During_IN 426_CD patient-years_NNS of_IN follow-up_JJ for_IN patients_NNS with_IN standard_JJ criteria_NNS ,_, 3.3_CD acute_JJ chest_NN syndromes_NNS ,_, 1.3_CD cerebrovascular_JJ events_NNS ,_, and_CC 1.1_CD osteonecrosis_NNS per_IN 100_CD patient-years_NNS were_VBD observed_VBN ._. In_IN 72_CD patients_NNS evaluated_VBN by_IN transcranial_JJ Doppler_NNP studies_NNS -LRB-_-LRB- TCD_NNP -RRB-_-RRB- ,_, 34_CD patients_NNS were_VBD at_IN risk_NN of_IN primary_JJ stroke_NN and_CC only_RB 1_CD had_VBD a_DT cerebrovascular_JJ event_NN after_IN a_DT follow-up_JJ of_IN 96_CD patient-years_NNS ._. HU_NNP was_VBD interrupted_VBN definitely_RB for_IN 12_CD patients_NNS :_: 7_CD underwent_VBD allogeneic_JJ bone_NN marrow_NN transplantation_NN ,_, 1_CD failed_VBD to_TO respond_VB after_IN 2_CD years_NNS of_IN treatment_NN ,_, 1_CD with_IN HbSC_NNP because_IN of_IN a_DT total_JJ Hb_NNP level_NN higher_JJR than_IN 13_CD g\/dL_NNP ,_, 2_CD refused_VBD to_TO pursue_VB the_DT treatment_NN after_IN 1_CD and_CC 2_CD years_NNS ,_, and_CC 1_CD patient_JJ developed_JJ malignancy_NN ._. Hydroxyurea_NN was_VBD also_RB transiently_RB stopped_VBD in_IN 3_CD cases_NNS ,_, 2_CD for_IN pregnancy_NN and_CC 1_CD for_IN systemic_JJ lupus_NN syndrome_NN ,_, diagnosed_VBD one_CD year_NN after_IN the_DT beginning_NN of_IN treatment_NN ._. Despite_IN a_DT sustained_JJ increase_NN in_IN HbF_NNP level_NN and_CC an_DT increase_NN in_IN hydroxyurea_NN doses_NNS to_TO 30_CD mg\/kg_NN per_IN day_NN ,_, a_DT girl_NN aged_VBN 13_CD months_NNS at_IN inclusion_NN with_IN previous_JJ history_NN of_IN splenic_JJ sequestration_NN ,_, was_VBD hospitalized_VBN 20_CD times_NNS during_IN these_DT 2_CD years_NNS of_IN HU_NNP therapy_NN and_CC continued_VBD to_TO present_VB multiple_JJ vaso-occlusive_JJ crises_NNS and_CC dactylitis_NNS during_IN the_DT first_JJ year_NN of_IN HU_NNP ._. She_PRP died_VBD during_IN the_DT second_JJ year_NN of_IN therapy_NN from_IN acute_JJ severe_JJ anemia_NN during_IN an_DT episode_NN of_IN splenic_JJ sequestration_NN ._. Acute_JJ promyelocytic_JJ leukemia_NN was_VBD diagnosed_VBN in_IN a_DT 21-year-old_JJ female_NN after_IN 8_CD years_NNS of_IN HU_NNP therapy_NN ._. Hospitalizations_NNS ,_, VOCS_NNP ,_, and_CC ACS_NNP were_VBD more_RBR frequently_RB observed_VBN after_IN 3_CD years_NNS of_IN treatment_NN than_IN during_IN the_DT first_JJ 3_CD years_NNS ._. Recurrent_JJ stroke_NN was_VBD observed_VBN once_RB in_IN an_DT 8-year-old_JJ girl_NN ,_, 6_CD years_NNS after_IN the_DT initial_JJ event_NN ._. There_EX were_VBD 8_CD patients_NNS followed_VBD for_IN a_DT median_NN of_IN 6_CD years_NNS -LRB-_-LRB- range_NN ,_, 3-9_CD years_NNS -RRB-_-RRB- ,_, with_IN 44_CD cumulative_JJ patient-years_NNS of_IN follow-up_JJ ,_, at_IN risk_NN of_IN secondary_JJ stroke_NN ._. Recurrent_JJ stroke_NN occurred_VBD in_IN one_CD patient_NN after_IN 6_CD years_NNS of_IN follow-up_JJ despite_IN evidence_NN of_IN good_JJ compliance_NN with_IN therapy_NN as_IN assessed_VBN by_IN hematologic_JJ response_NN to_TO HU_NNP ._. As_IN a_DT whole_NN ,_, 34_CD patients_NNS were_VBD considered_VBN at_IN risk_NN of_IN primary_JJ stroke_NN on_IN the_DT