To the editor:
The pathogenesis of osteonecrosis in sickle cell anemia (SCA) remains unknown. Blood hyperviscosity has been suggested as a factor involved in the genesis of osteonecrosis,1 but has not been studied until now. We hypothesized that abnormal hemorheology could play a role in this complication. Hematologic and hemorheologic parameters were assessed in SCA patients with (OST+; n = 30) or without (OST−; n = 67) osteonecrosis. Osteonecrosis was diagnosed as previously described.2 The study was conducted according to the Declaration of Helsinki guidelines and was approved by the Regional Ethics Committee. The results are reported in Table 1. OST+ patients were older than OST− patients (P < .05) and more had a history of vaso-occlusive crises (VOC) within the previous year (P < .05) and a higher frequency of α-thalassemia (P < .05), confirming previous studies.3-5 Although the OST+ group exhibited higher hemoglobin (Hb) and hematocrit and a lower hemolytic component than the OST− group (P < .01), blood viscosity was not significantly different between the 2 groups (P < .20). In contrast, red blood cell (RBC) deformability (P < .001) and aggregation (P < .05) were increased in the OST+ group. The hydroxyurea (HU) treatment frequency was not significantly different between the 2 groups (P < .20). As HU is known to modulate RBC deformability,6 we analyzed the data as a function of HU therapy independently of osteonecrosis and found that HU-treated patients had lower blood viscosity and greater RBC deformability (data not shown). Excluding HU-treated patients from the cohort did not change the results (Table 1).
General characteristics and hematologic and hemorheologic parameters in patients with (OST+) and without (OST−) osteonecrosis
| . | With patients undergoing HU treatment . | Without patients undergoing HU treatment . | ||
|---|---|---|---|---|
| OST− (n = 67) . | OST+ (n = 30) . | OST− (n = 57) . | OST+ (n = 22) . | |
| Age (y) | 32.5 ± 12.2 | 39.3 ± 13.1* | 32.0 ± 12.4 | 38.8 ± 11.9* |
| Gender (male/female) | 32/35 | 11/19 | 27/30 | 9/13 |
| HU (%) | 15.9 | 27.6 | — | — |
| α-Thalassemia (%) | 37.3 | 56.7* | 40.4 | 59.1 |
| Positive history of VOC (%) | 9.0 | 26.7* | 8.8 | 27.3* |
| HbF (%) | 7.9 ± 5.7 | 9.6 ± 6.2 | 7.5 ± 5.6 | 9.1 ± 6.1 |
| WBC (109/L) | 9.5 ± 2.0 | 8.7 ± 2.1 | 10.0 ± 2.7 | 9.0 ± 1.7 |
| RBC (1012/L) | 2.8 ± 0.6 | 2.9 ± 0.5 | 2.8 ± 0.6 | 3.1 ± 0.4 |
| PLT (109/L) | 404 ± 126 | 381 ± 136 | 414 ± 125 | 373 ± 144 |
| MCV (fL) | 83.5 ± 9.8 | 86.6 ± 10.1 | 81.4 ± 8.2 | 83.5 ± 7.4 |
| MCHC (g/dL) | 35.9 ± 1.1 | 35.6 ± 1.2 | 35.8 ± 1.1 | 35.5 ± 1.3 |
| Hb (g/dL) | 8.2 ± 1.3 | 9.0 ± 1.1** | 8.1 ± 1.2 | 9.1 ± 1.1*** |
| Hct (%) | 22.9 ± 3.7 | 25.2 ± 3.0** | 22.7 ± 3.4 | 25.6 ± 3.1*** |
| RET (%) | 8.5 ± 3.3 | 7.7 ± 2.7 | 8.6 ± 3.3 | 7.7 ± 2.3 |
| BIL (μmol/L) | 61.9 ± 44.1 | 52.6 ± 37.4 | 62.8 ± 46.1 | 54.7 ± 43.0 |
