Adverse Outcomes in Sickle Cell Anemia in Brazil

Heloisa Helena Arantes Gallo da Rocha*

E mail: heloisag@ism.com.br

·         Hematologist  from Serviço de Hemoterapia  do Hospital dos Servidores do Estado do Rio de Janeiro

·          Technical Counselor and Foundation of  “Associação dos Falcêmicos do Rio de Janeiro.”

·         Fellow of International Society of   Hematology

·         Member of American Society of Hematology

·         Member of  Sociedade Brasileira de Hematologia e Hemoterapia

·         Member of  Colégio Brasileiro de Hematologia

Area: Hematology










The author does an analytical study of the Brazilian patients with sickle cell disease (SCD).  She makes a comparison between the data with those described by Miller. She concludes that:

1)-The most common bad outcome variable is repetitive painful episode- more than twice a year needing Medical care

2) - The hemoglobin level  of Brazilian- SCD patients  is lower than those from Miller’s study. (Both obtained from   steady state. patients)

3) There was no difference statistically significant between the levels of hemoglobin in those SCD patients with or without bad prognosis variants (p= NS) 

4) The survival time was shorter in patients presenting systematically low levels of hemoglobin (lessen than 7 g/dl).

5) The patients who presented dactylites before their first anniversary   had repetitive painful crisis in their childhood and adulthood life. If these people are left to their own natural history the survival time will be shorter than the patients who do not have these complaints.

5) Children with platelets levels higher than 500 000/mm3   probably will have acute spleen sequestration syndrome.

6)  We noticed that SCD patients who had strokes had more repetitive painful crises (61%) and Thrombocytosis   (38, 4%).

7) The SC P who presented repetitive painful crises probably will have higher levels of platelets counts or vice- versa.

8) The high reticulocyte counts in SCD patients (> 10%) in steady state clinical situation is consistent with low survival time.

9) - The worse prognosis variables the sickle cell patients have the shorter is their survival.

Keywords; sickle cell disease, prognosis.


The sickle cell anemia (SCA) is a hereditary hemolytic anemia.  The polymerization of hemoglobin S (HbS) in the deoxygenated state in sickle cell disease (SCD) is the watershed event leading to vase- occlusion, factors that retard the transit time of sickle erythrocytes in the microcirculation and the cell’s adhesive properties. The change in shape is responsible for the enhancing of adhesion in endothelial cells and vase occlusion impeding the blood flux to tissues (1) .The vascular occlusion is the result of a fall in oxygen tension in cells presenting a major density (2). The researches in electronic microscopy had been shown that the irreversible sickle cells does not have almost anything of polymerized hemoglobin and so still retains the sickle shape. The sickle shape occurs because the protein fixation in membrane cytoskeleton and they must not be responsible for the irreversibility shape of sickle cell. (3). these processes are dependent of sickle hemoglobin concentration. Thus, the occurrence of dense, dehydrated, take a deleterious effect in sickle cell process. (4)

In the afro descendant population of Brazil the prevalence of sickle cell trait is about 8% (5). Those people with the sickle cell trait have 40% of S hemoglobin in electrophoresis and they are free of symptoms.

In the white population we have 3% of positivity for sickle cell trait (5).

In SCD, besides the chronic anemia some major complications of the disease itself include acute thoracic syndrome (AST), stroke, spleen and kidney dysfunction, painful crisis and priapism. The children with SCD are more susceptible to present bacterial infection (6).

 35% of SCD patients have coexisting alpha thalassemia, beta thalassemia, hemoglobin C resulting in clinically significant differences (5).

In order to understand the pathophysiology of SCD it is necessary to consider the basic information about the hemoglobin structure and genetic control.  The importance of the intracellular  hemoglobin S (HbS) , cell- cell interactions, high shear rates by red cell deformability ,  and the vase- occlusion  and the  ischemia effect in tissues(7). Last decade we had many studies about situations that could influence the prognostic and survival of patients with SCD. (8,9,10).Garner ( 11)  et al (apud Garner) concluded  that the variation in the proportion of fetal cells (FC), leukocytes, platelets and red blood cell are genetically determined. The combinations of common gens are the causes of bad prognosis features in SCD. (11)

Miller et al (10) studied prospectively 392 children with SCD (SS) or S Thalassemia Beta zero before 6 months of age until 10 years of age. He did laboratories and clinical studies in order to evaluate those symptoms and signs that could be predictive of a bad outcome.

