Você está na página 1de 9

Hematopathology / LABORATORY IDENTIFICATION OF CRYOGLOBULINEMIA

Laboratory Identification of Cryoglobulinemia From Automated Blood Cell Counts, Fresh Blood Samples, and Blood Films
Anne Fohlen-Walter, PhD,1 Christine Jacob, MD,2 Thomas Lecompte, MD,1 and Jean-Franois Lesesve, MD1
Key Words: Cryoglobulinemia; Automated blood cell count; Blood film artifact

Abstract
Four cases showing different means to detect cryoglobulins are reported: effects on blood cell counts performed on 2 technologically different automated hematology instruments and microscopic features in fresh blood samples and on MayGrnwald-Giemsastained blood films. These cases were chosen for their instructive value in depicting all artifacts associated with cryoglobulins. Laboratory recognition of the cryoglobulins is important to correct factitious results with automated blood cell counters, mainly pseudoleukocytosis and pseudothrombocytosis. Moreover cryoglobulininduced laboratory artifacts may be the first factor prompting the assessment for cryoglobulinemia and the diagnosis of the underlying cause.

Cryoglobulins are circulating immunoglobulins or immunoglobulin complexes characterized by reversible coldinduced precipitation occurring between 4C and 37C.1 Cryoglobulins have been reported in various infectious, renal, hepatic, autoimmune, hematologic (especially multiple myeloma and lymphoproliferative disorders), and neoplastic diseases, but they also can occur in the absence of any apparent relevant disease.2 Clinical manifestations, including purpura, Raynaud syndrome, arthralgia, peripheral neuritis, or renal disease, are inconstant, and this emphasizes the need for relevant laboratory screening tests.3 The in vitro effects of cryoglobulins are less well documented, especially for blood cell counts ascertained by automated cell counters. Erroneous WBC counts have been reported as the major artifact, and abnormal platelet counts have been mentioned occasionally.4 Recognition of the hematologic abnormalities associated with cryoglobulins may be the first clue leading to the diagnosis of cryoglobulinemia and eventually to its underlying cause. We report anomalies in blood count histograms and scatterplots in 4 patients with cryoglobulinemia and the appearance of cryoglobulins in microscopy.

Materials and Methods


Cases were obtained from the hematology laboratory of the University Hospital, Nancy, France, and were selected for their instructive value in demonstrating different manifestations of cryoglobulinemia on blood cell analysis. Peripheral blood specimens were obtained in vacuum tubes containing tripotassium EDTA (8.4 mmol/L final concentration; Vacutainer, Becton Dickinson, San Jose, CA) and maintained at room
606

Am J Clin Pathol 2002;117:606-614

American Society for Clinical Pathology

Hematopathology / ORIGINAL ARTICLE

Table 1 Clinical and Laboratory Manifestations at Room Temperature for Four Cases of Cryoglobulinemia*
Blood Count Abnormalities Case No. Bayer Technicon H*2 1 2 3 4
*

Manifestations Fresh Blood Sample Large clear amorphous material Blood Smear Extracellular pale pink amorphous material; multiple faintly basophilic neutrophil inclusions None None None

Coulter STKS Pseudoleukocytosis

Pseudoleukocytosis Pseudothrombocytosis; Pseudoleukocytosis Pseudothrombocytosis Pseudoleukocytosis

Pseudoleukocytosis; Thin bright deposits pseudothrombocytosis Pseudoleukocytosis Thin bright reticulum Not done Fusiform crystals

For proprietary information, see the Materials and Methods section.

