erythrocyte sedimentation rate during steady state and painful crisis in sickle cell anemia

Erythrocyte Sedimentation Rate During Steady State and Painful Crisis in Sickle Cell Anemia

CHRISTINE LAWRENCE, M.D. MARY E. FABRY, Ph.D. Bronx, New York

From the Division of Hematology, Department of Medicine, Albert Einstein College of Medicine, and the Bronx Municipal Hospital Center, Bronx, New York. This work was supported in part by Program Grant HL-21016 and Grant RR&O- GCRC from the National Institutes of Health. Re- quests for reprints should be addressed to Dr. Christine Lawrence, Jacobi Hospital, Room 314, Pelham Parkway, Bronx, New York 10461. Man- uscript submitted September 10, 1985, and ac- cepted October 11, 1985.

To define its diagnostic utility in sickle crisis, the erythrocyte sedimentation rates of oxygenated blood were studied in patients with sickle cell anemia and healthy normal subjects using the Guest-Westergren method. A normal range for sedimentation rate as a function of hematocrit was established in 22 normal subjects. Twenty-seven asymptomatic patients with sickle cell anemia had abnormally low sedimentation rates in relation to their hematocrits. Those low sedimentation rates were not increased by substituting plasma from healthy control subjects, which suggests that the low sedimentation rate was a cell- related phenomenon. Sedimentation rates measured in 28 patients with sickle cell anemia at the end of uncomplicated painful crisis increased to levels appropriate for their degree of anemia. In patients with sickle crisis and medical complications, the sedimentation rates were even higher. At the end of an uncomplicated painful crisis, the mean plasma fibrinogen level was significantly higher than at the onset (p <O.OOS). When red cells from patients with sickle cell anemia at the end of crisis were suspended in normal plasma from control subjects, the sedimentation rates remained high. it is concluded that the erythrocyte sedimentation rate of asymptomatic patients with sickle ceil anemia is abnormally low due to cellular factors, and the increase during painful crisis is due primarily to red cell changes, modified by plasma factors.

Patients with sickle cell anemia have unusually low etythrocyte sedimentation rates, particularly when the anemia is taken into account [l-4]. The sedimentation rates of normal persons are affected by several red cell and plasma factors that depend on their state of health, but little is known about why sedimentation is abnormal in patients with sickle cell anemia.

Acute-phase reactants, such as fibrinogen, but also elevated levels of gamma globulins, increase the erythrocyte sedimentation rate [5-lo]. The elevation of the fibrinogen level reported to occur during painful sickle cell crises [8,9] would be expected to increase the sedimentation rate during crisis if these patients’ red cell response to fibrinogen was similar to that ih normal persons. The observations that deoxygenation retards red cell sedimentation in patients with sickle cell anemia, and that deoxygenated sickle cells do not aggregate or form rouleaux [2, I 1- 131, suggest that cellular factors are important determinants of sedimentation rates in these patients.

The effect of painful sickle crisis on the erythrocyte sedimentation rate, either with or without medical complications, has not been system- atically investigated. In this report, we examine the effect of painful crisis, with and without medical complications, on the erythrocyte sedimentation rate of patients with sickle cell anemia and the relative contributions of cellular and plasma factors to the findings.

November 1986 The American Journal of Medicine Volume 81 801

SEDIMENTATION RATE IN SICKLE CELL ANEMIA-LAWRENCE and FAIRY

15Or f@ 0

130 -

120-

110 -

100 -

90 -

80 -

70 -

60 -

50 -

40 -

30 -

20 -

IO -

IO 20 30 40 50 Hct 1%)

igun 1. Erythrocfle sedimentation rates (ESR) by the uest-Westergren method in 22 normal subjects and 27 G

patients with sickle cell anemia. The normal subjects (women 0, men 0) had hematocrits (Hcf) between 39 and 50 percent. .The lower hematocrits in their blood samples tiere obtained by diluting with, autologous PI&ma. The bounda* llires enclose the values in native and diloied normal blood samples. The solid symbols show the sedL mentation rates in undiluted blood of 23 patients with sickle cell anemia (A) and four patients with sickle$” thalassemia(t).

