Search

With direct sequencing of the amplified cDNA, we analysed the transcript and mRNA splicing defect in a common Chinese beta-thalassemia mutant (IVS-II nt. 654 C-->T). The result shows that this mutant gene would not only produce abnormally processed beta-globin mRNA, but also transcribes a small amount of normally spliced mRNA, hence leading to beta+ thalassemia. The method described herein provides a simple and sensitive approach to the studies of gene expression and molecular defects in genetic diseases at transcriptional level.

Hydroxyurea is one of several cytostatic agents that increase fetal hemoglobin (HbF) production in some patients with sickle-cell disease, although their mechanisms of action remain to be defined. We have studied the effects of hydroxyurea in several hospitalized patients with sickle-cell disease treated for 3 months, who were then maintained on outpatient therapy. Among hydroxyurea-treated patients, we found that about 75% respond with at least a twofold increase in the percentages of F reticulocytes and HbF. Among the responders, HbF levels increased twofold to 10-fold, with three patients achieving levels of 10% to 15%. Statistical analysis of the three cellular variables that determine HbF levels in patients with sickle-cell disease--namely, increased F-cell production, F-cell survival, and HbF/F cells--disclosed that HbF production, as measured by an increase in F reticulocytes, accounted for about 70% of the HbF elevation, with smaller contributions coming from augmentation of HbF/F cells and preferential survival of F cells. Four responders were re-treated with their optimal weekly hydroxyurea dose, given either in daily fractions or over 4 consecutive days, after a 1- to 3-month washout period. Greater HbF responses were attained with the optimal hydroxyurea dose than with the dose-regimen escalation, and usually occurred after a lag period. Furthermore, increases in HbF and F-cell levels were more rapid in patients receiving therapy on 4 out of 7 days rather than on a daily schedule. Our calculations show that the increases in HbF/F and F cells and the decrease in the fraction of dense cells during limited hydroxyurea administration should cause a significant improvement in intracellular sickle hemoglobin polymerization tendency. Controlled prospective trials are necessary to establish whether these effects lead to clinical benefit. Alternate schedules of hydroxyurea administration, or its use in conjunction with other means to elevate HbF or reduce mean cell hemoglobin concentration, may achieve greater inhibition of polymerization and thus be more likely to result in unequivocal amelioration of disease manifestations.

Hb Q-Thailand [alpha 74(EF3)Asp-->His] is often found in Thailand, China, and other Southeast Asian countries. The alpha-Q-Thailand gene is strongly linked to an alpha gene deletion and has important implications in the identification and diagnosis of hemoglobinopathies and thalassemias. The alpha-Q-Thailand mutation was previously mapped to the alpha 1 gene in a study of Chinese patients. In this paper, a Thai patient with Hb Q-Thailand/Hb H disease and his mother were studied at the DNA level, and the gene organization of Hb Q-Thailand in the Thai patient was found to be the same as that of Chinese patients (i.e. the Hb Q-Thailand gene is located on the alpha 1 gene of chromosome #16, while the -4.2 kb or leftward deletion involves the alpha 2 gene). Also, the GAC-->CAC mutation proposed at codon 74, has been confirmed by DNA sequencing and a simple and accurate method for diagnosis of the Hb Q-Thailand variant has been developed based on restriction enzyme analysis. Since the GAC-->CAC mutation generates new cutting sites for both restriction enzymes Apa LI and Hgi AI, polymerase chain reaction amplification of a specific region around codon 74, followed by digestion with these enzymes and agarose gel electrophoresis of the digested products, permits rapid and accurate identification of Hb Q-Thailand.