Another retrospective study was done in Brazilians sickle cell patients (SS) in order to compare with the Miller data and compare the results with those of Miller. This study would classify Brazilian patients with SCD into bad or not bad prognosis. Those with bad prognosis would be matter of discussion about the medical strategies to be followed with them. So, the objective of this work is to get to know the clinical or laboratories variables that would indicate that patients with SCD would have a bad outcome.


 Characteristics of the patients: patients presenting SCD registered in Instituto Estadual de Hematologia/ Rio de Janeiro/ Brazil from 1969 through 1995. The patient’s ages in the study varied from 4 months to 65 years of age.

Clinical Events and Laboratory Methods:

Hemoglobin typing was performed at Instituto de Hematologia with the use of cellulose acetate and citrate agar electrophoresis and alkali denaturation (for the measurement of fetal hemoglobin) (12). Hemoglobin mean levels; platelet counts; reticulocyte counts; acute splenic sequestration;  strokes; dactylitis ; painful crisis .

Definition of Clinical Events: Stroke was defined as an acute neurologic syndrome due to vascular occlusion or hemorrhage in which neurologic symptoms or signs lasted more than 24 hours. Transient ischemic attacks and silent infarctions were not included. A painful event was defined as pain in arms and/or legs, back, abdomen, chest or head that lasted at least two hours, led to request for medical care  and was attributable to sickle cell disease .Dactylitis , the acute chest syndrome, the right upper quadrant syndrome, osteomyelitis and apendicitis were not counted as painful crisis.

Dactylitis was defined as pain and tenderness, with or without swelling, in the hands, feet, or both. Acute splenic sequestration was defined as a decrease from base line in the hemoglobin level or hematocrit of at least 20 percent plus a simultaneous increase in the size of spleen to at least 2 cm below the left costal margin.

Adverse Events:

The adverse events served as proxies for severe sickle cell disease:

Adverse Event 1: Hemoglobin levels (hb) less than 7g/dL (in steady – state of the disease)

Adverse Event 2: The steady state platelet count higher than 5000.000/mm3.

Adverse Event 3: Splenic Sequestration before one year of age.

Adverse Event 4: Stroke

Adverse Event 5: Dactylitis in patients with  less than 2 years age.

Adverse Event 6: Painful crisis two or more times per year, needing medical care.

Adverse event 7: Reticulocyte count higher than 10% (in steady state)

Statistical Analysis:

Means or median (for nonnormal sample distribution) +/- SD were calculated when appropriate.

In order to identify the patients at highest risk for severe disease, we adopted  the identicators of Miller et al (11)

The paired Student t test was applied to compare the values of the group with or without bad prognosis (according to Miller variables of bad prognosis). P values =/< 0.05 where considered significants.

Comparison between groups was done by analysis of variance (ANOVA).

Correlation Coefficient.

The steady state hemoglobin level  were based on measurements obtained during a median of three visits, and the steady state platelet count was based on measurements, obtained during a median of two visits. All means laboratory values were first examined as continuous covariates to explore potential linear and nonlinear relations. In order to identify the patients at highest risk for severe disease, we used the Miller’s indicator’s.


Characteristics of the Patients:

Of the 279 patients enrolled in the study, 177 (63%) had adverse outcomes that qualified their disease as severe. In this group the median duration of follow up was 10.2 +/- 7.6 years.  60% of the patients  in the group of severe disease were girls. The most common adverse event that resulted in the classification of the disease as severe was the repetitive painful crisis (37%) (Table I)

Table I – characteristics of bad outcome  registered in 279 patients (some with more than  one characteristics)


No of Patients


N =                                                                             304

100 %

Platelets count > 500 (x 10- 3  / mmm 3)






Dactylitis in patients with less than  2 years of ages



Acute splenic sequestration



Painful events >2/yr



Hemoglobin level  <7 g/dl



Reticulocytes >10 %




Variable I:

Hemoglobin levels < 7 g/dl

The  mean steady state hemoglobin level for the entire cohort was 7,5 g/dl. The steady state hemoglobin level was the mean value obtained during 3 routine visits.