Table 2 Hematologic Findings for Four Cases*


Case 1 RT WBC count ( 109/L) Bayer H*2 peroxidase channel Bayer H*2 basophil channel Coulter STKS Platelet count ( 109/L) Bayer H*2 Coulter STKS Mean platelet volume (fL) Bayer H*2 Coulter STKS 14.94 4.5 9.3R 213 213 9.4 8.0 37C RT 12.61 11.31 15.3RH 906 128R 4.6 7 .9R Case 2 37C RT Case 3 37C RT 7 .77 6.26 ND 272 ND 6.6 ND Case 4 37C

4.3 NS 4.4 234 204 9.0 7 .1

11.51 NS 10.9 112 85 8.2 10.0

3.08 NS 5.4R 173 106R 6.5 8.0R

3.02 NS 2.8 112 99 7 .7 8.0

6.23 6.24 ND 269 ND 7 .1 ND

ND, not done; NS, statistically not different from peroxidase channel; RT, room temperature; 37C, after 30 minutes at 37C. * Values for WBC and platelet counts are given in conventional units; conversions to Systme International are as follows: WBC count, divide by 0.001 to obtain value per microliter; platelet count, divide by 1.0 to obtain value 103/L. For proprietary information, see the Materials and Methods section. R, RH: Flagged results according to the instrument. First anomaly leading to the investigation of cryoglobulinemia.

temperature until analysis. Blood counts were performed using 2 technologically different blood cell counters: an electrical impedance counter (STKS, Beckman Coulter, Hialeah, FL) and an optical-based instrument (Technicon H*2, Bayer, Tarrytown, NY). Blood analysis was performed at room temperature and after heating the sample for 30 minutes at 37C. A drop of fresh blood was examined at room temperature with a phasecontrast microscope. This is readily accomplished by placing a small drop of blood beneath a coverslip on a glass side. Blood films were prepared from specimens obtained at room temperature and were stained according to the May-Grnwald-Giemsa technique. Cryoglobulinemia was confirmed by the detection of protein precipitates in the serum maintained at 4C during at least 7 days, which dissolved when heated at 37C.

Results
Cases 1 and 2 were associated with multiple myeloma and B-cell malignant lymphoma, respectively. Case 3 had a diagnosis of interstitial nephropathy. In case 4, cryoglobulinemia
American Society for Clinical Pathology

was transient, concomitant to a pulmonary infection. In all cases, the abnormalities on blood cell counts were observed before the clinical and biochemical diagnosis of cryoglobulinemia. Moreover, for case 2, investigation of the lymphocytosis-associated cryoglobulinemia (lymphocytes, 6,880/L [6.9 10 9 /L]) prompted the diagnosis of lymphoma. Cryoglobulinemia was confirmed by biochemical methods in serum samples for cases 1, 2, and 3. Interestingly, for case 2, a precipitate already was noted in the plasma of the sample used for blood cell counts after sedimentation at room temperature. Relevant hematologic findings are summarized in Table 1 and Table 2. RBC counts and hemoglobin measurements were slightly affected by the presence of the cryoglobulins. Case 2 was the only case associated with a flatness alarm on the H*2 analyzer, showing flow abnormality and leading to slightly decreased RBC count and hemoglobin measurement at room temperature compared with analysis of the sample after 30 minutes at 37C (room temperature: RBC count, 4.77 106/L [4.77 1012/L]; hemoglobin, 12.7 g/dL [127 g/L]; 37C: RBC count, 5.00 106/L [5.00 1012/L];
Am J Clin Pathol 2002;117:606-614
607

Fohlen-Walter et al / LABORATORY IDENTIFICATION OF CRYOGLOBULINEMIA

50

100

200 fL

300

400

50

100

200 fL

300

400

Perox

Perox

0-20 fL

0-20 fL

Figure 1 (Case 1) Histograms at room temperature (left) and 37C (right) for the Coulter STKS (A) and Bayer Technicon H*2 (B) counters. The arrows indicate the small particles at the left of the lymphocyte monocyte granulocyte diagram (Coulter STKS) and the interference on the perox channel (peroxidase activity) scatterplot (Bayer Technicon H*2). B, Leukocyte perox channel (top) and platelets curve (bottom). For proprietary information, see the Materials and Methods section.

hemoglobin, 13.4 g/dL [134 g/L]). Cases are illustrated in Figure 1, Figure 2, Figure 3, and Figure 4, respectively, depicting the principal manifestations of cryoglobulins on the blood count depending on the instrument used. For the STKS counter, the lymphocyte monocyte granulocyte (LMG) diagram showing the volume distribution of
608