| AST (IU/L) | 39.4 ± 14.8 | 37.0 ± 10.1 | 39.8 ± 14.3 | 37.1 ± 11.2 |
| LDH (IU/L) | 522 ± 166 | 433 ± 96** | 537 ± 161 | 442 ± 100** |
| Hemolytic component (relative unit) | 0.16 ± 1.10 | −0.35 ± 0.61** | 0.23 ± 1.08 | −0.33 ± 0.61** |
| ηb (mPa/s) | 7.64 ± 1.79 | 8.24 ± 2.01 | 7.80 ± 1.75 | 8.40 ± 2.16 |
| RBC deformability at 3 Pa (a.u. × 100) | 15 ± 6 | 20 ± 5*** | 15 ± 5 | 19 ± 5*** |
| RBC aggregation (%) | 52 ± 9 | 57 ± 8* | 52 ± 10 | 55 ± 7 |
| RBC disaggregation threshold (s−1) | 306 ± 148 | 262 ± 108 | 309 ± 152 | 265 ± 116 |
| . | With patients undergoing HU treatment . | Without patients undergoing HU treatment . | ||
|---|---|---|---|---|
| OST− (n = 67) . | OST+ (n = 30) . | OST− (n = 57) . | OST+ (n = 22) . | |
| Age (y) | 32.5 ± 12.2 | 39.3 ± 13.1* | 32.0 ± 12.4 | 38.8 ± 11.9* |
| Gender (male/female) | 32/35 | 11/19 | 27/30 | 9/13 |
| HU (%) | 15.9 | 27.6 | — | — |
| α-Thalassemia (%) | 37.3 | 56.7* | 40.4 | 59.1 |
| Positive history of VOC (%) | 9.0 | 26.7* | 8.8 | 27.3* |
| HbF (%) | 7.9 ± 5.7 | 9.6 ± 6.2 | 7.5 ± 5.6 | 9.1 ± 6.1 |
| WBC (109/L) | 9.5 ± 2.0 | 8.7 ± 2.1 | 10.0 ± 2.7 | 9.0 ± 1.7 |
| RBC (1012/L) | 2.8 ± 0.6 | 2.9 ± 0.5 | 2.8 ± 0.6 | 3.1 ± 0.4 |
| PLT (109/L) | 404 ± 126 | 381 ± 136 | 414 ± 125 | 373 ± 144 |
| MCV (fL) | 83.5 ± 9.8 | 86.6 ± 10.1 | 81.4 ± 8.2 | 83.5 ± 7.4 |
| MCHC (g/dL) | 35.9 ± 1.1 | 35.6 ± 1.2 | 35.8 ± 1.1 | 35.5 ± 1.3 |
| Hb (g/dL) | 8.2 ± 1.3 | 9.0 ± 1.1** | 8.1 ± 1.2 | 9.1 ± 1.1*** |
| Hct (%) | 22.9 ± 3.7 | 25.2 ± 3.0** | 22.7 ± 3.4 | 25.6 ± 3.1*** |
| RET (%) | 8.5 ± 3.3 | 7.7 ± 2.7 | 8.6 ± 3.3 | 7.7 ± 2.3 |
| BIL (μmol/L) | 61.9 ± 44.1 | 52.6 ± 37.4 | 62.8 ± 46.1 | 54.7 ± 43.0 |
| AST (IU/L) | 39.4 ± 14.8 | 37.0 ± 10.1 | 39.8 ± 14.3 | 37.1 ± 11.2 |
| LDH (IU/L) | 522 ± 166 | 433 ± 96** | 537 ± 161 | 442 ± 100** |
| Hemolytic component (relative unit) | 0.16 ± 1.10 | −0.35 ± 0.61** | 0.23 ± 1.08 | −0.33 ± 0.61** |
| ηb (mPa/s) | 7.64 ± 1.79 | 8.24 ± 2.01 | 7.80 ± 1.75 | 8.40 ± 2.16 |
| RBC deformability at 3 Pa (a.u. × 100) | 15 ± 6 | 20 ± 5*** | 15 ± 5 | 19 ± 5*** |
| RBC aggregation (%) | 52 ± 9 | 57 ± 8* | 52 ± 10 | 55 ± 7 |
| RBC disaggregation threshold (s−1) | 306 ± 148 | 262 ± 108 | 309 ± 152 | 265 ± 116 |
Values are means ± SD. All patients were at steady state at the time of the study, ie, no blood transfusions in the previous 3 months and absence of acute episodes at least two months before inclusion into the study. Measurements of 4 hemolytic markers (BIL, bilirubin; LDH, lactate dehydrogenase; AST, aspartate aminotransferase; RET, reticulocytes) were performed using standard methods, and a principal component analysis was used to derive a hemolytic component value from these markers.9 This standard statistical data reduction approach uses conventional clinical measurements to explain the maximum-shared variance among these indirect measures of hemolysis. The hemolytic component