PATIENTS AND METHODS

For measurement of blood counts and erythrocyte sedimentation rates, venous blood specimens were collectad in 7 ml dry ethylene diamine totraacetate (EDTA)-containing Va- cutainer tubes. For red cell density gradient determinations, blood was collected in dry heparinized Vacutainer tubes. The blood specimens were obtained with informed consent from 50 homozygous patients with sickle cell anemia and four patients with sickle@’ thaiassemia. The findings were essentially the same in both groups so we will refer to all of them as having sickle cell disease or disorders. The patients ware classified into five groups. Group I patients had no symptoms or signs of a crisis or complicating illness and were seen for a routine follow-up examination. Group 2

patients had the onset less than 72 hours earlier, of a typical painful crisis requiring hospitalization. Group 3 patients were heterogeneous; most were seen repeatedly in the emergency room with recurring mild to moderate pain that did not require hospitalization; most of them were probably addicted to narcotics. Group 4 patients were hospitalized for a painful crisis that lasted four to 14 days; the laboratory data given herein correspond to their peak erythrocyte sedimentation rates. Group 5 patients had major infiamma- tory illnesses together with painful crises. Nine of the Group 5 patients had serious infections (six bacterial and three viral) and four had major thromboemboiic problems. The normal control subjects were healthy nonanemic subjects, 12 male and 10 female. They were physicians and medical students, aged 21 to 57 years (average 33 years). Plasma samples were stored at -2O’C until used. All heterologous blood samples that were mixed were typed, cross-matched, and compatible. Blood counts were performed using a Coulter Couhter S Plus IV.

All sedimentation rates in native specimens were measured within 12 hours of drawing blood, during which the sedimentation rate is stable when EDTA is the anticoagu- lant [ 141. To minimize sickling, 2 ml of well-mixed EDTA- blood was placed in IO ml stoppered tubes and rotated on a wheel at 10 RPM for 15 minutes before being placed in the sedimentation rate tubes. That aeration uniformly produced at least 97 percent oxygen saturation (Corning 178 pH/ Blood Gas Analyzer, Corning Medical and Scientific, Med- field, Massachusetts). Sedimentation rates were measured at one hour with the Guest-Westergren method, using 230 mm long plastic tubes with an internal diameter of 2.55 mm (manufactured by Guest Medical and Dental Products, Zug, Switzerland, and distributed by American Scientific Prod- ucts, McGraw Park, Illinois) [ 151.

in experiments in which sickle and normal red cells and plasma were mixed, the ceils were centrifuged at 1,200 RPM for two minutes. The supernate plasma was removed, and the ceils were washed twice with 0.9 percent sodium chloride before being rinsed, centrifuged, and then resus- pended in autologous or compatible plasma. The hematocrit was adjusted to within f 5 percent of that in the ‘donor cell’ blood. Measurements of serum total protein, albumin, total globulins, quantitative IgG, IgA, and IgM, and plasma fibrinogen were performed in the hospital’s special chemis- try laboratory by standard methods.

Mean values are reported f 1 SD and the significance of differences was estimated with a two-tailed t test for un- paired or paired values as appropriate. Statistical analysis was performed on an Apple II computer using the program Statistics with Daisy (Rainbow Computing, Torrance, Caii- fornia).

RESULTS

To provide a reference for the effect of anemia on the Guest-Westergren method, we measured the sedimentation rate in 22 normal nonanemic control subjects, in undiluted blood, and in specimens diluted with autologous plasma to yield hematocrits down to 6 percent. Figure 1 depicts the sedimentation rates in 22 control subjects.

802 November 1986 The American Journal of Medicine Volume 81


Flgure 2. Erythrocyte sedimentation rates (ES@ in three pairs of blood group- compatible patients with sickle cell an.+ mia and normal subjects during progressive dilution or concentration with autologous sick/e (--) or normal (- - - -) plasma. Normal red cells (a). sickle red ceils (A). Hct = hematocrit.