Intracoronary stents may potentially alleviate some of the problems associated with coronary angioplasty. Since the anatomy and physiology of swine coronary arteries closely resemble those of humans, the response to implantation of the Glanturco-Roubin, balloon-expandable, flexible stent was studied in this model. Additionally, the effect of aspirin, 1 mg/kg/day orally, on this response was evaluated. Eighteen Hanford minature swine underwent stenting of the left anterior descending coronary artery. Two died within 24 hours of stent implantation. The 16 survivors were put to death at 4 (n = 4), 11 (n = 4), 28 (n = 4), 56 (n = 3), and 180 (n = 1) days. Angiographically, reduction of stent lumen diameter of 0.1 to 1.3 mm was observed and was maximum at 11 days, with gradual improvement at subsequent time periods. Scanning electron microscopy, transmission electron microscopy, and light microscopy showed early disruption of subjacent endothelium, and adherence of platelets to exposed subendothelium and stent wires. Microthrombi were readily apparent. At 11 days, intimal thickening, made up predominantly of smooth muscle cells with abundant extracellular matrix, was observed and covered the stent wires. At 28 days, regression of intimal thickening was apparent and a confluent endothelium with flow-directed orientation was seen. At 56 and 180 days, the luminal surface was smooth; intimal thickening averaged 525 microns over the stent wires and 55 microns away from the wires. Findings in aspirin-treated animals were similar to results in those that did not receive aspirin. Thus in this swine model, stent implantation results in a time-dependent and self-limited vascular response.

Early thrombosis has complicated human stent implantation in several trials. To determine the best anticoagulation/antiplatelet therapy to maintain stent patency after percutaneous transluminal coronary angioplasty, we implanted the flexible balloon-expandable coil stent into the left anterior descending coronary artery of 28 atherosclerotic 8-month-old Hanford miniature swine. Animals were randomly assigned to one of three treatment groups: group A, aspirin (1 mg/kg/day) and dipyridamole (1 mg/kg three times a day); group B, aspirin and dipyridamole (same doses) plus Coumadin (dose required to prolong prothrombin time 1.3-1.5-fold that of normal); and group C, control. Adjuvant therapy was begun 3 days before stenting. Two pigs (one from group A and one from group B) died during implantation, both without thrombosis. Twenty-six animals survived until follow-up angiography and sacrifice at 1 month. No occlusive thrombosis of the stent occurred in survivors. Reduction of the stent lumen diameter was observed in every case at follow-up. Percent lumen reduction was 19% in group A, 26% in group B, and 24% in group C. Marked smooth muscle cell hyperplasia was seen by light and transmission electron microscopy at stent struts. Scanning electron microscopy of the luminal surface showed a variable morphology consisting of normal endothelium, adherent leukocytes, stellate periluminal cells, and occasional fibrin strands and red blood cells. Luminal narrowing was not affected by anticoagulation therapy, antiplatelet drugs, cholesterol level, or stent sizing. We conclude that occlusive thrombosis does not complicate stent implantation in this model but that substantial luminal narrowing due in part to smooth muscle hyperplasia does occur. The significance of luminal narrowing at the stent site requires further study.

Because fetal hemoglobin contains gammaglobin chains instead of beta chains, it is not affected by the genetic defect that causes sickle cell disease. Increased levels of fetal hemoglobin decrease the tendency toward intracellular polymerization of sickle hemoglobin that characterizes this disease. Hydroxyurea is one of several cytostatic agents that have been shown to increase the production of fetal hemoglobin in some patients with sickle cell disease. We studied the effects of hydroxyurea administration in 10 hospitalized patients with sickle cell disease, each of whom was treated for three months. Seven patients responded with a 2- to 10-fold increase in fetal hemoglobin, from a mean (+/- SD) of 1.6 +/- 1.6 percent of total hemoglobin to 6.8 +/- 4.7 percent; three patients had fetal-hemoglobin levels of 10 to 15 percent of total hemoglobin. Three did not respond to treatment. Four of the patients who responded were retreated with hydroxyurea after one to four months without treatment and were found to have larger increases in fetal-hemoglobin levels. In most patients, levels were still rising at the end of the study, even after 90 days of therapy. Fetal-hemoglobin levels tended to peak at dosages of hydroxyurea that were myelosuppressive. In the patients who responded to treatment, there were significant increases in the percentage of reticulocytes and erythrocytes containing fetal hemoglobin and in the amount of fetal hemoglobin within these cells. The percentage of dense red cells decreased in the patients who responded to treatment. The tendency toward intracellular polymerization at physiologic oxygen saturation was reduced by about 33 percent in the cells containing fetal hemoglobin, whereas there was no change in the other cells. We conclude that hydroxyurea is effective in increasing the production of fetal hemoglobin, which in this study was found to be associated with a small decrease in hemolysis and an increase in hemoglobin levels despite myelosuppression. Controlled, prospective trials are necessary to establish whether these effects will lead to clinical benefit.

