Platelet Satellitosis: A Case Report with Laboratory Characterization and Quantitative Evaluation*

Transcription

1 ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 9, No. 2 Copyright 1979, Institute for Clinical Science, Inc. Platelet Satellitosis: A Case Report with Laboratory Characterization and Quantitative Evaluation* GENE L. GULATI, Ph.D., BONG H. HYUN, M.D., D.Sc.f and MARY W OODALL, B.S., M.T.(ASCP)SH Department o f Pathology, Muhlenberg Hospital, Plainfield, N] and Rutgers Medical School, Piscataway, NJ ABSTRACT In an attem pt to understand better the phenom enon of platelet satellitosis (PS) and its pathogenesis and clinical relevance, in vitro and in vivo studies were performed. Quantitative evaluation of PS was carried out by (1) determ ining the com bined percentage of neutrophils and bands showing PS and (2) calculating the PS score in a m anner similar to the leukocyte alkaline phosphatase score. Among ethylenediam ine tetraacetic acid (EDTA)- treated, heparinized, citrated and non-anticoagulated blood samples, only films made from EDTA-treated blood revealed PS. By light microscopy, PS was characterized by adherence of platelets to neutrophils and bands. Examination of films prepared at tim ed intervals from EDTA-treated blood kept atroom tem perature (RT) showed maximal PS at one hour. A significant decline in PS ensued after the blood had stood at RT for four hours or longer. Day-to-day variations in the degree of PS followed a cyclic pattern with a cycle length of approximately 20 days, but they did not correlate with the patient s clinical status or any other laboratory findings. Results of in vitro studies suggested that the patient s plasma contained a substance which would induce p latelet satellitosis in normal EDTA anti coagulated blood and in buffy coats harvested from the patient s citrated blood. Over the past 25 years the phenom enon of platelet adherence to leukocytes, generally referred to as p latelet satellitosis * Presented in part in the Hematology Check Sample Program, CCE Council on Hematology, American Society of Clinical Pathologists (Hematology No. H-78, 1976). f Address all correspondence and reprint requests to: Bong H. Hyun, M.D., Department of Pathology, Muhlenberg Hospital, Plainfield, NJ (PS) or platelet satellitism, has been described in a small but clinically diverse group of patients.1-16 The exact nature of this phenom enon, its pathophysiology and clinical significance are not w ell understood. Recently PS was observed in the blood film of a hospitalized patient while the present authors performed the routine differential leukocyte count. In /79/ $01.20 Institute for Clinical Science, Inc.

2 110 GULATI, H Y UN AND W OODALL vivo and in vitro studies carried out on this patient in an attem pt to investigate the nature and the pathogenesis of PS are reported in this com m unication. Procedures developed and utilized to evaluate PS quantitatively are also detailed. Case Report A 67 year old, obese, white female was discharged from the hospital following treatment for cerebrovascular insufficiency, hypertension, diabetes mellitus with Kimmelstiel-Wilson syndrome, hypothyroidism and genitourinary tract infection (E. coli). She was readmitted one month later because of an episode of syncope followed by weakness. Physical examination revealed psoriatic lesions on the legs. Initial important laboratory findings were as follows: hemoglobin 11.0 g per dl (1.71 mmol per 1), total leukocyte count 12,000 per / I (12 x 109 per 1), with a differential count of 46 percent neutrophils, 18 percent bands, 21 percent lymphocytes, 4 percent monocytes, 8 percent eosinophils, 2 