WBCs, with the anomaly of partially lysed leukocytes, is represented. Volume conductance scattering and platelet volume distribution histograms are not shown because they were not altered by the presence of cryoglobulins. The STKS counter, based on an impedance principle, mainly detects cryoglobulins as small particles (<35 fL) at the left
American Society for Clinical Pathology

Am J Clin Pathol 2002;117:606-614

Hematopathology / ORIGINAL ARTICLE

50

100

200 fL

300

400

50

100

200 fL

300

400

Perox

Perox

0-20 fL

0-20 fL

Figure 2 (Case 2) Histograms at room temperature (left) and 37C (right) for the Coulter STKS (A) and Bayer Technicon H*2 (B) counters. The arrows indicate the small particles at the left of the lymphocyte monocyte granulocyte diagram (Coulter STKS) and the triangle-shaped histogram of platelet volumes (Bayer Technicon H*2). B, Leukocyte perox channel (peroxidase activity) (top) and platelets curve (bottom). For proprietary information, see the Materials and Methods section.

of the LMG diagram, leading to spuriously elevated WBC counts (R* flag). For the optical-based H*2 analyzer, the peroxidase activity (perox channel scatterplot) and the histogram of platelet volume distribution are represented. In scatterplots using the perox channel, forward light scatter, largely determined by cell volume (y-axis), is
American Society for Clinical Pathology

plotted against light absorbance largely determined by the intensity of the peroxidase reaction (x-axis). Cryoglobulins interfere mainly on the histogram of platelet volume distribution, generating a characteristic triangular figure. Platelet counts are spuriously elevated, and the mean platelet volume is usually absurd (too small for true platelets)
Am J Clin Pathol 2002;117:606-614
609

Fohlen-Walter et al / LABORATORY IDENTIFICATION OF CRYOGLOBULINEMIA

50

100

200 fL

300

400

50

100

200 fL

300

400

Perox

Perox

0-20 fL

0-20 fL

Figure 3 (Case 3) Histograms at room temperature (left) and 37C (right) for the Coulter STKS (A) and Bayer Technicon H*2 (B) counters. The arrows indicate the small particles at the left of the lymphocyte monocyte granulocyte diagram (Coulter STKS) and the interference on the histogram of platelet volumes (Bayer Technicon H*2). B, Leukocyte perox channel (peroxidase activity) (top) and platelets curve (bottom). For proprietary information, see the Materials and Methods section.

(Figures 2 and 3). Cryoglobulins rarely manifest on the perox channel scatterplot, leading to falsely elevated WBC counts (Figures 1 and 4). The basophil/lobularity channel, analyzing WBCs according to their volume following differential cytoplasmic stripping, is not affected by cryoglobulins.
610

Examination of fresh blood samples at room temperature revealed that cryoglobulins might take different appearances depending on the patients. For example, a large flake of clear amorphous material is shown (case 1) Image 1, as are thin refringent precipitates (cases 2 and 3) Image 2 and fusiform crystals of different thicknesses (case 4) Image 3.
American Society for Clinical Pathology

Am J Clin Pathol 2002;117:606-614

Hematopathology / ORIGINAL ARTICLE

Perox

Perox

0-20 fL

0-20 fL

Figure 4 (Case 4) Bayer Technicon H*2 counter histograms at room temperature (left) and 37C (right). The arrows indicate the interference on the perox channel (peroxidase activity) scatterplot. Leukocyte perox channel (top) and platelets curve (bottom). For proprietary information, see the Materials and Methods section.

May-Grnwald-Giemsastained blood films usually were normal (cases 2-4). But in case 1, the film revealed some pinkish amorphous deposits between RBCs Image 4 and multiple basophilic inclusion bodies in the majority of neutrophils Image 5. These inclusion bodies sometimes were so abundant that they compressed and distorted the nucleus. Table 3 proposes a guide to screen for and definitively diagnose the presence of cryoglobulins.