With the technique of laser-Doppler velocimetry, cutaneous blood flows in the forearm of patients with stable sickle cell disease after graded periods of proximal ischemia were compared with normal subjects matched for age, race, and sex, and with patients with anemia caused by beta(+)-thalassemia. In the sickle cell patients the reactive hyperemia was characterized by an increased time interval between the release of the occlusion and the peak amplitude response (time-to-peak) and by a greater period of blood flow above the base-line value (payback ratio) compared with controls. In addition, prolongation of the occlusion period led to an augmentation in the magnitude of the characteristic basal flow oscillations or an induction of this phenomenon at sites not exhibiting it before ischemia. Base-line or ischemia-provoked flow oscillations of either this magnitude or frequency were only observed in normal or thalassemic controls during brief intervals in the rapidly decaying portion of the hyperemic response and in one subject with homozygous hemoglobin C disease. These results would support a model of a local integrative control of microcirculatory blood flow, which appears to become augmented, synchronized, and sustained in sickle cell subjects.

Warfarin sodium (Coumadin) has been used as an effective anticoagulating agent in human medicine for many years, although careful monitoring of its effects are necessary to avoid excessive anticoagulation. Previous experience with this drug for chronic anticoagulation therapy in miniature swine has been limited. The effect of warfarin sodium was studied by measuring prothrombin time in twelve 8-month-old Hanford miniature swine. The pigs had been fed a high-cholesterol diet and had undergone a prior coronary artery abrasion procedure for development of an atherosclerotic coronary disease model. Atherosclerosis was induced by feeding a high-cholesterol diet. Baseline prothrombin time ranged from 12.8 to 15.0 s (13.7 s mean). Prothrombin time was determined daily for the first 5 days of treatment and at least twice weekly thereafter until the animals were sacrificed. Animals received warfarin for 37-41 days. Prothrombin time could be increased 33-50% by once daily oral administration of warfarin 0.04-0.08 mg/kg. Oral administration of more than 0.08 mg/kg as a maintenance dose resulted in the death of two pigs. Most animals responded well to 0.08 mg/kg for the first 3 days of treatment followed by a maintenance dose of 0.06 mg/kg. Dosage was adjusted periodically when prothrombin times exceeded 50% above baseline. It is our experience that monitoring prothrombin time at least twice weekly and adjusting the maintenance dose can eliminate death losses due to warfarin intoxication.

With the availability of these noninvasive imaging modalities, one can now obtain more objective information about microcirculatory pathophysiology. As a result, previously held pathophysiologic paradigms in sickle cell disease will undoubtedly require modification or abandonment. These approaches should be viewed as supplements, not substitutes, for more traditional evaluations. Nonetheless, since these techniques are amenable to sequential application, these approaches promise to define disease severity more precisely, in quantitative terms, and should provide the means to follow the results of therapy unambiguously.

Irreversibly sickled cells (ISCs), considered by some to be of major pathophysiologic significance, have been reported to comprise between 5-50% of the total red cell population in patients with homozygous sickle cell anemia. Since the deformation of erythrocytes containing sickle hemoglobin is highly dependent on the concentration of hemoglobin in the deoxy conformation, any method established to enumerate the true ISC count requires the hemoglobin to be in the full oxy or liganded conformation. Because the oxygen dissociation curve for sickle erythrocytes is significantly shifted to the right, extremely high partial pressures of oxygen are required to approach full saturation. On the other hand, fully liganding the hemoglobin in the oxy conformation with carbon monoxide (CO) can be readily accomplished. We found that there is a significant reduction in the average number of sickled forms in the peripheral blood of sickle cell anemia patients after incubation in CO (to a value of 6.5 +/- 3.5%) when compared to conventional methods for ISC preparations. These results suggest that fully liganded erythrocytes should be used in quantitating ISCs in studies of the pathophysiology of this disease, especially since ISCs are likely to affect rheology differently from reversibly sickling cells.