160

140

120

A t

t \ \ I i l

‘\ \ . l

‘\ \

‘\\ \ . ’

‘, i

\ ‘. “ . *

- & - - -

IO 20 30 40

r El

IO 20 30 40 Hct(%)

r C

80

70 F

0 0

60 c

Figure 3. Sequential sedimentation rates (ESR) during 23 uncomplicated painful sickle crises. Initial values: he mozygous sickle cell disease ( l ), sickle- p” thalassemia ( + ); follow-up values: 1 ’ sickle cell (0), sickle-thalassemia (0). a

The rates in undiluted blood ranged from 0 to 15 mm per hour (mean 5.7 f 4.4 mm per hour) and dilution progressively increased the rate. The contrasting low sedimentation rates in 27 Group 1 and 2 patients with sickle cell disorders at their native hematocrits are also illustrated in Figure 1. These patients’ sedimentation rates were between 0 and 4 mm per hour, well within the “normal range,” but clearly below the values in diluted control blood specimens with comparable hematocrits.

Figure 2 depicts the findings in cell/plasma mixing experiments in three pairs of blood type-compatible asymptomatic patients with sickle cell anemia and normal subjects (hemoglobin AA). Progressive dilution of saline- washed normal red cells with either their own plasma, or compatible plasma from a patient with sickle cell anemia in steady state, resulted in similar increases in sedimenta-

2 3 4 5 6 7 8 9 IO II 12 13 ! DAY OF CRISIS

tion rates. In contrast, in the reverse experiment, when saline-washed sickle cells were diluted with their own or compatible normal plasma, the sedimentation rates remained abnormally low (less than 2 mm per hour), even at hematocrits as low as 8 percent.

Figure 3 shows the serial sedimentation rates in 23 patients with sickle cell disorders between one and 14 days after the onset of an uncomplicated painful crisis. The sedimentation rates increased progressively without a significant decrease of hemoglobin level or hematocrit. The mean hemoglobin concentration at the beginning of crisis was 9.1 f 1.5 g/dl, and at the time of peak sedimentation rate was 8.9 f 1.6 g/dl. The mean hematocrit at the beginning of crisis was 26.9 f 4.7 percent, and at the time of peak sedimentation rate was 26.0 f 4.7 percent. Sedimentation rates increased an average of



50

40

30

20

IO

IO 20 30 40 50 Hct (%I

i 7gure 4. Peak sedimentation rates (ESR) during uncom

plicated and complicated painful sickle crises. The solid symbols show 28 uncomplicated crises (Group 4): homozygous patients with sickle cell disease (A), patients with sickle-Do-thalassemia ( + ). The open symbqls show 14 crises with medical complications (Group 5): patients with sickle cell disease (A), patients with sickle-thalassemia ( 0 ). The shaded areas shows the range in native and diluted normal blood (Figure 1). Hct = hematocrit.

4.2 mm per day of crisis (p <O.OOl), as estimated by linear regression analysis. The percent of the variance of the daily increase in sedimentation rate determined by the correlation was 42 percent (i.e., r* = 0.42).

Figure 4 depicts the peak sedimentation rates in two groups of 42 patients with sickle cell disorders who were followed through uncomplicated painful crises (Group 4, 28 patients) or crisis with other intercurrent illnesses (Group 5, 14 patients), compared with the normal range of sedimentation rate versus hematocrit (Figure 1). In uncomplicated crises, the sedimentation rates rose to values that were within the normal range for subjects with hemoglobin A. The sedimentation rates in the 14 patients

with major medical complications rose above that “normal range. ’ ’

Figure 5 illustrates the findings in two of six studies in which end-of-crisis saline-washed sickle cells were diluted with either normal or autologous plasma. Dilution of the sickle celfs with normal plasma (from subjects with hemoglobin A) resulted in progressively increased sedimentation rates that fell within the lower portion of the normal curve. That contrasted sharply with the failure of dilution to affect the sedimentation of precrisis sickle cella (Figure 2). When the end-of-crisis autologous sickle plasma was used to dilute the end-of-crisis sickle cells, the sedimentation rates were even higher, consistent with the increased fibrinogen in that plasma.

