to be corrected by multiplication with the actual hematocrit divided by 45% (Fig. 1, down right). The physiological range of the twbc in avian species with a lymphocytic differential count is higher than in species with a granulocytic differential count. Estimated counts between 15,000 and 45,000 leucocytes/µl for the first group, and 5,000 and 25,000 leucocytes/µl for the second group, are considered physiologically to moderately elevated for adult, healthy individuals. For thrombocytes, estimated values between 20,000 und 40,000 cells/µl are considered physiological for all avian species (Table 4).

    Concluding Remarks

    The benefit of the presented methodology depends essentially on a proper technical preparation of the blood film. Special emphasis has been given to the description of artifacts, as these are the most frequent cause for errors. The WGS stain is preferred over the widely-used quick stains, since the quick stains lead to staining artifacts in heterophils and basophils.

    The use of a meandering search between the borders of the film takes into consideration the natural uneven distribution of blood cells in a blood film. In samples prepared with the wedge smear technique (Fig. 2), lymphocytes tend to concentrate in the central horizontal area, whereas heterophils and monocytes tend to accumulate towards the borders (26). Blood samples from species with a physiological lymphocytic differential count are more affected than samples from species with a predominantly granulocytic differential count. In cases of cluster formation of heterophils and monocytes, a full passage border-center-border has to be examined.

    The interpretation of numerical changes in the blood panel requires a comparison to physiological reference values. The physiological reference range for a particular test is defined as the 95% confidence interval of all obtained measurements. Depending on the variation, the sample size for a meaningful range needs to be 40-60. The measurements need to be independent and the individuals must be representative for the total population. Variation is influenced by internal factors, such as species, age, gender, and metabolic status, and external factors, such as husbandry, housing, diet, climatic conditions, methods of specimen collection, analytical procedure, and the examiner evaluating the sample.

    Taking these aspects into account, reference intervals published in the literature cannot be applied without caution to the patient population of specific practices or laboratories. Furthermore, with few exceptions, published data rarely fulfill the requirements for a representative sample size and standardized experimental conditions. Therefore, these datasets can only serve as rough guidelines, and should be replaced by practice specific reference intervals based on experience. Species-specific hematologic reference values for direct methods with a counting chamber have been published in textbooks and databases (11, 13, 14, 17, 27, 29), dissertations (1, 12, 16, 25, 32), proceedings, and original papers (2, 4-6, 8-10, 18, 19, 28, 31).

    In comparison to direct counts from native blood, blood film estimations are time- and material-saving alternatives, but require a high quality blood film as an essential prerequisite. Various methods for the determination of estimates are described in the literature (3, 20), and corresponding reference values are rarely cited (1, 34). A comparative study between hemocytometer methods and estimations in pigeons, chicken, and budgerigars revealed higher and wider reference intervals for the latter. Accuracy of estimations decreased with increasing total numbers (34). As a consequence, present estimations cannot replace direct counting methods. However, these estimations allow an approximate classification as physiological, increased, or decreased. The intervals given in Table 4 are deliberately wide and general, with the intention to be used as a starting point for the establishment of practice specific ranges. In cases of very valuable birds, individual profiles generated by regular health checks are recommended.

    1. Bürkle M. Untersuchungen zu hämatologischen Referenzwerten unter besonderer Berücksichtigung der Morphologie des weiβen Blutbildes für drei Arten Rotschwanzsittiche (Pyrrhura cruenta, P. h. gaudens, P. perlata). Diss med vet, Gieβen 2003.

    2. Calle PP, Stewart CA. Hematologic and serum chemistry values of captive Hyacinth Macaws (Anodorhynchus hyazinthinus ). J Zoo An Med 1987; 2-3: 98-99.

    3. Campbell TW, Ellis C. Avian and Exotic Animal Hematology and Cytology, 3rd ed. Ames: Blackwell Publishing Professional 2007.

    4. Clubb SL, Schubot RM, Joyner K, Zinkl JG, Wolf S, Escobar J, Clubb KJ, Kabbur MB. Hematologic and serum biochemical reference intervals in juvenile Eclectus Parrots (Eclectus roratus). J Ass Avian Vet 1990; 4 (4): 218-225.

    5. Clubb SL, Schubot RM, Joyner K, Zinkl JG, Wolf S, Escobar J, Kabbur MB. Hematologic and serum biochemical reference intervals in juvenile Cockatoos. J Ass Avian Vet 1991; 5 (1): 16-26.

    6. Clubb SL, Schubot RM, Joyner K, Zinkl JG, Wolf S, Escobar J, Kabbur MB. Hematologic and serum biochemical reference intervals in juvenile Macaws (Ara sp. ). J Ass Avian Vet 1991; 5 (3): 154-162.

    7. Dein FJ. Avian hematology: erythrocytes and anemia. Proc Ann Meet Ass Avian Vet 1983; 10-23.

    8. Dobsinsky O, Dobsinska E. Jahreszeitliche Veränderungen des Blutbilds beim erwachsenen Fasan. Zbl Vet Med R A 1976; 23: 609-615.

    9. Drew ML, Joyner K, Lobinger R. Laboratory reference intervals for a group of captive Thick-billed Parrots (Rhynchopsitta pachyrhyncha ). J Ass Avian Vet 1993; 7 (1): 35-38.