The multivariate analysis included  three variables that were significantly associated with na adverse clinical course: early dactylitis, repetitive painful crisis and stroke, there were no significant interactions  between them when we talk about hemoglobin levels less than 7 g/dl. (table II)

Table II: Mean  Hemoglobin Levels in Patients with Miller’s Severe Sickle Cell Disease.


No of patients

Mean hemoglobin Level




Repetitive painful crisis



Dactylitis presenting in SC patients with less than two years of age.



P Value


0,2563     NS



Age at death in patients with mean hemoglobin levels < 7g/dl was 17, 6 years

Age of death in patients with mean hemoglobin levels > 7g/dl   was 34 years. One tail p:  0, 05 (significant)

Variable 2-


The statistic analysis of the patients with sickle cell disease who presented platelets counts higher than 500.000/mm3 with or without repetitive painful crisis did not showed any difference. (Pair T analysis –P value=NS. So, the higher number of platelet did not influence the painful events .in our population. (Table I)

The higher platelet counts is a risk factor for the occurrence of acute splenic sequestration.

Variable 3:

Acute splenic sequestration :

 The SC patients that presents acute splenic sequestration have higher platelets count than those who did not presented the acute splenic sequestration  ( p≤0.05 ).

Variable 4


15 patients had stroke. From these 15 patients we just could study 13 patients.

These 13 patients age, varied from 2 to 43 years. The mean age at Stroke was 18 years old.

9/13 patients (70%) presented stroke with less or equal 12 years old (mean: 6, 5 yrs).Still in this group, 5 patients (55%) died with an average age of 16, 6 years. From 13 patients who presented stroke, 5 died (38%)

Four (4) patients presented stroke with ages higher than 28 years (average: 38, 2 years) .None died. (Table III).

The patients who presented stroke presented repetitive painful crisis and dactylitis before two years of age (Table IV)

Table III: Prevalence of stroke in different ages:

Age ≤ 6years


Age ≤ 15years


Age ≥  40 years


Table IV: Univariate analysis of the Risk of Severe Sickle Cell Disease:


30,7% (4/13)

Repetitive Painful Crisis

61,5% (8/13)

Hemoglobin levels  < 7 g/dl

15,3% (2/13)

Count Platelets > 500( X 10 -3  /mm3)        

38,4% (5/13)




The patients who presented dactylitis before the age of two years old also presented hemoglobin level (g/dl) < 7 when compared with the group of patients with SCD who did not presented dactylitis (p = 0,03).

The patients with repetitive painful crisis (> two times /yr, with admittance in Unit Emergency), had also dactylitis before the age of two years)- p value: 0,04.

The patients who presented early dactylitis had a survival age minor (16 years old) than the people that did not presented dactylitis (26years old) (p: 0, 05).

The patients, who presented dactylitis before the age of one year, came to the first medical consultation and admittance at hospital early than the group that did not presented this event.  (p = 0, 05).

In the group of patients who presented stroke, 30% also presented early dactylitis.




Table VII Comparison between patients with or without  early dactylitis presenting

 count platelets> 500 (X 10-3/mm 3) (T test paired) :









P(T<=t) bi-tailoredl




Critical T bi-tailored




The platelets count higher 500(x10-3/mm3) did not influenced the occurrence of dactylitis.

Variable: 6

Repetitive Painful Crisis:

When one compares the  group of patients who presented  repetitive painful crisis with the group of patients who did not presented it and make a correlation of both group, with the platelet counts higher than 500(x10-3/mm3 )the T test of  Student results in a p value significant in one tail, showing that there is a bias

When one compares the group that presented  repetitive painful crisis with the group that did not presented it there is a bias the firs group to have a shorter survival time( 24,2 years old) . P=NS

The patients with hemoglobin levels <  7 g/dl  compared with the group with hemoglobin levels > 7g/dl, when analyzed with the variable  dactylitis before the age of one year, the statistical study  was not significant ( p value= NS)

 Variable 7:

 Reticulocytes count > 10 %:

The Regression Curve (regression analysis) demonstrated a relationship between the higher number of reticulocytes and shorter survival time.