Durjng the early days of crisis, there was a zone of “fuzziness” in the plasma just ab.ove the packed red cells. When the sedimentation rate reached its maximum,

lop -

go- 1

e E 80-

z 70- E

60 -

IO 20 30 40 50 Hct (%I

Flgure 5. Erythrocyte sedimentation rates (ES!?) using end-of-crisis red cells of two patients with siakle cell- anemia (0 and A) diluted with compqtible normal plasma (- - - ), and with the patient’s own end+f-crisis plasma (--). Hct = hematocrit.


SEDIMENTATION RATE IN SICKLE CELL ANEMIA-LAWRENCE and FABRY

TABLE I Serum Total Protein, Albumin, Globulin, and Plasma Fibrinogen Concentrations in Sickle Cell Disorders”

Total Protein? Albumin Globulin Plasma Fibrinogen Wdl) WI) WI) @WI)

Normal control subjects (15) 301.9 f 38.1 Patients with sickle cell disease

Group 1 (8)x 7.36 f 0.40 4.53 f 0.45 2.84 f 0.43 (11) 287.3 f 52.7 Group 2 (19) 7.17 z!z 0.62 4.17 f: 0.45 3.00 f 0.42 (7) 331.4 f 36.5§ Group 3 (17) 7.59 f 0.50 4.36 f 0.34 3.29 f 0.39tt (16) 374.8 f 53.0tl Group 4 (16) 7.17 f 0.63 4.04 f 0.34* * 3.13 f: 0.49 (11) 476.4 f 123.8tt: Group 5 (7) 7.63 f 0.66 3.84 f 0.43* * 3.79 f 0.72” * (13) 655.9 f 130.8$$

* The numbers in parentheses refer to numbers of patients. Values are given as the mean f I SD. + None of the intergroup means are significantly different. 1 Group 1, asymptomatic; Group 2, admission with early crisis; Group 3, repeated emergency room visits; Group 4, end of crisis; Group 5, end of crisis with medical complications. 5 p versus Group <0.05 1. * * p <O.Ol versus Group 1. t+ p <O.OOl versus Group 1. U p <O.OOl versus Group 1 and normal subjects.

TABLE II Serum lmmunoglobulin Concentrations in Sickle Cell Disorders*

W (ma/dW

W (maIdlIt

MM (ma/dl)$

Normal control subjects (13) 1,024.6 f 196.1 154.6 f 69.3 170.2 zt 89.3 Patients with sickle cell anemia

Group 1 (7) 1,522.g f 446.4 (p <0.02)+ 319.6 f 170.1 (p <0.05) 117.7 f: 74.6 Group 2 (12) 1,508.7 f 282.3 (p <O.OOl) 325.5 f 172.5 (p <O.Ol) 152.4 f 135.0 Group 3 (14) 1,773.6 f 324.2 (p <O.OOl) 349.9 f 120.6 (p <O.OOl) 117.8 f 48.0 Group 4 (8) 1,728.8 f 393.4 (p <0.005) 309.8 f 176.5 (p <0.05) 126.4 f 31.4 Group 5 (7)s 2,168.3 f 1,122.3 (p <0.05) 410.8 Jo 219.5 (p <0.05) 116.6 f 38.0

’ The numbers in parentheses refer to the numbers of subjects. Values are given as the mean f 1 SD. + There were no significant differences between any of the sickle cell groups. P values compare the sickle cell group means with the normal control values. t There were no significant differences between any of the groups. § The mean IgA level in Group 5 is derived from six values.

there was a sharp separation between the plasma and the red cells, just as in nonsickle blood with a high sedimentation rate.

Tables I and II summarize the serum protein and plasma fibrinogen data in our patients. There were no significant differences in serum total protein between any of the groups (Table I), but the mean serum albumin level was significantly lower in the peak-crisis patients (Group 4), and in those with other illnesses (Group 5), than in the asymptomatic Group 1 patients (p <O.Ol). Mean serum total globulin was significantly higher in the Group 5 patients (3.8 g/dl) than in the Group 1 asymptomatic patients (2.8 g/dl, p <O.Ol). Mean plasma fibrinogen level was significantly higher in all the symptomatic patients with sickle cell anemia compared with Group 1. The mean fibrinogen concentrations in Groups 1 and 2 were not significantly different from the normal mean, but the mean values in Groups 3, 4 and 5 were significantly higher. Plasma fibrinogen level increased during uncomplicated crises from 333 f 34 mg/dl (Group 2) to 476 f