basis_NN of_IN abnormal_JJ TCD_NNP ,_, and_CC 7_CD of_IN the_DT 21_CD explored_VBN by_IN MRI\/MRA_NNP had_VBD moderate\/severe_JJ arterial_JJ stenosis_NN ._. Only_RB 1_CD of_IN these_DT 34_CD patients_NNS presented_VBD a_DT cerebrovascular_JJ event_NN -LRB-_-LRB- seizures_NNS -RRB-_-RRB- after_IN an_DT evaluation_NN of_IN 96_CD patient-years_NNS ._. One_CD patient_NN who_WP died_VBD was_VBD described_VBN previously_RB ._. Toxicities_NNS of_IN HU_NNP were_VBD reported_VBN with_IN 2_CD transient_JJ episodes_NNS of_IN thrombocytopenia_NN -LRB-_-LRB- 2_CD infants_NNS -RRB-_-RRB- that_WDT required_VBD temporary_JJ discontinuation_NN of_IN therapy_NN ._. There_EX were_VBD 5_CD episodes_NNS observed_JJ of_IN severe_JJ anemia_NN -LRB-_-LRB- Hb_NNP ,_, -LRB-_-LRB- 6_CD g\/dL_NNP -RRB-_-RRB- that_WDT required_VBD transfusion_NN -LRB-_-LRB- 4_CD infants_NNS -RRB-_-RRB- ._. Aplastic_JJ crisis_NN associated_VBN with_IN a_DT parvovirus_NN infection_NN was_VBD documented_VBN in_IN one_CD patient_NN ._. During_IN the_DT trial_NN period_NN ,_, 75_CD ACS_NNP events_NNS occurred_VBD and_CC were_VBD 2_CD times_NNS more_RBR frequent_JJ in_IN the_DT placebo_NN arm_NN -LRB-_-LRB- 51_CD events_NNS -RRB-_-RRB- than_IN in_IN the_DT HU_NNP arm_NN -LRB-_-LRB- 25_CD events_NNS -RRB-_-RRB- ._. In_IN transfused_VBN patients_NNS ,_, the_DT prevalence_NN of_IN recurrent_JJ stroke_NN has_VBZ been_VBN reported_VBN at_IN 13_CD %_NN to_TO 22_CD %_NN -LRB-_-LRB- with_IN one_CD recurrence_NN for_IN each_DT 24-41_CD patient-years_NNS -RRB-_-RRB- and_CC the_DT prevalence_NN of_IN TIA_NNP is_VBZ 19_CD %_NN ._. In_IN particular_JJ ,_, the_DT Belgian_JJ experience_NN with_IN chronic_JJ transfusion_NN is_VBZ very_RB disappointing_JJ because_IN of_IN a_DT high_JJ rate_NN of_IN severe_JJ alloimmunization_NN ,_, mainly_RB explained_VBN by_IN the_DT absence_NN of_IN blood_NN donors_NNS of_IN African_JJ origin_NN ._. Repeated_VBN vaso-occlusive_JJ crises_NNS are_VBP sometimes_RB present_JJ from_IN infancy_NN and_CC early_JJ childhood_NN ._. Furthermore_RB ,_, dactylitis_NN ,_, anemia_NN ,_, and_CC high_JJ WBC_NNP counts_NNS in_IN infancy_NN predict_VBP a_DT severe_JJ outcome_NN ,_, and_CC this_DT could_MD trigger_VB an_DT early_JJ start_NN of_IN HU_NNP in_IN these_DT patients_NNS ._. Of_IN interest_NN ,_, 4_CD of_IN the_DT children_NNS included_VBN in_IN the_DT registry_NN at_IN younger_JJR than_IN 2_CD years_NNS developed_VBD episodes_NNS of_IN severe_JJ acute_JJ anemia_NN ._. One_CD patient_NN died_VBD of_IN a_DT fatal_JJ episode_NN of_IN splenic_JJ sequestration_NN before_IN completing_VBG 2_CD years_NNS of_IN treatment_NN ._. During_IN the_DT dose_NN titration_NN ,_, blood_NN counts_NNS consistent_JJ with_IN a_DT finding_NN of_IN marrow_NN depression_NN were_VBD observed_VBN at_IN least_JJS once_RB in_IN 35_CD percent_NN of_IN the_DT patients_NNS who_WP received_VBD placebo_NN ._. Treatment_NNP was_VBD temporarily_RB stopped_VBN in_IN almost_RB all_DT patients_NNS in_IN the_DT hydroxyurea_NN group_NN because_IN of_IN marrow_NN depression_NN ;_: blood_NN counts_NNS usually_RB recovered_VBD within_IN two_CD weeks_NNS ._. Treatment_NNP