percent basophils and 1 percent metamyelocytes. Platelet satellitosis was noted on peripheral blood films. A platelet count performed one week after admission was 325,000 per /xl (0.325 x 1012 per 1). The blood glucose was 150 mg per dl (8.25 mmol per 1), urea nitrogen 26 mg per dl (9.3 mmol per 1), uric acid 6.6 mg per dl (0.39 mmol per 1), serum albumin 2.5 g per dl (0.39 mmol per 1), alkaline phosphatase 232 IU per 1, lactate dehydrogenase 240 IU per 1, and glutamic oxaloacetic transaminase (SGOT) 85 IU per 1. Urinalysis revealed more than 100 white cells per high-power field, moderate protein and many bacteria. During the course of hospitalization (10 weeks) the patient received a total of 26 drugs including diuretics, ataractics, analgesics, cardiotonics, anticoagulants, hormonal preparations, antibiotics, antihypertensive agents and insulin. One month after admission she developed gangrene of one foot requiring amputation. Subsequently the patient deteriorated, and died on the 76th hospital day. The major findings at autopsy included cerebral atherosclerosis with marked stenosis of the right internal carotid artery, acute infarction of the right cerebral hemisphere and cholangiocarcinoma with metastases to para-aortic and peri pancreatic lymph nodes. TABLE I Q u a n tita tiv e E v a lu a tio n o f P l a t e le t S a t e l l i t o s i s 3 core 0 N o n e F r a c t i o n o f C e l l C i r c u m f e r e n c e S u r r o u n d e d b y P l a t e l e t s 1+ O n e q u a r t e r 2+ O n e h a l f 3+ T h r e e q u a r t e r s 4+ E n t i r e Methods and Materials Q u a n t i t a t i v e E v a l u a t i o n o f P l a t e l e t S a t e l l i t o s i s Peripheral blood (PB) films prepared by th e c o n v en tio n al slid e m eth o d and stained with W right stain were evaluated for PS quantitatively by (1) determ ining the com bined percentage of neutrophils and bands showing platelet satellitosis (% CPS)and (2) calculating the PS score in a m anner similar to the leukocyte alkaline phosphatase score. As shown in table I, a scale of 0 (zero) to 4+ was used to score cells. On initial screening of some PB films, it was noted that the PS-positive cells (neutrophils and bands), though seen in both thick and thin areas of the film, were present in relatively larger numbers in the thick area. Therefore, a total of 400 neutrophils and bands combined (200 each from th e th in and th ick areas) w ere counted to obtain the %CPS and PS score per 100 cells. I n V iv o S t u d y D u rin g the course of 10 w eeks of hospitalization, the patient s peripheral blood films were studied and prepared routinely 19 times as part of the CBC. Each film was evaluated for %CPS and the PS score. These were then plotted for the entire 10 week period on linear graph paper. This study perm itted assessment of variations in the degree of PS under in vivo circum stances and of the relation w ith the clinical course and the pertinent laboratory findings. I n V it r o S t u d i e s S tu d y 1: Blood sam ples w ere collected from the patient by direct finger stick in ethylenediam ine tetraacetic acid (EDTA), in sodium citrate and in heparin. Film s w ere prepared by the conventional slide m ethod within 15 m inutes of sample