124 mg/dl (Group 4, p <O.OOS). Figure 6 depicts the relation between sedimentation rates and plasma fibrinogen concentrations in our five groups of patients with sickle cell anemia. There were no significant differences between any of the sickle cell groups with respect to their levels of serum IgG, IgA, or IgM (Table II), but the mean IgG and IgA levels in all the sickle cell groups were significantly higher than in the control subjects. Although the differences were not statistically significant, mean IgG and IgA levels in the Group 5 patients were higher than in the other patients.

COMMENTS

The erythrocyte sedimentation rate is affected by the hematocrit [ 16,171 and other red cell and plasma factors [5-7,181. Red cell sedimentation is thought to be partially determined by factors affecting aggregation, such as cell deformability [l l-131, surface charge [ 19,201, and the presence of proteins capable of neutralizing surface



120 . . .

TA a+ A 20 +++ ++

+A + + A L I I rdw~~ e

2;O I I I I I

100 300 400 500 600 700 800 900 FIBRINOGEN (mg/dl)

charge [2 l-231. Heterogeneity of red cell size and shape tends to slow red cell sedimentation, presumably by inter- fering with rouleaux formation [18]. Increased red cell rigidity, as in iron-deficiency anemia [ 171, slows the sedimentation rate, perhaps also by inhibiting red cell aggregation.

The magnitude of the effect of hematocrit on sedimentation rate varies with the method used for measurement. We defined a normal range of sedimentation rates as a function of hematocrit for the Guest-Westergren method for a group of 22 healthy persons homozygous for hemoglobin A, both at their native hematocrit and after dilution with autologous plasma (Figure 1).

If hematocrit (or anemia) is considered, the erythrocyte sedimentation rate in patients with steady-state sickle cell anemia is inappropriately low (0 to 4 mm per hour) (Figure 1) [i-4]. This effect has usually been ascribed mainly to interference with rouleaux formation by the abnormalities and variations of red cell size, shape, and rigidity [2,1 l- 13,181. Deoxygenation retards sedimentation in sickle cell anemia, and deoxygenated sickle cells fail to form rouleaux or aggregates [2,1 l-131. This factor was mini- mized in our studies because all our blood specimens were oxygenated before their sedimentation rates were measured.

In addition to anemia, the sedimentation rate is also influenced by plasma constituents. Albumin retards, and hypoalbuminemia permits, more rapid erythrocyte sedimentation [ 10,241. In contrast, large asymmetric mole- cules tend to accelerate cell sedimentation [21-231. At least part of their effect may be attributable to masking the negative charges on the red cell membrane, which nor-

Figure 6. Plasma fibrinogen and peak erythrocyte sedimentation (ESR) values in the five groups of patients with sickle cell anemia. Group 1, asymptomatic ( l ); Group 2, early crisis (0); Group 3, repeated emergency room visits (+); Group 4, end of crisis (A); Group 5, end of crisis with medical complications (A).

mally inhibit rouleaux formation. Fibrinogen has a striking accelerant effect [5-7, lo] and is largely responsible for the very high sedimentation rates found during pregnancy and various acute infectious and inflammatory disorders [25,26]. Increases of immunoglobulins also accelerate the sedimentation rate, and are thought to be responsible for the increased red cell rouleauxing and sedimentation in dysproteinemias such as multiple myeloma and macro- globulinemia [IO].

That the low sedimentation rates in blood from asymptomatic patients with sickle cell anemia were due to cellular factors was shown by suspending their cells in plasma from normal persons, and normal cells (hemoglobin AA) in plasma from patients with sickle cell anemia (Figure 2). Even at very low hematocrits, sickle cells failed to sediment in either sickle or normal plasma, and sickle plasma did not retard sedimentation of normal cells. Indeed, in one case (panel A), in which the patient with sickle cell anemia had a higher fibrinogen level (435 mg/ dl) than the normal subject (343 mg/dl), more rapid sedimentation of the diluted normal cells was observed.