The enzyme glucose-6-phosphate dehydrogenase (G6PD) was assayed in the circulating erythrocytes of 38 20-to 30-year-old subjects and 29 80- to 90-year-old subjects. The red blood cell G6PD activity of an aged-unbiased erythrocyte population in the young group was found to be significantly higher than in the older group (5.46 +/- 1.06) vs (4.31 +/- 0.99) (P less than 0.001), respectively (mean +/- SD). No significant difference was observed between the mean enzyme values of the male and female subjects in the two age groups. Since the mature red blood cell is devoid of regenerative capabilities, one may propose an explanation for this observation that implicates either an alteration in G6PD synthesis in the erythrocyte precursor cells in aged hosts, and/or the presence of an inhibitory factor that has the effect of decreasing the assayed activity for this enzyme.

We have demonstrated that the extent of intracellular polymerization of deoxyhemoglobin S can be predicted from knowledge of intracellular hemoglobin concentration, composition and oxygen saturation. Furthermore, we have demonstrated that polymer, which appears to be the main determinant of abnormal red cell rheology, can be detected in sickle erythrocytes at high oxygen saturation values and is not significantly affected by membrane and other cellular constituents. Some of the factors which modify the pathophysiology of sickle cell anemia can be classified as genetic or cellular. To analyze in more detail the genetic factors, we examined 12 sickle syndromes. When the effects of these genotype differences are analyzed for their changes in hemoglobin composition and concentration, we found that polymer formation can account for 80% of the variation in hemolytic and clinical severity. Cell heterogeneity can also modify polymer formation. The premature increases in erythrocyte density (intracellular hemoglobin concentration) in sickle cell anemia increases polymerization tendency. Homozygous alpha-thalassemia in sickle cell patients reduces this increase in cell heterogeneity and improves the hemolytic aspect of the sickle cell disease. For homozygous sickle cell patients we find that the broader density distributions (higher degree of cell heterogeneity) are associated with those cell populations with greater tendency of polymer formation. However, the major utility of our knowledge of intracellular polymerization appears to be its value in defining quantitatively the goals of the major therapeutic approaches with respect to how much inhibition of polymerization would be necessary to achieve various levels of amelioration of disease processes. The primary determinant of the amount of polymer formation within the SS erythrocyte is the extent of oxygen saturation. We measured intracellular polymer formation in SS erythrocytes using carbon-13/proton double nuclear magnetic resonance. As the oxygen saturation is decreased below about 90% oxygen saturation, we begin to see the appearance of polymer which steadily increases with decreasing oxygen saturation. The total intracellular hemoglobin concentration also affects the amount of polymer formed. By examining polymer formation in fractionated subpopulations of SS erythrocytes at various density values (or intracellular hemoglobin concentrations) we demonstrated that the polymer fraction increased with increasing intracellular hemoglobin concentration for any given oxygen saturation.(ABSTRACT TRUNCATED AT 400 WORDS)

Neurologic dysfunction is a significant source of morbidity and mortality in the sickle cell diseases, occurring with a prevalence of 6% to 34%. Because changes in brain glucose metabolism may precede gross functional or morphologic alterations, we recently applied the technique of positron emission tomography with fluorodeoxyglucose F 18 in an exploratory study to compare six patients with sickle cell disease without prior neurologic abnormalities (and with normal cranial computed tomographic scans) with six healthy age-matched controls, with respect to overall and regional cerebral metabolic rates for glucose. We found no significant difference in the global metabolic rates for the two groups. However, we observed an unusual clustering of abnormal regional cerebral metabolic rates for glucose in the frontal lobes of these subjects. These results support previous observations that frontal lobe involvement may be quite prevalent in sickle cell disease, even among individuals with normal computed tomographic scans.