3 111 P L A T E L E T S A T E L L IT O S IS collection, stained w ith W right stain and evaluated for PS. S t u d y 2: P a tie n t s E D T A -tre a te d blood was allow ed to stand at room tem p erature (RT) up to a maximum of 28 hours. F ilm s p r e p a r e d from th is sa m p le at specified intervals (15 m inutes, 1, 2, 3, 4, 12 and 28 hours) w ere exam ined for PS after staining w ith W right stain. S t u d y 3: E D T A - a n tic o a g u la te d w hole blood from an ABO -com patible healthy subject and buffy coats isolated from patient s citrated blood w ere incu bated separately after mixing w ith either platelet-rich plasm a (PRP) or plateletpoor plasm a (PPP) harvested from pa tie n t s ED TA -treated blood. Incubations w ere carried out at room tem perature and at 37 C, some for 30 m inutes and others for two hours. Film s prepared from these in c u b a te d m ix tu re s w e re s ta in e d w ith W right stain and exam ined for PS. Results I n V iv o S t u d y total 19 PB films exam ined over the 70 day period, one show ed m arked PS (%CPS = 88, PS score = 268), 10 m oderate d eg ree of PS (%CPS = 20 to 50, PS score = 50 to 108), 4 m ild (%CPS = 8 to 13, PS score = 10 to 21), and the rem aining 4 revealed e ith e r an in s ig n ific a n t d e g re e o f PS (% CPS = 2 to 3, PS score = 3 to 4) or none at all. T he PB films w ith m oderate to m arked degrees of PS also revealed a few aggregates of PS-positive cells (figure 2). The PS score always paralleled the %CPS (figure 1). I n V it r o S t u d i e s A n tic o a g u la n ts a n d th e P h enom enon o f PS. Among the films prepared from n on-anticoagulated, E D T A -treated, ci trated and hep arin ized blood sam ples, only those prepared from ED TA -treated blood were found to be positive for PS. These involved m ainly neutrophils w ith a small num ber of bands but none of the other leukocytes (lym phocytes, m ono cytes, eosinophils, basophils, m yelocytes and metamyelocytes). V a ria tio n s in the D egree o f PS D u rin g H o sp ita l C ourse. The degree of PS as V a ria tio n s in th e D egree o f PS w ith T im e In V itro. R esu lts o f th is study m easured by % CPS and the PS score fol low ed a cyclic pattern w ith a cycle length of approxim ately 20 days (figure 1). O f the are illustrated in figure 3. A film prepared from ED TA -treated blood at 15 m inutes from the tim e of collection revealed a F i g u r e 1. Variations in the degree of platelet satellitosis noted in the film s prepared from E D T A -treated blood sam p les o f the p atien t over a period of 10 weeks of hospitalization Percentage of c e lls (n eu tro p h ils and bands) show ing platelet satellitosis (%CPS)... o Platelet satel litosis score (PS score). DAYS OF HOSPITALIZATION

4 112 G U L A T I, HYUN AND W O O D A L L F i g u r e 2. Peripheral b lood film sh ow in g marked degree o f plate let sa te llito sis and the tendency o f neutrophils with platelet satellitosis to aggregate (Wright stain, x 250). %CPS value of 45 and a PS score of 84. By one hour the degree of PS had reached the peak level w ith a %CPS of 78 and a PS score of 188. This level of PS was m ain tained for the subsequent two hours. It then appeared to decline, w ith the film prepared at four hours showing a %CPS of 60 and PS score of 143. Poor cell m orphol ogy encountered in films prepared at 12 and 28 hour intervals made it difficult to analyze the degree of PS, but examination of these films gave the im pression that the degree of PS had significantly and pro gressively declined. In-V itro Induction o f PS. A significant degree of PS (%CPS of 10 or higher, PS IN C U B A T I O N T I M E, H O U R S F i g u r e 3. Changes in the d eg ree o f p la telet sa tellitosis revealed by the examination o f films prepared at serial inter vals from the p a tie n t s E D T A -treated b lood sam ple fo llo w in g in cu bation at room tem per ature Percentage of c e lls (n eu trop h ils and bands) show ing platelet satellitosis (%CPS) o o P latelet satel litosis score (PS score), r Values at these points are rough estim ates, made necessary by poor cell morphology.