was_VBD interrupted_VBN in_IN four_CD patients_NNS in_IN the_DT placebo_NN group_NN because_IN of_IN increased_VBN bilirubinemia_NN -LRB-_-LRB- bilirubin_NN ,_, -RRB-_-RRB- 10_CD mg_NN per_IN deciliter_NN -LRB-_-LRB- 103_CD umol_NN per_IN liter_NN -RRB-_-RRB- -RRB-_-RRB- ._. Hair_NN loss_NN ,_, rash_NN ,_, fever_NN ,_, and_CC gastrointestinal_JJ disturbance_NN were_VBD as_RB common_JJ in_IN patients_NNS receiving_VBG placebo_NN as_IN in_IN those_DT taking_VBG hydroxyurea_NN ._. In_IN six_CD patients_NNS -LRB-_-LRB- one_CD in_IN the_DT hydroxyurea_NN group_NN -RRB-_-RRB- parvovirus_VBZ b19_JJ infection_NN developed_VBN during_IN treatment_NN ._. The_DT aplastic_JJ crises_NNS caused_VBN by_IN the_DT virus_NN were_VBD not_RB prolonged_VBN ,_, and_CC all_PDT the_DT patients_NNS recovered_VBD uneventfully_RB ._. We_PRP did_VBD not_RB address_VB the_DT reversibility_NN of_IN chronic_JJ organ_NN damage_NN ;_: it_PRP is_VBZ unknown_JJ whether_IN the_DT inhibition_NN of_IN sickling_NN could_MD affect_VB such_JJ preexisting_JJ lesions_NNS ._. Hematopoietic_JJ depression_NN did_VBD occur_VB during_IN therapy_NN ,_, as_IN anticipated_VBN ,_, but_CC it_PRP was_VBD short_RB lived_VBN ._. Adverse_JJ reactions_NNS were_VBD equally_RB common_JJ in_IN both_DT treatment_NN groups_NNS ._. There_EX is_VBZ concern_NN that_IN long-term_JJ hydroxyurea_NN therapy_NN may_MD be_VB carcinogenic_JJ or_CC leukemogenic_JJ ,_, because_IN some_DT other_JJ antineoplastic_JJ agents_NNS have_VBP such_JJ effects_NNS ._. Adhesion_NN molecules_NNS on_IN the_DT surface_NN of_IN erythrocytes_NNS ,_, leukocytes_NNS and_CC platelets_NNS are_VBP involved_VBN in_IN vascular_JJ occlusion_NN in_IN sickle_NN cell_NN anemia_NN ._. On_IN the_DT other_JJ hand_NN ,_, hydroxyurea_NN treatment_NN reduced_VBD the_DT percentage_NN of_IN reticulocytes_NNS -LRB-_-LRB- 8.44_CD %_NN vs_VBZ 4.46_CD %_NN -RRB-_-RRB- and_CC of_IN erythrocytes_NNS positive_JJ for_IN CD36_NNP ,_, CD71_NNP ,_, CD49d_NNP and_CC annexin_NNP V_NNP -LRB-_-LRB- Figures_NNS 1_CD and_CC 2_CD -RRB-_-RRB- ._. Hydroxyurea_NN treatment_NN reduced_VBD the_DT percentage_NN of_IN annexin_FW V_FW +_FW erythrocytes_NNS in_IN all_DT patients_NNS ,_, but_CC one_CD -LRB-_-LRB- Figure_NNP 1A_NNP -RRB-_-RRB- ._. Hydroxyurea_NN treatment_NN significantly_RB reduced_VBD total_JJ leukocyte_JJ counts_NNS from_IN pretreatment_NN values_NNS ._. The_DT percentage_NN of_IN annexin_JJ V-labeled_JJ platelets_NNS was_VBD reduced_VBN during_IN hydroxyurea_NN treatment_NN -LRB-_-LRB- Table_NNP 2_CD -RRB-_-RRB- in_IN all_DT but_CC two_CD patients_NNS -LRB-_-LRB- Figure_NNP 1B_NNP -RRB-_-RRB- ._. The_DT increase_NN in_IN MCV_NNP was_VBD positively_RB correlated_VBN with_IN the_DT increase_NN in_IN F_NN cells_NNS in_IN the_DT circulation_NN and_CC in_IN hemoglobin_NN concentration_NN and_CC negatively_RB correlated_VBN with_IN the_DT percentage_NN of_IN reticulocytes_NNS and_CC of_IN cells_NNS expressing_VBG the_DT adhesive_NN molecules_NNS CD36_NNP ,_, CD49d_NNP and_CC annexin_NNP V_NNP -LRB-_-LRB- Figure_NNP 4_CD -RRB-_-RRB- ._. This_DT observation_NN suggests_VBZ that_IN the_DT increase_NN in_IN MCV_NNP may_MD