Discussion
Cryoglobulins may precipitate at temperatures less than 37C; the precipitation temperature is highly variable from one cryoglobulin to another.1 An interference of the presence of cryoprecipitates on blood cell counts is thus likely to occur if cryoprecipitation develops rapidly and at room temperature (18C-25C). In the 4 cases reported, environmental cooling of the blood from venipuncture to analysis was sufficient to permit cryoprecipitates to form. Of clinical
American Society for Clinical Pathology

interest, and as in our case records, cryoglobulinemia may be diagnosed several years before the underlying cause, especially for hematologic diseases.1,5-7 It has been reported that electronic cell counters may mistake the precipitated cryoglobulins for blood cells when the size of the precipitates falls within the range of the aperture through which cells in suspension are drawn. 8-10 However, comparison of automated blood cell counts for patients with cryoglobulinemia using instruments with different technologies has seldom been reported in the medical literature. In almost all reported cases, blood cell counts were performed on a Coulter counter Model S8-14 or a Coulter counter Model S Plus, 15,16 leading to falsely elevated WBC counts. We showed that cryoglobulins may interfere in WBC and/or platelet counts, even with current analyzers. The degree of interference in WBC or platelet counts seems to depend not only on the instrument used but also on the type of precipitate. Indeed, spuriously elevated cell counts are due to the presence of particles of cryoglobulin being counted as WBCs or platelets in relation to such physical properties as their size, structure, and shape.8-10
Am J Clin Pathol 2002;117:606-614
611

Fohlen-Walter et al / LABORATORY IDENTIFICATION OF CRYOGLOBULINEMIA

Image 1 (Case 1) Large flake of clear amorphous material (contrast microscope, original magnification 650).

Image 2 (Cases 2 and 3) Thin refringent precipitates (contrast microscope, original magnification 650).

Image 3 (Case 4) Fusiform crystals (contrast microscope, original magnification 400).

Image 4 (Case 1) Blood smear showing pinkish amorphous deposits between RBCs (May-Grnwald-Giemsa, original magnification 1,000).

Cryoglobulins manifested in the present study as erroneously elevated WBC counts on the STKS counter. Interference on the platelet count was much less frequent and of mild importance on this machine. On the H*2, the apparent largest cryoglobulin particles (cases 1 and 4) interfered in WBC counts, and the smaller ones interfered in platelet counts (cases 2 and 3). Of interest, we noted that the WBC count as assessed in the basophil/lobularity channel was never affected by cryoglobulins. The main differential diagnoses of cryoglobulins are platelet clumps, giant platelets, nucleated RBCs, and chylomicrons, which also interfere in the left part of the
612

LMG diagram on the Coulter STKS counter. Interference on the perox channel plots on the H*2 also may correspond to platelet clumps or chylomicrons (the latter also are visible on the basophil/lobularity channel). RBC values generally are unaffected in cryoglobulinemias.10 Nevertheless, an apparent increment in hemoglobin because of decreased light transmittance in the measurement cuvette has been reported.11,17 According to our experience, cryoglobulin did not interfere in the hemoglobin measurement. We noted only 1 case of slightly reduced hemoglobin owing to a slightly decreased RBC count because of a flow
American Society for Clinical Pathology

Am J Clin Pathol 2002;117:606-614

Hematopathology / ORIGINAL ARTICLE

Image 5 (Case 1) Blood smear showing multiple basophilic inclusion bodies in neutrophils (May-Grnwald-Giemsa, original magnification 1,000).