Although the pathophysiologic manifestations of sickle cell disease have been assumed to result from microvascular occlusion consequent to in situ sickling of erythrocytes, actual blood vessel obstruction have been rarely demonstrated in vivo. Recent observations utilizing sophisticated biophysical techniques to study the intracellular hemoglobin S polymerization process has led to major revisions in this previously held pathophysiologic paradigm, but in vivo correlations are still lacking. With the development of new noninvasive imaging and para-imaging methods, it is now technically possible and feasible to characterize both regional organ perfusion and tissue biochemistry in quantitative terms. In addition, these modalities promise to clarify pathogenesis of the disease through definition of the events responsible for the progression from tissue ischemia and infarction through the resolution phase. Since these noninvasive techniques are amenable to sequential applications, they should facilitate objective evaluations of clinical trials of therapeutic agents designed to prevent or delay the vaso-occlusive manifestations of sickle cell disease.

We have applied the technique of laser-Doppler velocimetry to compare patterns of cutaneous blood flow in the forearms of patients with stable sickle-cell disease, with the patterns in normal subjects matched for age, race, and sex, and in patients with anemia due to beta+-thalassemia. The mean resting blood flow in the patients with sickle-cell disease was comparable to that of the control groups but was associated with large, local oscillations in flow with periods of 7 to 10 seconds and peak-to-trough magnitudes about half the mean flow. Oscillations occurred simultaneously at sites separated by 1 cm but were independent in phase and frequency. Since these laser-Doppler measurements represent the average flow pattern in about 1 mm3 of skin (i.e., in approximately 50 to 70 capillary loops), these results suggest that microcirculatory flow in patients with sickle-cell disease proceeds by synchronization of rhythmic flow in large domains of microvessels. These findings indicate that periodic flow may be a compensatory mechanism to offset the deleterious altered rheology of erythrocytes containing polymerized hemoglobin S, and suggest that laser-Doppler velocimetry may be a useful method to investigate microvascular physiology in patients with sickle-cell disease.

The use of miniature swine as a model for cardiovascular diseases of humans is becoming more popular for many reasons. One of the problems involved in using swine is their propensity for fatal cardiac arrhythmias during surgical procedures requiring general anesthetics, especially cardiac procedures. In preparation for use as a model of human atherosclerotic coronary artery disease, 30 three-month-old (15 kg) Hanford miniature swine underwent left heart catheterization, coronary angiography, and abrasion of the left anterior descending coronary artery. All pigs were treated with diltiazem HCl 30-60 mg (2-4 mg/kg) three times daily and aspirin 25 mg once daily, both given orally for three days before surgery. General anesthesia was induced with a combination of ketamine HCl 25 mg/kg, atropine sulfate 0.1 mg/kg, and acepromazine 0.22 mg/kg, all given intramuscularly. Halothane 1-2% and nitrous oxide 30% were used to maintain general anesthesia after endotracheal intubation. After successful cannulation of the femoral artery, all animals were given 200 units/kg heparin and 5 mg/kg bretylium tosylate intravenously. The electrocardiogram and mean blood pressure were monitored throughout the procedure. No significant change in blood pressure was noted during the procedure. Intracoronary nitroglycerin 200 micrograms was administered prior to abrasion of the left anterior descending with a 2.0-mm angioplasty balloon. The pigs were allowed to recover after routine closure of the incision. One pig (1/30) died during the abrasion procedure as a result of ventricular fibrillation. This represents a low, acceptable mortality using the present regimen compared with other regimens.

An increasing body of experimental evidence demonstrates that intracellular hemoglobin concentration and composition is a primary determinant of pathophysiology in sickle cell disease. To quantitate more precisely the heterogeneous distribution of intracellular hemoglobin concentrations in a given individual with this disease, we have calibrated the phthalate ester separation technique by using discontinuous Stractan density gradients to isolate subpopulations of red cells of relatively uniform corpuscular hemoglobin concentration values. We find that blood from individuals with sickle cell anemia exhibits a markedly broader distribution of corpuscular hemoglobin concentration values, containing both very light and very dense cells, than the red cell density profile from normal individuals. This increased breadth of cell densities in patients with sickle cell anemia remains even after exclusion of the very light and very dense subpopulations. In patients with stable sickle cell anemia, there appears to be minimal variation in the distribution of cell densities that are unimodal but skewed toward higher density values. The phthalate ester method can conveniently be used to follow changes in cell densities during vaso-occlusive events, to monitor therapy targeted at modifying intracellular hemoglobin S concentrations, and in sequential applications in large field trials designed to determine the relationship between red cell heterogeneity and specific manifestations of the sickle cell syndromes.