have_VB a_DT direct_JJ effect_NN on_IN the_DT adhesive_JJ phenotype_NN of_IN sickle_NN cells_NNS and_CC that_IN monitoring_VBG the_DT increase_NN of_IN MCV_NNP may_MD be_VB used_VBN as_IN indirect_JJ evidence_NN of_IN the_DT decrease_NN of_IN the_DT adhesive_JJ phenotype_NN of_IN erythrocytes_NNS in_IN patients_NNS with_IN sickle_JJ cell_NN anemia_NN treated_VBN with_IN hydroxyurea_NN ._. In_IN the_DT present_JJ study_NN we_PRP observed_VBD that_IN the_DT greatest_JJS PS_NNP exposure_NN occurred_VBD on_IN the_DT largest_JJS erythrocytes_NNS ,_, as_IN demonstrated_VBN by_IN flow_NN cytometry_NN -LRB-_-LRB- larger_JJR FSC_NNP -RRB-_-RRB- -LRB-_-LRB- Figure_NN 3_CD -RRB-_-RRB- ._. This_DT population_NN also_RB showed_VBD a_DT concentration_NN of_IN erythrocytes_NNS expressing_VBG CD71_NNP ,_, CD36_NNP and_CC CD49d_NNP and_CC of_IN reticulocytes_NNS -LRB-_-LRB- data_NNS not_RB shown_VBN -RRB-_-RRB- ._. Reticulocytes_NNS are_VBP the_DT cells_NNS with_IN the_DT greatest_JJS adhesiveness_NN to_TO vascular_JJ endothelium_NN in_IN sickle_NN cell_NN anemia_NN ._. This_DT increased_VBN adhesiveness_NN has_VBZ been_VBN attributed_VBN to_TO the_DT adhesion_NN molecules_NNS CD36_NNP and_CC CD49d_NNP and_CC ,_, more_RBR recently_RB ,_, to_TO PS_NNP exposure_NN on_IN the_DT cell_NN membrane_NN ._. PS_NNP exposure_NN on_IN the_DT surface_NN of_IN sickle_NN cells_NNS ,_, in_IN addition_NN to_TO affecting_VBG erythrocyte_JJ adhesion_NN to_TO the_DT vascular_JJ endothelium_NN ,_, exacerbates_VBZ anemia_NN by_IN enhancing_VBG phagocyte_JJ recognition_NN and_CC removal_NN of_IN these_DT cells17_CD and_CC favors_VBZ the_DT development_NN of_IN a_DT thrombophilic_JJ state_NN ._. There_EX is_VBZ a_DT direct_JJ correlation_NN between_IN PS_NNP exposure_NN on_IN sickle_NN erythrocytes_NNS and_CC the_DT generation_NN of_IN thrombin_NN ._. Six_CD patients_NNS with_IN persistently_RB abnormal_JJ TCD_NNP results_NNS developed_VBD stroke_NN ._. Persistent_JJ TCD_NNP elevation_NN signals_VBZ ongoing_JJ stroke_NN risk_NN ._. Two_CD patients_NNS who_WP had_VBD quit_VBN transfusion_NN subsequently_RB died_VBD of_IN sepsis_NN ._. From_IN the_DT SC_NNP group_NN ,_, 2_CD patients_NNS developed_VBD stroke_NN soon_RB after_IN closure_NN of_IN the_DT trial_NN and_CC started_VBD transfusion_NN -LRB-_-LRB- 1_CD at_IN 2_CD weeks_NNS and_CC another_DT at_IN 10_CD weeks_NNS after_IN the_DT trial_NN -RRB-_-RRB- ._. In_IN addition_NN to_TO the_DT 12_CD stroke_NN events_NNS in_IN the_DT STOP_NNP trial_NN ,_, 6_CD patients_NNS developed_VBD stroke_NN during_IN the_DT posttrial_JJ follow-up_JJ -LRB-_-LRB- 5_CD from_IN the_DT SC_NNP group_NN and_CC 1_CD from_IN the_DT TX_NNP group_NN -RRB-_-RRB- ._. From_IN the_DT SC_NNP group_NN ,_, aside_RB from_IN the_DT 2_CD patients_NNS who_WP had_VBD stroke_NN in_IN the_DT immediate_JJ posttrial_JJ period_NN prior_RB to_TO choosing_NN treatment_NN ,_, 1_CD patient_NN who_WP refused_VBD transfusion_NN and_CC 2_CD who_WP started_VBD transfusion_NN during_IN the_DT posttrial_JJ follow-up_JJ developed_JJ stroke_NN ._. The_DT only_JJ patient_NN from_IN the_DT TX_NNP group_NN who_WP developed_VBD stroke_NN had_VBD