5 PLA T E LE T SATELLITOSIS 113 score of 14 or higher) was observed in films prepared from room tem perature incubated m ixtures containing p atien ts EDTA plasma (PRP or PPP) and either patient s citrated buffy coat or normal ED TA-anticoagulated whole blood. PS was not seen in films prepared from similar m ixtures incubated at 37 C. Discussion O ur findings th a t (1) only ED TA - treated blood revealed PS and (2) the phenom enon of PS involved prim arily neutrophils with small num ber of bands b u t none of the other leukocytes are essentially in accord w ith the earlier reports of Crom e and B arkhan,3 K jeldsberg,6 K jeldsberg and Sw anson,7 Prchal and Blakely,12 and Signy and G reen.15 The findings do not agree w ith F ield and MacLeod,4 who reportedly found PS also in film s o f c itrate d, oxalated and heparinized blood, or Ravel and Bassart,13 who observed PS in films prepared from nonanticoagulated blood in addition to those from EDTA-treated blood. Thus, the concept proposed by Field and Mac Leod4 that PS is probably an in vitro p h en o m en o n is su p p o rte d by sev eral3,6,7,15 but not all of the reports in the literature, particularly not by those5,12 indicating a parallel course betw een the phenom enon of PS and the clinical condition of the patient. That PS is a transient phenom enon is suggested by the observation that incubation of EDTA-treated blood at RT for four hours or longer resulted in significant d e cline in both the %CPS and the PS score. A marked increase in the degree of PS seen during the initial hour of incubation of the EDTA-treated blood at RT necessitated the preparation of films consistently at a preset interval of preferably one to three hours from the time of blood collection, for the study of day-to-day variations in the degree of PS. Results of the study carried out to m onitor such variations, utilizing PB films prepared usually one to two hours after blood collection, revealed a cyclic pattern in %CPS and the PS score with the cycle length being approximately 20 days. A similar pattern of about the same cycle length has been previously described by Morley for blood neutrophil concentration in healthy m en.11 W hether or not these two phenom ena are related in any way is not known. Variations in %CPS and the PS score did not correlate with the status of the patient or abnormal hematologic findings such as leukocytosis, neutrophilia, eosinophilia, shift to the left, or apparent throm bocytosis, as judged by scan of PB films. Additional abnormal laboratory findings the patient exhibited w ere elevated levels for blood glucose, urea nitrogen, alkaline phosphatase, glutam ic oxaloacetic transaminase as w ell as triglycerides (type IV hyperlipoproteinemia) and low levels of serum album in w ith normal total protein content. These determ inations and platelet counts were not performed frequently enough to be of value in establishing a relation betw een any of these findings and the phenom enon of PS. Attempts to relate the phenom enon of PS to the effect of a specific drug or group of drugs have also b een reported as unsuccessful.2,7 In our patient, such attempts were hindered by the fact that her com plete therapeutic regim en consisted of a long list of drugs (26 total). Results of our in vitro Study 3 indicate that PRP and PPP prepared from the patient s EDTA-treated blood were capable of inducing PS in normal EDTA-treated whole blood and in buffy coats harvested from the citrated blood of the patient at RT but not at 37 C. In similar studies carried out by Bolton and Boyd, it was found that cell-free EDTA-plasma from their patient (known to exhibit PS) induced PS in normal EDTA-treated blood to a greater degree at 4 C than at RT or at 37 C.2 Transfer of PS activity to normal whole blood by the plasm a of patients known to exhibit PS has also been reported by