When  we compared the age of death in the group of patients that presented reticulocyte higher than 10% with the patients who presented reticulocyte counts less than  10% percent, we observed that  both statistical analysis:  T paired  study and Regression Analysis showed us  a  linear enhancement  in death occurrence according to reticulocyte count  elevation. (p= 0, 01)

It was done a comparative study of  death occurrence in  SC patients that presented more bad outcome variables ( 1, 2, 3, 4 e 5 ); The age of death was lower in SC patients with two bad outcome variables  in comparison with those patients that presented only one bad outcome variable, but  in T student test it was not statistically significant.

When we compared SC patients presenting one or three bad outcome variables the p value was 0, 08, putting altogether there are a bias .The age group average was different between these patients. The age at death in those patients that presented one bad outcome variable was 21 years of age; the age at death in those patients that presented two bad outcome variables was: 19 years; the age at death in those patients that presented three variables was 13 years old. With a bigger number of patients we think that it will be statistically significant, meaning that:

The more the numbers of bad outcome variables, earlier will be the death.

Also, according to Graphic Regression II, the more the number of bad outcome variables in patients with SCD the greater probability of early death

From 13 patients that presented stroke, 5 died. (38%)

Table VIII- Deaths and ages of whole group of SC patients:


Number (%)

Average (yrs)

           death<10 years

8 (8.5%)


          death>10 years



           death>20 years







In this observational study of patients with sickle cell anemia, we found that repetitive painful crisis as the most common of bad outcome variables (37%).

There was association between early dactylites and repetitive painful crisis. So there is a correlation of early dactylites (before one year of age) and sever disease, in accordance with the Miller’s population. The clinical importance of this fact is that when we, hematologists meet a child with SC anemia with early dactylites we must go on and become worry about the future of this child for he probably will present painful crisis in infancy and adolescence. At the moment we have this insight, we are oblige to act in order to give a better life quality for him/her. We will have to use Hidroxiurea or EACA inhibitors or what else appear not to permit the pain to become a constant in him/her life. We know that in Africa, sickle cell anemia is called just as an onomatopoeic “pain cry “.

Hydroxyurea   stimulates the production of fetal cells within the red blood cells.

The intracellular polymerization of sickle hemoglobin is the most important determinant of clinical and biological manifestations of sickle cell anemia. The fetal hemoglobin at this point has the most important role.  (2). When there is fetal hemoglobin there is no production of S polymers and its intracellular expression in sickle cells inhibit the polymerization.

The homozygous state of SC anemia from Saudi Arabian  is clinically milder for they have higher levels of fetal because there is a gene mutation in globinic beta chain  in 11 chromosome.(5)

Hereditary persistence of fetal hemoglobin (HPFH) is the name given to a group of conditions characterized by persistence of fetal hemoglobin synthesis into adult life in the absence of significant hematological manifestations. This state has much in common with b and d b thalassemias. There are different forms of HPFH. When the heterozygous state combines with hemoglobin S, there are usually no anemia and little evidence of because of the uniformly high level of HbF per cell that renders the interaction with HPFH and HbS so mild. This is an important observation because it provides valuable insights into the level of HbF in each red cell that would be required to control sickle cell anemia by pharmacological stimulation of HbF synthesis (9). The level of HbF varies between 15 and 42%. The mean level of HbF is about 29%.

A high level of HbF in SCD correlates with increased life expectancy and a decrease in the incidence of vase- occlusive crises. HbF levels at birth vary between 60 and 80 per cent and are similar in newborns with SS disease and in matched AA controls. The rate of fall in HbF after birth is slower in SS disease compared to AA controls. As the HbF falls start the symptoms of the disease fatigue, pallor, jaundice and dactylitis (13) .An analysis of 105 patients with SCD found that the average in HbF in each group is higher than in general population. The age group correspondent to 0 to 5 years old the levels of HbF are the highest (mean: 13, 8% and 12, 3%). At older ages the levels fell. In those patients with ages between 30 and 40 years old HbF increased up (mean: 8, 45%). Over the age of 40 years no trends was apparent with a HbF mean of 3, 13.

The leukocytes levels, mainly if there are also a monocytosis must be considered in ambulatory patients because they are related to a severe disease. The monocytes must be respected because of their action in conjunction with endothelial cells playing an important role in stroke and in painful crises (3).