discontinued_VBN transfusion_NN after_IN the_DT trial_NN ._. All_DT stroke_NN events_NNS were_VBD ischemic_JJ infarctions_NNS ._. All_DT 6_CD patients_NNS who_WP developed_VBD stroke_NN during_IN the_DT posttrial_JJ follow-up_NN had_VBD a_DT last_JJ interpretable_JJ TCD_NNP that_WDT was_VBD abnormal_JJ prior_RB to_TO stroke_NN regardless_RB of_IN transfusion_NN status_NN ._. The_DT data_NNS confirm_VBP that_IN stroke_NN risk_NN increases_NNS with_IN TCD_NNP velocity_NN ._. Elevated_JJ cerebral_JJ blood-flow_NN velocities_NNS in_IN SCD_NNP are_VBP related_VBN to_TO severe_JJ anemia_NN ,_, vessel_NN stenosis_NN ,_, and_CC cerebral_JJ vasodilatation_NN caused_VBN by_IN tissue_NN hypoxia_NN ._. Discontinuation_NN of_IN transfusion_NN after_IN at_IN least_JJS 30_CD months_NNS in_IN patients_NNS who_WP have_VBP converted_VBN to_TO normal_JJ velocities_NNS was_VBD associated_VBN with_IN high_JJ risk_NN of_IN reversion_NN to_TO abnormal_JJ and_CC chance_NN of_IN stroke_NN ._. Those_DT with_IN persistently_RB abnormal_JJ TCD_NNP results_NNS despite_IN transfusion_NN often_RB have_VBP severe_JJ vasculopathy_NN and_CC are_VBP at_IN much_RB higher_JJR risk_NN for_IN stroke_NN ._. Our_PRP$ report_NN demonstrated_VBD in_IN 2_CD patients_NNS that_WDT persistently_RB abnormal_JJ TCD_NNP results_NNS while_IN on_IN transfusion_NN is_VBZ ominous_JJ and_CC may_MD predict_VB the_DT uncommon_JJ stroke_NN that_WDT occurs_VBZ on_IN transfusion_NN therapy_NN ._. Treatment_NNP was_VBD also_RB withdrawn_VBN in_IN 5_CD of_IN 6_CD children_NNS who_WP had_VBD developed_VBN hypersplenism_NN ,_, in_IN 3_CD because_IN of_IN a_DT pathological_JJ transcranial_NN Doppler_NNP ,_, and_CC in_IN 2_CD after_IN a_DT stroke_NN ._. One_CD 18-year_JJ old_JJ patient_NN ,_, who_WP had_VBD received_VBN hydroxyurea_NN for_IN 11_CD months_NNS and_CC had_VBD only_RB a_DT slight_JJ reduction_NN in_IN the_DT frequency_NN of_IN painful_JJ episodes_NNS ,_, died_VBD suddenly_RB from_IN asystole_NN while_IN he_PRP was_VBD hospitalized_VBN for_IN a_DT painful_JJ crisis_NN ._. No_DT autopsy_NN was_VBD performed_VBN and_CC death_NN was_VBD attributed_VBN to_TO a_DT probable_JJ underlying_JJ SCD-related_JJ myocardiopathy_NN ._. In_IN one_CD child_NN ,_, already_RB reported_VBN ,_, a_DT diagnosis_NN of_IN acute_JJ lymphoblastic_JJ leukemia_NN with_IN evidence_NN of_IN Philadelphia_NNP chromosome_NN was_VBD made_VBN 1.5_CD months_NNS after_IN starting_VBG hydroxyurea_NN ._. Of_IN the_DT five_CD patients_NNS switched_VBN to_TO hydroxyurea_NN after_IN normalization_NN of_IN their_PRP$ TCD_NNP under_IN transfusion_NN ,_, two_CD developed_JJ velocities_NNS above_IN 200_CD cm\/sec_NN and_CC were_VBD again_RB prescribed_VBN regular_JJ exchange_NN blood_NN transfusions_NNS ._. Treatment_NNP was_VBD also_RB stopped_VBN in_IN another_DT patient_NN because_IN of_IN the_DT first_JJ occurrence_NN of_IN pathological_JJ TCD_NNP velocities_NNS ._. Hydroxyurea_NN was_VBD initiated_VBN in_IN six_CD patients_NNS because_IN of_IN history_NN of_IN stroke_NN ,_, which_WDT was_VBD the_DT sole_JJ indication_NN in_IN two_CD of_IN them_PRP ._. There_EX were_VBD two_CD stroke_NN