The single and multiple dose pharmacokinetics of oral nifedipine capsules, 10 mg, have been examined in five patients with peripheral vasospasm. After a single dose, nifedipine was rapidly absorbed in three and slowly absorbed in two patients. Mean bioavailability parameters included a tmax of 2.9 h, a Cmax of 33.3 ng ml-1 and a AUC0-8 h of 113.3 ng h ml-1. After multiple dosing with either 10 or 20 mg every 8 h for 10 days the mean tmax at steady state was 2.1 h while the mean dose-corrected (to 10 mg) Cmax and AUC0-8 h were 51.9 ng ml-1 and 146.9 ng h ml-1, respectively. The mean elimination rate constant was 0.173 h-1 after both single and multiple doses. The mean extent of accumulation of nifedipine, defined as the ratio of AUC0-8 h (steady state)/AUC0-8 h (single dose), was 1.3; we concluded that nifedipine accumulates in the body when it is administered every 8 h. This should be taken into account when predicting steady state serum concentrations and haemodynamic effects of nifedipine from single dose kinetic data.

An 86-year-old black woman admitted for an elective cataract extraction was found to have moderate hypochromic microcytic anemia. Hematologic evaluation disclosed the presence of hemoglobin SC disease and heterozygous alpha-thalassemia-2 (alpha alpha/alpha-). A red cell density profile of the patient's peripheral blood revealed an absence of the typical uniform shift toward higher-density values seen in hemoglobin SC disease, indicating a "normalization" in the distribution of intracellular hemoglobin types. It is suggested that the co-inheritance of hemoglobin SC disease and heterozygous alpha-thalassemia-2, probably by decreasing the tendency toward intracellular hemoglobin S polymerization, contributed to her prolonged survival and relatively mild clinical course.

A consecutive series of 22 stable adult inpatients with sickle cell anaemia were examined for the presence and severity of spontaneous 'comma' signs of the conjunctiva. Fifteen patients had severe conjunctival signs (more than 10 commas in the worse eye). The presence of severe conjunctival signs was associated with a broader distribution of intraerythrocytic haemoglobin concentrations (p = 0.0005). The patient group with severe conjunctival signs was not found to be significantly different from the group without such signs for age, sex, haemoglobin value, reticulocyte count, alpha-globin gene number, percentage fetal haemoglobin, or the proportion of very dense cells (CHC greater than 37 g/dl). Thus the singular heterogeneity of the erythrocytes in sickle cell disease may be indicative of the factor(s) responsible for the diagnostic comma sign.

Individuals with sickle cell disease and its genetic variants demonstrate a modest to severe hemolytic anemia. They are also prone to both acute and chronic vaso-occlusive phenomena, which may lead to irreversible tissue damage, organ failure, and premature death. To better understand these vaso-occlusive phenomena, it would be useful to determine the extent to which arteriolar microvascular obstruction might contribute to the pathogenesis of sickle cell disease. We studied the hematologic and ocular effects of orally administered nifedipine in several steady-state patients and compared these with a case-control group of patients not receiving nifedipine. In the nifedipine-treated patients, a striking reversal in ischemic retinal and conjunctival changes was observed, as well as a significant improvement in color vision performance. In addition, the treated subjects showed a slight but significant decline in the levels of both indirect serum bilirubin and plasma hemoglobin, suggesting a concurrent amelioration in the rate of hemolysis. These observations are consistent with the hypothesis that inappropriate vaso-constriction, or frank vasospasm, perhaps in response to the altered rheology of red cells containing polymerized hemoglobin S, may play an important role in the pathophysiology of sickle cell disease. Furthermore, selective vasodilation may prove to be a useful adjunctive approach to the treatment of the manifestations of the sickle syndromes.