disorder in relation to the presence of numerous thin precipitates in blood. Of practical interest, reliable automated counts can be obtained by warming the blood specimen at 37C (for at least 30 minutes) before analysis. If this method is not sufficient (partial redissolution occurs), collection of blood in warmed tubes and maintenance at 37C until analysis should be performed. We also report on the variable morphologic appearance of cryoglobulins in a fresh blood drop and on a blood film. The morphologic appearance of cryoglobulins in fresh blood samples has been described rarely. According to our experience, cryoglobulins mostly appear as thin, bright deposits (cases 2 and 3). Care must be taken to not mistake them for platelet clumps, particularly since EDTA-associated platelet

clumps may sometimes disappear at 37C. In practice, cryoglobulin precipitates are thinner and rougher than platelet clumps. We also observed 2 cases of crystal formation (cases 1 and 4). Crystallization has seldom been reported. In such cases, the morphologic appearance of the crystal was varied from needle-like, boat-shaped, cubic, flake-like, rhomboid, or hexagonal to rectangular.11,13-14 The coincidence of the appearance of long (50-150 m), thin crystals on blood films with the detection of a falsely elevated leukocytosis has been suggested.11,13 Only a few reports on the manifestations of cryoglobulins on blood films have been published.7,14,15,18 Indeed, according to our experience, May-Grnwald-Giemsastained blood films usually are normal. In addition, the blood film examinations were negative in 3 of 4 cases (cases 2-4). Yet, cryoglobulins may sometimes become visible as extracellular or intracellular material. Extracellular materials reported are small grayish precipitates,14 fusiform or needle-shaped crystals,11,13 faintly basophilic or pinkish globules, droplets, flakes, or shapeless aggregates.9,15 Only 7 cases of leukocyte cytoplasmic inclusions in cryoglobulinemia have been reported in the literature15,18 (plus 2 cases, personal communication, J.M. Jackson, reported in Lucas et al4). It has been demonstrated that cryoglobulin-containing plasma might induce the generation of neutrophil inclusions in neutrophils from a healthy donor.18 Cytochemical staining has been reported on blood films from 2 cases: the inclusions were negative for Sudan black B, oil red O, Gram stain, myeloperoxidase, and periodic acidSchiff. Ultrastructural and immunofluorescence studies have shown that the cytoplasmic inclusions in neutrophils corresponded to the cryoglobulin, likely phagocytosed by these cells.15,18 Morphologically, cryoglobulin inclusions should be distinguished from other neutrophil inclusions, such as Dhle bodies, abnormal mucopolysaccharides, and fragments of RBCs ingested in hemolytic anemias. Usual neutrophil cytoplasmic inclusions such as Dhle bodies,

Table 3 Guide for Screening and Diagnosis of Cryoglobulins


Guide 1. Key anomalies in automated WBC and/or platelet counts 2. Visualization of the cryoglobulins 2.1. Fresh blood sample (phase-contrast microscope) 2.2. May-Grnwald-Giemsastained blood film Comments Differential diagnosis: platelet clumps, giant platelets, nucleated RBCs, chylomicrons Before warming the blood Thin bright deposits or crystal of various shapes Differential diagnosis, platelet clumps Extracellular material or leukocyte cytoplasmic inclusions Examination is seldom positive Differential diagnosis, cytoplasmic neutrophil vacuoles, toxic granulations, fragments of ingested RBCs, Dhle bodies, abnormal mucopolysaccharides, Alder-Reilly anomaly, Chdiak-Higashi syndrome, pyknosis, and karyorrhexis In case of partial redissolution, keep the blood at 37C from venipuncture to analysis Blood collection, coagulation, and centrifugation must be performed at 37C

3. At 37C, disappearance of artifacts on blood cell counts 4. Biochemical analysis of the serum: isolation, purification, and immunochemical analysis