    10. Ernst RA, Coleman TH, Kulenkamp AW, Ringer RK, Pangborn S. The packed cell volume and differential leukocyte count of Bobwhite Quail (Colinus virginianus). Poul Sci 1971; 50: 389-392.

    11. Fudge AM, ed. Laboratory Medicine Avian and Exotic Pets. Philadelphia, London, Montreal, Sydney, Tokyo: Saunders 2000.

    12. Guba E. Untersuchungen über Beziehungen zwischen Blutbild und Rasse, Leistungen und Umwelt bei Hühnern. Diss med vet, München 1955.

    13. Gylstorff I. Blut, Blutbildung und Blutkreislauf. In: Handbuch der Geflügelphysiologie. Mehner A, Hartfiel W, Hrsg. Jena: Fischer 1983; 280-395.

    14. Harrison GJ, Lightfoot TL. Clinical Avian Medicine. Palm Beach Florida: Spix Publishing 2006; 607ff, 1005ff.

    15. Hauska H, Scope A, Vasicek L, Reauz B. Vergleichende Untersuchungen zur Färbung aviärer Blutcelln. Tierärztl Prax 1999; 27 (K): 280-287.

    16. Hauska H. Untersuchungen über das rote Blutbild klinisch gesunder Exemplare einiger ausgewählter Papageienspezies. Diss med vet, München 1994.

    17. International Species Information System (ISIS), 12101 Johnny Cake Ridge Road, Apple Valley, MN 55124, USA, www.worldzoo.org.

    18. Joyner KL, de Berger N, Lopez EH, Brice A, Nolan P. Health parameters of wild psittacines in Guatemala: A preliminary report. Proc Ann Conf Ass Avian Vet, New Orleans 1992; 287-303.

    19. Lane RA, Rosskopf WJ Jr, Allen KL. Avian pediatric hematology: Preliminary studies of the transition from the neonatal hemogram to the adult hemogram in selected psittacine species: African Greys, Amazons & Macaws. Proc Ann Meet Ass Avian Vet Houston/Texas 1988; 231-238.

    20. Lane RA. Avian Hematology: Basic Cell Identification and WBC Count Determination. Proc Int Cont Zool Avian Med, Oahu/Hawaii, 1987; 290-297.

    21. Latimer KS. Response to avian hematology letters. J Ass Avian Vet 1991; 5 (3): 121-124.

    22. Lucas AM, Jamroz C. Atlas of Avian Hematology. Washington: US Dep Agric Monograph 25; 1961.

    23. Pendl H, Reball H. Hämatologische Schnelldiagnostik beim Vogel. Institutsinterner Kurs an der Klinik für Vögel, Ludwig-Maximilians-Universität München, 2003 (nicht publiziertes Kursmaterial).

    24. Pendl H. Praxisnahe Hämatologie beim Vogel. Teil 2: Pathologie und Interpretation. Tierärztl Prax 2008 (in Vorbereitung).

    25. Radzikowski HP. Einfluss von Haltungssystem und Belastung auf einige hämatologische und serologische Parameter bei weiβen Leghornhennen. Diss med vet, München 1979.

    26. Reauz B, Scope A, Hauska H, Vasicek L. Vergleich hämatologischer Untersuchungsmethoden bei Vögeln. Tierärztl Prax 1999; 27 (K): 65-70.

    27. Ritchie BW, Harrison GJ, Harrison LR, eds. Avian Medicine: Principles and Application. Lake Worth Florida: Wingers Publishing Inc 1994; 1330ff.

    28. Rosskopf WJ jr, Woerpel RW, Rosskopf G, Van de Water D. Normal hematologic and blood chemistry values for pet avian species. Proc Ann Meet Ass Avian Vet 1982; 162-168.

    29. Samour JH. Avian Medicine. London, Edinburgh, Philadelphia, St. Louis, Sydney, Tokyo: Mosby 2000; 320ff.

    30. Samour JH. Persönliche Mitteilung 2002.

    31. Vanderheyden N. The hematology of nestling raptors and psittacines. Proc Ann Meet Ass Avian Vet 1986; 347-353.

    32. Veil K. Feststellung von Blutwerten bei verschiedenen Vogelarten. Diss med vet, München 1978.

    33. Weiss DJ. Uniform evaluation and semiquantitative reporting of hematologic data in veterinary laboratories. Vet Clin Pathol 1984; 13: 27-31.

    34. Wiskott M. Vergleich verschiedener Methoden zur Leukozytenzählung bei Vögeln. Diss med vet, Wien 2002.

    PendlLab – Diagnostic Microscopy
    Hematology cytology and histology in birds and reptiles
    Dr. med. vet. H. Pendl
    Eschenweg 14
    CH-6312 Steinhausen

    References »

    NEXT: Surgical Views — Suspensory Ligament Rupture Technique During Ovariohysterectomy in Small Animals [CE]


    Did you know... Disseminated intravascular coagulation is the most commonly seen hemostatic disorder in horses.Read More

    These Care Guides are written to help your clients understand common conditions. They are formatted to print and give to your clients for their information.

    Stay on top of all our latest content — sign up for the Vetlearn newsletters.
    • More