recurrences_NNS in_IN our_PRP$ cohort_NN ._. One_CD occurred_VBD after_IN 3_CD months_NNS of_IN treatment_NN in_IN a_DT child_NN in_IN whom_WP stroke_NN was_VBD the_DT only_JJ indication_NN for_IN hydroxyurea_NN ._. The_DT second_JJ one_CD occurred_VBD in_IN a_DT child_NN whose_WP$ past_JJ history_NN of_IN stroke_NN was_VBD not_RB the_DT reason_NN advocated_VBN by_IN his_PRP$ physician_NN for_IN hydroxyurea_NN treatment_NN ,_, this_DT reason_NN being_VBG the_DT existence_NN of_IN painful_JJ crises_NNS ._. This_DT child_NN had_VBD a_DT stroke_NN after_IN 8_CD years_NNS of_IN treatment_NN ._. Hypersplenism_NNP was_VBD observed_VBN in_IN six_CD Hb_NNP SS_NNP patients_NNS ._. Before_IN hydroxyurea_NN treatment_NN ,_, three_CD of_IN them_PRP had_VBD splenomegaly_RB and_CC two_CD a_DT previous_JJ history_NN of_IN splenic_JJ sequestration_NN ._. The_DT diagnosis_NN was_VBD based_VBN on_IN an_DT increase_NN in_IN spleen_NN size_NN and_CC concomitant_JJ thrombocytopenia_NN ;_: two_CD cases_NNS ,_, aged_VBN 6_CD and_CC 7_CD ,_, had_VBD recurrent_JJ splenic_JJ sequestration_NN crises_NNS ._. Five_CD patients_NNS were_VBD splenectomized_VBN ,_, three_CD of_IN them_PRP have_VBP restarted_VBN hydroxyurea_NN with_IN good_JJ tolerance_NN ._. Hydroxyurea_NN was_VBD stopped_VBN in_IN seven_CD of_IN these_DT patients_NNS -LRB-_-LRB- one_CD because_IN of_IN headache_NN ,_, one_CD because_IN of_IN hypersplenism_NN ,_, one_CD because_IN TCD_NNP velocity_NN became_VBD abnormal_JJ ,_, and_CC four_CD because_IN of_IN treatment_NN failure_NN -RRB-_-RRB- ._. Hypersplenism_NNP ,_, which_WDT was_VBD diagnosed_VBN in_IN six_CD patients_NNS ,_, is_VBZ a_DT classical_JJ complication_NN of_IN SCD_NNP ,_, but_CC recurrent_JJ splenic_JJ sequestration_NN episodes_NNS are_VBP unusual_JJ in_IN children_NNS aged_VBN 6_CD and_CC 7_CD years_NNS ,_, Hb_NNP SC_NNP children_NNS excepted_VBD ,_, and_CC suggest_VBP hydroxyurea-induced_JJ hypersplenism_NN ._. All_PDT these_DT observations_NNS converge_VBP to_TO suggest_VB that_DT hydroxyurea_NN can_MD prevent_VB or_CC delay_VB functional_JJ asplenia_NN ,_, increasing_VBG the_DT risk_NN and_CC lengthening_VBG the_DT at-risk_JJ period_NN for_IN acute_JJ splenic_JJ sequestration_NN episodes_NNS ._. We_PRP ,_, therefore_RB ,_, recommend_VB careful_JJ evaluation_NN of_IN spleen_NN size_NN and_CC blood_NN tests_NNS at_IN each_DT evaluation_NN in_IN children_NNS with_IN previous_JJ splenomegaly_NN or_CC a_DT past_JJ history_NN of_IN splenic_JJ sequestration_NN before_IN starting_VBG hydroxyurea_NN ._. One_CD malignancy_NN occurred_VBD in_IN our_PRP$ cohort_NN ,_, but_CC analysis_NN of_IN this_DT case_NN led_VBD us_PRP to_TO conclude_VB that_DT ,_, most_RBS probably_RB ,_, the_DT leukemia_NN pre-existed_VBD the_DT hydroxyurea_NN administration_NN ,_, since_IN the_DT leukemia_NN was_VBD diagnosed_VBN 1.5_CD months_NNS after_IN starting_VBG hydroxyurea_NN therapy_NN ._. This_DT short_JJ interval_NN strongly_RB suggested_VBD that_IN the_DT bone_NN pains_NNS which_WDT had_VBD prompted_VBN hydroxyurea_NN treatment_NN were_VBD in_IN fact_NN the_DT first_JJ manifestation_NN of_IN the_DT leukemia_NN ._. However_RB ,_, a_DT special_JJ warning_NN must_MD be_VB made_VBN concerning_VBG the_DT potential_JJ occurrence_NN of_IN splenic_JJ twenty-five_JJ SCD_NNP children_NNS hypersplenism_NN ._. When_WRB the_DT acute_JJ chest_NN syndrome_NN was_VBD diagnosed_VBN ,_, patients_NNS had_VBD hypoxia_NN ,_, decreasing_VBG hemoglobin_NN values_NNS ,_, and_CC progressive_JJ multilobar_JJ pneumonia_NN ._. Patients_NNS who_WP were_VBD 20_CD or_CC more_JJR years_NNS of_IN age_NN had_VBD a_DT more_RBR severe_JJ course_NN than_IN those_DT who_WP were_VBD younger_JJR ._. Neurologic_JJ events_NNS occurred_VBD in_IN 11_CD percent_NN of_IN patients_NNS ,_, among_IN whom_WP 46_CD percent_NN had_VBD respiratory_JJ failure_NN ._. Among_IN the_DT specific_JJ causes_NNS were_VBD pulmonary_JJ fat_JJ embolism_NN and_CC 27_CD different_JJ infectious_JJ pathogens_NNS ._. Eighteen_CD patients_NNS died_VBD ,_, and_CC the_DT most_RBS common_JJ causes_NNS of_IN death_NN were_VBD pulmonary_JJ emboli_NNS and_CC infectious_JJ bronchopneumonia_NN ._. Infection_NN was_VBD a_DT contributing_JJ factor_NN in_IN 56_CD percent_NN of_IN the_DT deaths_NNS ._. Among_IN patients_NNS with_IN sickle_JJ cell_NN disease_NN ,_, the_DT acute_JJ chest_NN syndrome_NN is_VBZ commonly_RB precipitated_VBN by_IN fat_JJ embolism_NN and_CC infection_NN ,_, especially_RB community-acquired_JJ pneumonia_NN ._. Among_IN older_JJR patients_NNS and_CC those_DT with_IN neurologic_JJ symptoms_NNS ,_, the_DT syndrome_NN often_RB progresses_VBZ to_TO respiratory_JJ failure_NN ._. When_WRB we_PRP analyzed_VBD only_JJ patients_NNS who_WP had_VBD hypoxia_NN before_IN transfusion_NN -LRB-_-LRB- defined_VBN as_IN an_DT oxygen_NN saturation_NN of_IN less_JJR than_IN 91_CD percent_NN -RRB-_-RRB- ,_, the_DT values_NNS increased_VBD from_IN 86_CD percent_NN to_TO 93_CD percent_NN -LRB-_-LRB- P_NNP -LRB-_-LRB- 0.001_CD -RRB-_-RRB- ._. Respiratory_JJ failure_NN was_VBD documented_VBN in_IN 13_CD percent_NN of_IN patients_NNS ._. Neurologic_JJ events_NNS occurred_VBD in_IN 11_CD percent_NN of_IN patients_NNS ._. The_DT most_RBS common_JJ were_VBD altered_VBN mental_JJ status_NN -LRB-_-LRB- in_IN 56_CD percent_NN -RRB-_-RRB- ,_, seizures_NNS -LRB-_-LRB- 11_CD percent_NN -RRB-_-RRB- ,_, and_CC neuromuscular_JJ abnormalities_NNS -LRB-_-LRB- 8_CD percent_NN -RRB-_-RRB- ._. Anoxic_JJ brain_NN injury_NN occurred_VBD in_IN three_CD patients_NNS ,_, central_JJ nervous_JJ system_NN hemorrhage_NN in_IN three_CD ,_, and_CC infarction_NN in_IN three_CD ._. Respiratory_JJ failure_NN developed_VBN in_IN 46_CD percent_NN of_IN patients_NNS with_IN neurologic_JJ complications_NNS ;_: 92_CD percent_NN of_IN these_DT patients_NNS underwent_VBD transfusion_NN ,_, and_CC 23_CD percent_NN died_VBD ._. Half_NN the_DT complications_NNS consisted_VBD of_IN a_DT transient_JJ decline_NN in_IN oxygen_NN saturation_NN ._. Laryngospasm_NNP occurred_VBD in_IN 10_CD patients_NNS ,_, and_CC pneumothorax_NNS developed_VBN in_IN 2_CD patients_NNS ,_, both_DT of_IN whom_WP had_VBD had_VBN the_DT acute_JJ respiratory_JJ distress_NN syndrome_NN before_IN the_DT bronchoscopy_NN ._. Eight_CD patients_NNS underwent_VBD