American Society for Clinical Pathology

Am J Clin Pathol 2002;117:606-614

613

Fohlen-Walter et al / LABORATORY IDENTIFICATION OF CRYOGLOBULINEMIA

Alder-Reilly anomaly, and Chdiak-Higashi syndrome are morphologically distinct.4 The principal cause of neutrophil vacuolization is bacterial septicemia, which may induce clear cytoplasmic neutrophil vacuoles, toxic granulations, and Dhle bodies. 18 The presence of a distorted neutrophil nucleus should not be mistaken for nuclear pyknosis or karyorrhexis. Of note, inclusions in monocytes from patients with cryoglobulinemia also have been found.18 The present study indicates that significant anomalies in automated blood cell counts may be related to the presence of cryoglobulin precipitates. The observation of such abnormal histograms must prompt microscopic examination of fresh blood samples and May-Grnwald-Giemsastained blood films, which may permit the visualization of specific features of these precipitates. In such cases, the clinician should be informed promptly, as such anomalies may be the first indication of the presence of a cryoglobulinemia.
From the 1Hematology Laboratory, and 2Biochemistry Laboratory, Centre Hospitalier Universitaire Nancy, Vandoeuvre, France. Address reprint requests to Dr Lesesve: Hmatologie biologique, CHU Nancy-Brabois, F-54511 Vandoeuvre, France. Acknowledgment: We thank Marie-Christine Bn, PhD, for careful review of the manuscript and helpful discussion.

References
1. Brouet JC, Clauvel JP, Danon F, et al. Biologic and clinical significance of cryoglobulins: a report of 86 cases. Am J Med. 1974;57:775-788. 2. Barnett EV, Bluestone R, Cracchiolo A. Cryoglobulinemia and disease. Ann Intern Med. 1971;73:95-107. 3. Gorevic PD, Kassab HJ, Levo Y, et al. Mixed cryoglobulinemia: clinical aspects and long-term follow-up of 40 patients. Am J Med. 1980;69:287-308.

4. Lucas FV, Miller ML, Savage RA, et al. Cryoglobulinaemia. Hematology. 1984;26:1-7. 5. Invernizzi F, Galli M, Serino G, et al. Secondary and essential cryoglobulinemias: frequency, nosological classification, and long-term follow-up. Acta Haematol. 1983;70:73-82. 6. La Civita L, Zignego AL, Monti M, et al. Mixed cryoglobulinemia as a possible preneoplastic disorder. Arthritis Rheum. 1995;38:1859-1860. 7. Lesesve JF, Goasguen J. Cryoglobulin detection from a blood smear leading to the diagnosis of multiple myeloma. Eur J Haematol. 2000;65:77. 8. Emori HW, Bluestone R, Goldberg LS. Pseudo-leukocytosis associated with cryoglobulinemia. Am J Clin Pathol. 1973;59:202-204. 9. Abela M, McArdle B, Qureshi M. Pseudoleucocytosis due to cryoglobulinaemia [letter]. J Clin Pathol. 1980;33:796. 10. Montecucco C, Chri-Lignire EL, Ascari E. Pseudoleukocytosis in essential mixed cryoglobulinemia may simulate chronic lymphocytic leukemia [letter]. Clin Exp Rheumatol. 1985;3:278-279. 11. Taft EG, Grossman J, Abraham GN, et al. Pseudoleukocytosis due to cryoprotein crystals. Am J Clin Pathol. 1973;60:669-671. 12. Haeney MR. Erroneous values for total white cell count and ESR in patients with cryoglobulinemia. J Clin Pathol. 1976;29:894-897. 13. Krause JR, Nitiyanant P, Rabin BS. Cryocrystalglobulinemia in hairy cell leukemia. Cancer. 1978;42:2798-2801. 14. Shah PC, Rao K, Noble V, et al. Transient pseudoleukocytosis caused by cryocrystalglobulinemia. Arch Pathol Lab Med. 1978;102:172-173. 15. Hambley H, Vetters JM. Artefacts associated with a cryoglobulin. Postgrad Med J. 1989;65:241-243. 16. Luzar MJ, Camisa C, Neff JC. Essential mixed cryoglobulinemia (type II) with pseudoleukocytosis. Arthritis Rheum. 1984;27:353-355. 17. Nosanchuck JS, Roark MF, Wanser C. Anemia masked by triglyceridemia. Am J Clin Pathol. 1974;62:838-839. 18. Maitra A, Ward PCJ, Kroft SH, et al. Cytoplasmic inclusions in leukocytes: an unusual manifestation of cryoglobulinemia. Am J Clin Pathol. 2000;113:107-112.

614

Am J Clin Pathol 2002;117:606-614

American Society for Clinical Pathology

Você também pode gostar