6 114 GULATI, HYUN AND W O ODA LL M cg regor and associates.10 R ecently, Greipp and Gralnick have successfully dem onstrated transfer of PS activity to normal whole blood and buffy coat using serum at 37 C and 25 C.5 These reports and ours suggest that the plasma and/or serum of such patients contains a factor (we prefer to call it PS factor, PSF) which is capable of inducing PS in vitro. Studies designed to investigate the nature of PSF have b een few and have yielded variable results,2,5,10 PSF has been reported to be an immunoglobulin (IgG) by G reipp and Gralnick5 and a cryoprotein composed partly of cryofibrinogen by McGregor et al.10 From our own lim ited experim ental data, it can only be hypothesized that the phenom enon of PS is m ediated by PSF, which is probably a slow-acting humoral factor present in the plasma, that it somehow requires EDTA for its effect under in vitro conditions and that it is reactive at RT. That RT is probably the optimum tem perature for dem onstrating PS activity in vitro is also supported by studies12,15 in w hich investigators found a significantly lower degree of PS or none at all in EDTA-treated blood at 4 C, 28 C, 33 C and/or 37 C than at 20 C or 23 C. Further studies are necessary to determ ine the source, physicochemical nature and mechanism (s) of action of PSF. Other important findings pertaining to PS and deserving m ention here are that (1) PS causes spurious thrombocytopenia in vitro when the counts are performed with electronic particle counters,7 (2) PS may be an antibody-m ediated phenom enon and could be associated w ith throm bocytopenia in vivo5 and (3) clusters of PSpositive neutrophils are falsely interpreted as single large leukocytes with high peroxidase activity rep resen tin g im m ature or possibly abnorm al neutrophils by the Technicon Hemalog D.8 The infrequent platelet counts on our patient s blood were performed w ith a phase m icroscope soon after collection of the blood specim ens, thereby elim inating the possibility of encountering spurious throm bocytopenia. O ur attem pts to correlate the oscillations in the degree of PS w ith variations in the patient s condition have m et with unrew arding results, as have sim ilar attempts by many other investigators.2,4,7 Nevertheless, Prchal and Blakely in 1973 described a patient with long-standing Behget s disease, in whom the degree of PS paralleled the clinical course.12 More recently G reipp and Gralnick have d escribed two patients in whom the developm ent of throm bocytopenia and pulm onary distress could be directly related to the phenom enon of PS.5 The variety of clinical conditions represented by the patients reported to have shown P S 1,2,3, 5,6,7,8.9,12,13,15,16 js ^00 large even to attem pt to establish a relation betw een the phenom enon of PS and any one or more clinical diagnoses. It may be noted, however, that a significant num ber of these patients (including ours) suffered from throm bosis or throm botic ten d e n c y prior to, during and/or follow ing the period PS was observed in their blood films. A cause-effect relation betw een the throm botic problem s and the PS rem ains to be established. References 1. B a u e r, H. M.: In vitro platelet-neturophil adherence. Amer. J. Clin. Path. 63: , BOLTON, F. G. and B o y d, J.: Platelet adherence to polymorphs. Brit. Med. J. 2:747, C r o m e, P. E. and B a r k h a n, A.: Platelet adherence of polymorphs. Brit. Med. J. 2:871, F i e l d, E. J. and M a c L e o d, I.: Platelet adherence of polymorphs. Brit. Med. J. 2:388, G r e ip p, P. R. and G r a l n i c k, H. R.: Platelet to leukocyte adherence phenomena associated with thrombocytopenia. Blood 47: , KjELDSBERG, C. R.: Platelet satellitism. N. Eng. J. Med. 290:165, K jeld SBERG, C. R. and SWANSON, J.: Platelet satellitism. Blood 43: , LARSON, J. H. and PIER R E, R. V.: Platelet satellitism as a cause of abnormal Hemalog D differential results. Amer. J. Clin. Path. 68: , 1977.

7 PLA T E LE T SATELLITOSIS M c D o n a l d, G. A., D o d d s, T. C. and C r u ic k - SHANK, B.: Atlas of Hematology, 3rd ed. Baltimore, Williams and Wilkins, 1970, p M c G r e g o r, D. H., Liu, P. I., D a v is, J. W., e t a l : Platelet satellitism: Clinical, ultrastructural and experimental studies suggesting its etiology. Amer. J. Path. 78:64a, MORLEY, A. A.: A neutrophil cycle in healthy individuals. Lancet 2: , P r c h a l, J. T. and B l a k e l y, J.: Granulocyte platelet rosettes. N. Eng. J. Med. 289:1146, R a v e l, R. and B a s s a r t, J. A.: Platelet satellitosis and phagocytosis by leukocytes. Lab. Med. 5:41-42, R e is m a n, L. E., S a b e s in, S. M., G o r d o n, G., e t AL: Platelet satellitism: A phenomenon involving platelet phagocytosis by polym orphonuclear leukocytes. Clin. Res. 2:586, SlGNY, A. G. and G r e e n, A. E.: Platelet adherence to polymorphs. Brit. Med. J. 2:624, S i l b e r g l e i t, A.: A study of platelet-leukocyte aggregation in coronary thrombosis and cerebral thrombosis. Thromb. Diath. Haemorrh. Supplement 42: , 1970.