As Miller, we did not observed correlation in the platelet number and painful crises.

We could not compare the clinical and laboratorial features of acute chest syndrome in Brazilian SCD patients, but they are very common and sometimes worse or even began with the use of Tramadol, and the author related three cases of patients that began with painful crises, went to Emergency Room and the medical in charge prescribed Tramadol and all of them developed an acute chest syndrome, one of the patients died.


The hemoglobin level of our population was 7.5 g/dl, while in the Miller’s was 8.1g/dl

We found correlation between the platelet counts higher than 500 000 /mm3 and acute spleen sequestration P< 0, 05.

 70% of SC patients had stroke at ages equal or less than 12years. 55% of SC patients who presented stroke died (mean age: 16 years). 23% of SC patients presented stroke after 40 years old. None died.

The patients who presented early dactylitis had a decrease in survival time. (16 years old was the average death age of this group).

It was not found correlation between platelets count higher than 500 000/mm3 and dactylitis p: NS

 In the patients who presented repetitive painful crises, we observed a positive correlation with higher platelet counts in one tail. (Miller compared with platelet counts higher than 400 000/mm3)

It was not observed that the patients who presented repetitive painful crises had a lower survival time. This fact is in accordance with Miller’s study.

The regression graphic and the T paired test were consistent with:  the higher the reticulocyte counts is, the decreased is, the survival time. This statement is discordant with Miller’s wok.

Miller described that the probability of interactions of more than one bad outcome variables is important for us to begin a therapeutic strategy for in those cases the survival time is decreased. The regression curve of our patients also showed in agreement with this statement.

According Miller the most important bad outcome variables were: dactylitis before the first birthday; Hb<7g/dl; acute spleen sequestration; leukocytes counts constantly elevated; STA.

In the Brazilian patients the most important bad outcomes variables were: stroke; dactylitis before the first birthday and the percentile of reticulocyte constantly higher than 10%. The platelet count higher than 500 000 /mm3 was correlated with the acute splenic sequestration event.

 The bad outcome variables leukocytes counts high and acute chest syndrome was not available for us to study.



1-      Beutler,E: Hemolytic Anemia in Disorders of Cell Metabolism; Topics in hematology; Plenum Medical Book Co; 1978

2-      Ballas SK: Larnerj et al: Rheological predictors of the severity of painful sickle cell crisis. Blood 72: 1216- 1223; 1986

3-      Henderson, AB: Sickle Cell Anemia: clinical study of 54 cases. Am J. Med. 9: 757-765;1950

4-      Setty, BNY; Stuart MJ: Vascular cell adhesion molecule I is involved in mediating hipoxia- induced sickle cell red blood cell adherence to endothelium: Potential role in sickle cell disease. Blood; 2311- 2310; 1996

5-      Gallo da Rocha HHAG: Apostolado de Anemia Falciforme: L Lamina; maio de 2000.

6-      Gallo da Rocha, HHA; Cursino MB: Anemia Falciforme. Estudo de 68 casos; Ver. Inst Est Hemat. 9:5- 11; 1992.

7-      Powars, DR; Wilson,B; Imbus C et al : The natural history of stroke in sickle cell disease.  Am. J. Med 65: 461- 470; 1978.

8-      Chang, YP; Maier- Redelperg M; Smith, KD; et al: The relative importance of the X linked FCP locus and beta globin haplotypes in determining HF levels: a study of SS patients homozygous for bS haplotypes. Brit. J. Haemat.: 96: 4; 806- 814; 1997

9-      Baum, K; Maude GH; Serjeant; GR: The painful crisis of homozygous sickle cell disease: A study of risks factors. Arch. Int. Med. 147: 1231- 1234; 1987

10-  -Miller,ST, Sleeper SD, Pegelow, CH et al:  Prediction of ad verses outcomes in children with sickle cell disease;  NEJM, 342: 83- 89; 2000

11-  Garner, C; Tatu, T; Reittie, E et al: Genectic influence on F cells and other hematological variables: a twin heritability study: Blood; 95: 342- 346, 2000.

12-  Kleihauer, E; Braun, H and Betke, - K: Demonstration von fetalem Hamoglobin in den Erythrocyten eines Blutausstrichs. Klinische Wochenschrift, 35: 637- 644; 1957