intubation_NN because_IN of_IN bronchoscopic_JJ complications_NNS ._. As_IN compared_VBN with_IN the_DT patients_NNS who_WP underwent_VBD bronchoscopy_NN without_IN incident_NN ,_, the_DT patients_NNS who_WP had_VBD complications_NNS were_VBD more_RBR likely_JJ to_TO present_VB with_IN dyspnea_NN -LRB-_-LRB- 61_CD percent_NN vs._IN 36_CD percent_NN ,_, P_NNP =_SYM 0.04_CD -RRB-_-RRB- and_CC higher_JJR peak_NN respiratory_NN rates_NNS -LRB-_-LRB- mean_NN ,_, 47_CD vs._IN 41_CD per_IN minute_NN ;_: P_NNP =_SYM 0.03_CD -RRB-_-RRB- ._. Eighteen_CD patients_NNS died_VBD ._. The_DT primary_JJ causes_NNS of_IN death_NN were_VBD respiratory_JJ failure_NN from_IN pulmonary_JJ emboli_NNS -LRB-_-LRB- bone_NN marrow_NN ,_, fat_NN ,_, or_CC thrombotic_JJ -RRB-_-RRB- in_IN six_CD patients_NNS and_CC bronchopneumonia_NN in_IN six_CD ._. The_DT causes_NNS of_IN death_NN in_IN the_DT remaining_VBG patients_NNS were_VBD pulmonary_JJ hemorrhage_NN ,_, cor_FW pulmonale_FW ,_, hypovolemic_JJ shock_NN from_IN splenic_JJ sequestration_NN ,_, overwhelming_JJ sepsis_NN ,_, intracranial_JJ hemorrhage_NN ,_, and_CC seizure_NN ._. Overall_RB ,_, infection_NN was_VBD a_DT contributing_JJ factor_NN in_IN 10_CD deaths_NNS ._. The_DT organisms_NNS identified_VBD included_VBN Streptococcus_NNP pneumoniae_NN ,_, Escherichia_NNP coli_NNS ,_, Haemophilus_JJ influenzae_NN ,_, legionella_NN ,_, cytomegalovirus_NN ,_, Staphylococcus_NN aureus_NN ,_, and_CC chlamydia_NN ._. A_DT specific_JJ cause_NN -LRB-_-LRB- either_CC pulmonary_JJ fat_JJ embolism_NN or_CC an_DT infectious_JJ agent_NN -RRB-_-RRB- was_VBD identified_VBN in_IN 256_CD -LRB-_-LRB- 38_CD percent_NN -RRB-_-RRB- episodes_NNS ._. As_IN in_IN previous_JJ studies_NNS ,_, we_PRP found_VBD that_IN involvement_NN of_IN the_DT lower_JJR lobes_NNS predominated_VBD ,_, radiographic_JJ abnormalities_NNS progressed_VBD ,_, and_CC oxygenation_NN and_CC hemoglobin_NN levels_NNS declined_VBD ._. Neurologic_JJ complications_NNS developed_VBN in_IN 22_CD percent_NN of_IN the_DT adults_NNS in_IN our_PRP$ study_NN ._. In_IN nearly_RB half_PDT these_DT patients_NNS ,_, respiratory_JJ failure_NN subsequently_RB developed_VBD ._. The_DT cause_NN of_IN the_DT acute_JJ chest_NN syndrome_NN was_VBD established_VBN in_IN 38_CD percent_NN of_IN episodes_NNS ,_, and_CC pulmonary_JJ infarction_NN was_VBD the_DT presumed_JJ cause_NN in_IN another_DT 16_CD percent_NN ._. In_IN contrast_NN to_TO prior_JJ studies_NNS ,_, in_IN which_WDT specific_JJ causes_NNS were_VBD seldom_RB identified_VBN ,_, infection_NN and_CC emboli_NNS were_VBD common_JJ causes_NNS in_IN our_PRP$ study_NN ._. Chlamydia_NNP was_VBD the_DT most_RBS common_JJ isolate_VBP and_CC was_VBD associated_VBN with_IN an_DT increased_VBN rate_NN of_IN vaso-occlusive_JJ events_NNS ,_, although_IN the_DT rate_NN of_IN infection_NN with_IN this_DT organism_NN was_VBD similar_JJ to_TO that_DT in_IN the_DT general_JJ population_NN ._. Mycoplasma_NNP and_CC viral_JJ pneumonia_NN ,_, including_VBG that_IN caused_VBN by_IN parvovirus_NN ,_, occurred_VBD in_IN all_DT age_NN groups_NNS ,_, but_CC predominated_VBD among_IN young_JJ children_NNS ._.