The changes in the erythrocyte sedimentation rate during painful crises in sickle cell anemia were not previ- ously well documented. Limited data suggested that it increased [4], and several investigators have shown that plasma fibrinogen levels rise during crises [8,9]. For the purpose of a systematic study of sedimentation rate in this disease, we classified the patients with sickle cell anemia into five groups defined in Patients and Methods. The sedimentation rate in patients with sickle cell anemia increased as the crisis progressed (Figure 3). In patients without complicating illnesses, that increase occurred



without a significant change in hemoglobin level or hematocrit, and their peak sedimentation rates fell within the normal range for their hematocrit (Figure 4).

At the end of six uncomplicated crises, sickle cells were diluted with their own plasma or with the compatible plasma of a healthy person homozygous for hemoglobin A. All sedimentation rates fell within the normal range corrected for hematocrit (Figure 5), demonstrating that changes in the red cells and/or red cell population were responsible for the increased sedimentation rate. The sedimentation rates were higher when the end-of-crisis cells were diluted with their own (high fibrinogen) plasma (Figure 5). That suggests that although the increase in sedimentation rate during crisis was due primarily to a change in the cells, it was also partly due to a plasma change. Plasma fibrinogen is a potent aggregator of red cells, promoting their sedimentation. Several other studies have shown that plasma fibrinogen levels increase during painful crises [8,9], and our findings confirm that (Figure 6, Table I). When medical complications accompanied crises (Group 5), the mean fibrinogen levels were quite high, 656 f 131 mg/dl (p <0.005 versus Group 4).

Total serum globulin was also significantly increased in the Group 5 patients with medical complications (mean 3.79 f 0.72 g/dl) compared with the asymptomatic Group 1 patients (mean 2.84 f 0.43 g/dl, p <O.Ol, Table I). However, the mean concentrations of the individ- ual major immunoglobulin groups did not differ significantly among the various sickle cell groups (Table II). All the groups had significantly higher mean serum IgG and IgA concentrations than the normal subjects. Perhaps that reflects their repeated tissue injury during crisis and their frequent infections, and the latter may explain the higher

globulin levels in Group 3, which included a higher per- centage of addicted patients (Table I).

In conclusion, in well-oxygenated samples of blood from patients with sickle cell anemia in steady state, sedimentation rates are inappropriately low when hematocrit is taken into account. The low sedimentation rate is primarily due to cellular factors. During uncomplicated painful crisis, the sedimentation rate increases as a result of at least two changes from the steady state: (1) the modification in the red cell composition is alone sufficient for the erythrocyte sedimentation rate to become appro- priately elevated for the degree of anemia (Figure 4), and (2) the elevated plasma fibrinogen level (Table I) also contributes to the peak sedimentation rate. In patients who have other complicating illnesses, increases of serum globulin level also contribute to the increased erythrocyte sedimentation rate [24,25].

The finding of an elevated erythrocyte sedimentation rate in a patient with sickle cell disease suggests that the patient has, or has recently had, a crisis or a chronic or intercurrent illness, and is not in a steady state. In our experience, the finding of a persistently low etythrocyte sedimentation rate (less than 5 mm per hour measured in well-oxygenated blood) and a normal plasma fibrinogen level in a patient with sickle cell disease suggests that the patient has not had a recent significant painful crisis.

ACKNOWLEDGMENT

We thank Dr. Ronald L. Nagel for his helpful suggestions and critical review of the manuscript, and Rose Grossman and Joyce Johnson for their expert technical assistance. The cooperation of the medical house staff officers and laboratory personnel at Jacobi Hospital was greatly ap- preciated.

REFERENCES

1. Wintrobe MM, Landsberg JW: A standardized technique for 8. Gordon PA, Breeze GR, Mann JR, Stuart J: Coagulation the blood sedimentation test. Am J Med Sci 1935; 189: fibrinolysis in sickle-cell disease. J Clin Pathol 1974; 27: 102-114. 485-489.