Methodologies now exist that promise to disclose fundamental and critical information about the microvascular status of patients with sickle cell disease. Several of these modalities are in an embryonic stage of development, whereas others are close to being ready for application to clinical trials. It is important that these sophisticated diagnostic techniques not supplant our clinical approach to the patient. However, it is hoped that these strategies will allow us to understand better the extreme spectrum of disease manifestations, as well as to provide objective means of monitoring the response to therapy in patients with sickle cell disease.

Cryoprecipitate was administered to a patient with severe adult respiratory distress syndrome secondary to an amniotic fluid embolus, diagnosed cytologically. Following the administration of cryoprecipitate, cardiopulmonary and hematologic status markedly improved, and the patient recovered without apparent sequela. She is the sixth surviving patient reported to have an amniotic fluid embolus confirmed cytologically. On the basis of accumulating data on the relationship between fibronectin levels and the integrity of the reticuloendothelial system, it is quite possible that fibronectin (cold-insoluble globulin), and not fibrinogen, played the key role in her dramatic improvement and may well have been responsible for the clinical improvement in earlier patients treated with blood products containing fibronectin.

The extent of intracellular polymerization of hemoglobin S, leading to loss of erythrocyte deformability and eventual morphological sickling, is primarily determined by oxygen saturation and intracellular hemoglobin concentration and composition. Epidemiological analysis of sickle cell disease severity among the sickle syndromes and studies of the biophysics of intracellular polymerization were used to estimate the potential clinical benefit of various therapeutic strategies. These strategies include those designed to increase deoxyhemoglobin S solubility; to increase erythrocyte volume or water content, thereby reducing the intracellular hemoglobin concentration; or, most recently, to decrease the proportion of hemoglobin S by increasing the amount of non-S hemoglobin. Increasing levels of hemoglobin F is of particular interest due to its "sparing" effect in inhibiting polymerization, the well-characterized epidemiological associations of high levels of hemoglobin F with reduced disease severity, and recent studies of drug-induced increases in hemoglobin F. Our analyses of equilibrium polymer formation at physiological oxygen saturation values suggest that small decreases in polymer formation at intermediate levels of hemoglobin F may give rise to a small decrease in anemia (as associated with homozygous alpha-thalassemia coexistent with sickle cell anemia), but that greater reductions in polymer formation may be necessary to effect a significant improvement in disease severity. Current studies of hydroxyurea-induced increases of hemoglobin F give cautious optimism that therapeutically useful levels may be attainable.

We report 2 patients with sickle cell anemia who showed retroequatorial localized red retinal lesions. In 1 patient these lesions were associated with closure of the retinal precapillary arteriole and capillary bed. In both patients these lesions receded during the course of therapy with a selective arteriolar vasodilator, nifedipine. These observations are discussed in relation to the possible pathogenesis of the retinopathy seen in patients with sickle cell anemia.

Homozygous alpha-thalassemia has the beneficial effect in sickle cell anemia of reducing the hemolytic severity while changing several other hematological parameters. We examined in detail the cellular basis of some of these hematologic alterations. We find that the broad distribution in erythrocyte density and the large proportion of dense cells associated with sickle cell anemia are both reduced with coexisting alpha-thalassemia. Measurements of glycosylated hemoglobin levels as a function of cell density indicate that the accelerated increase in cell density, beyond normal cell aging, in sickle cell anemia is also reduced with alpha-thalassemia. The patients with homozygous alpha-thalassemia and sickle cell disease have slightly lower levels of hemoglobin F than the nonthalassemic sickle cell patients. Examination of hemoglobin F production revealed that the proportion of hemoglobin F containing reticulocytes remained unchanged, as did the proportion of hemoglobin F in cells containing hemoglobin F (F cells). Preferential survival of F cells occurs in sickle cell anemia, with or without alpha-thalassemia, and the slight difference in hemoglobin F levels appear to reflect differences in numbers of circulating F cells. Thus, in sickle cell disease with coexisting alpha-thalassemia, the change in the erythrocyte density profile, possibly due to inhibition of polymerization-related increases in cell density, explains the hematological improvement.