2. Bunting H: Sedimentation rates of sickled and non-sickled 9. Richardson SGN, Breeze GR, Stuart J: Hyperfibrinogenae- cells from patients with sickle cell anemia. Am J Med Sci mia and hyperviscosity in sickle-cell crisis. J Clin Pathol 1939; 198: 191-193. 1976; 29: 890-893.

3. Diggs LW, Bibb J: The erythrocyte in sickle cell anemia. 10. Talstad I, Haugen HF: The relationship between the erythro- JAMA 1939; 112: 695-700. cyte sedimentation rate (ESR) and plasma proteins in

4. Abel CC, Beier L: Erythrocyte sedimentation rate in various clinical materials and models.. Stand J Clin ‘Lab Invest hemoqlobinopathies. Am J Med Sci 1961; 242: 463-467. 1979: 39: 519-524.

5. Fahraeus R: The suspension-stability of the blood. Acta Med Stand 1921; 55: 1-228.

6. Gilligan DR, Ernstene AC: The relationship between the erythrocyte sedimentation rate and the fibrinogen content of plasma. Am J Med Sci 1934; 187: 552-556.

7. Ropes MW, Rossmeisl E, Bauer W: The relationship between ESR and plasma proteins. J Clin Invest 1939; 18: 791-798.

11. Winsor T, Burch GE: Rate of sedimentation of erythrocytes in sickle cell anemia. Arch Intern Med 1944; 73: 41-52.

12. Usami S, Chien S, Scholz PM, Bertles JF: Effect of deoxygenation on blood rheology in sickle cell disease. Micro- vast Res 1975; 9: 324-334.

13. Jan KM, Usami S, Smith JA: Influence of oxygen tension and hematocrit reading on ESRs of sickle cells. Arch Intern Med 1981; 141: 1815-1818.

November 1986 The American Journal of Medicine Volume 81 a07


14. Gambino SR, DiRe JJ, Monteleone M, Budd DC: The Wester- gren sedimentation rate, using Ks EDTA. Am J Clin Pathol 1965; 43: 173-180.

15. Niejadlik DC, Engelhardt C: An evaluation of the Guest method for determining erythrocyte sedimentation rate. Am J Clin Path01 1977; 68: 766-768.

16. Bull BS, Brecher G: An evaluation of the relative merits of the Wintrobe and Westergren sedimentation methods in- cluding hematocrit correction. Am J Clin Pathol 1974; 62: 502-510.

17. Poole JCF, Summers GAC: Correction of ESR in anaemia. Br Med J 1952; 1: 353-356.

18. Phear D: The influence of erythrocyte factors on their sedimentation rate. J Clin Pathol 1957; IO: 357-359.

19. Chien S, Jan JM: Red cell aggregation by macromolecules: roles of surface absorption and electrostatic repulsion. J Supramolecular Structure 1973; 1: 385-409.

20. Jan KM, Chien S: Role of surface electric charge in red blood cell interactions. J Gen Physiol 1973; 61: 638-654.

21. Chien S, Luse SA, Jan KM, Usami S, Miller LH, Fremownt H: Effects of red cell suspensions. In: Dietzel J, Lewis D, eds. Proceedings of the Sixth European Conference on Micro- circulation. Aalborg, Denmark, 1970. Basel: Karger Press, 1971; 29-34.

22. Chien S, Jan KM: Ultrastructural basis of the mechanism of rouleaux formation. Microvasc Res 1973; 5: 155-160.

23. Kernick D, Jay AW, Rowlands S, Skibo L: Experiments on rouleaux formation. Can J Physiol Pharmacol 1973; 51: 690-699.

24. Gray SB, Mitchell EB: Effect of purified protein fractions on sedimentation rates of erythrocytes. Proc Sot Exp Biol Med 1942; 51: 402-406.

25. Zacharski LR, Kyle RA: Significance of extreme elevation of erythrocyte sedimentation rate. JAMA 1967; 202: 116- 118.

26. Wyler DJ: Diagnostic implications of markedly elevated erythrocyte sedimentation rate: a reevaluation. South Med J 1977; 70: 1428-1430.


erythrocyte sedimentation rate during steady state and painful crisis in sickle cell anemia

Documents