has recently been demonstrated to reflect intravascular hemolysis,9 had a mean of 0 (standard deviation = 1.0), and predicted 49.2% of the variation among all 4 measured variables (eigenvalue = 1.97). Blood viscosity, RBC deformability, and aggregation properties were determined as previously described.10 Polymerase chain reaction (Gap-PCR) was used to detect the 6 common α-thalassemia deletions, including −α3.7 and −α4.2 alleles, and triplication defects of the α-globin genes.
HbF, fetal Hb; Hct, hematocrit; MCHC, mean corpuscular hemoglobin concentration; MCV, mean cell volume; PLT, platelets; VOC, vaso-occlusive crisis; WBC, white blood cell; ηb, blood viscosity.
Significant difference between the 2 groups: *P < .05; **P < .01; ***P < .001.
A binary (OST−/OST+) multivariate logistic model was used to identify factors associated with osteonecrosis in SCA patients and included age, Hb, RBC aggregation and deformability, hemolytic component, α-thalassemia status, and previous history of VOC as covariates. The overall model was significant (χ2 = 30.192; df = 7; P < .0001) and retained age (odds ratio [OR]: 1.06; 95% confidence interval [CI]: 1.01-1.12; P < .05), Hb (OR: 2.24; 95% CI: 1.19-4.18; P < .05), and RBC deformability (OR: 1.15; 95% CI: 1.01-1.33; P < .05) as independent factors statistically associated with osteonecrosis. Two other binary multivariate logistic models were tested: one included the previous parameters plus blood viscosity and HU therapy and the other excluded all HU patients. The results were similar to those in the first model (data not shown).
Our study demonstrates that increased RBC deformability is associated with osteonecrosis in SCA. Irregularly shaped, deformable sickle RBCs were previously shown to be more adherent than rigid, irreversibly sickle RBCs,7 hence triggering vascular occlusion.8 The greater RBC deformability found in the OST+ group is probably caused by the greater frequency of patients with α-thalassemia in this group because patients with α-thalassemia had greater RBC deformability (0.18 ± 0.05) than patients without RBC deformability (0.15 ± 0.06, P < .05). Although higher Hb levels were observed in patients with osteonecrosis, the data do not support a significant role for blood viscosity in the pathogenesis of this complication, even after excluding data from patients under HU therapy. Further studies will be required to delineate the mechanisms by which RBC deformability raises the risk for osteonecrosis.
Authorship
Acknowledgments: The authors thank Dr Martine Torres for her critical review of the manuscript and editorial assistance. Y.L. is funded by the Region of Guadeloupe.
Contribution: N.L., Y.L., M.R., M.-D.H.-D., D.M., X.W., B.T., M.-L.L.-M., M.E.-J., and P.C. designed the research; N.L., Y.L., M.R., M.M.-M., M.-D.H.-D., D.M., X.W., and P.C. performed the experiments; N.L., Y.L., M.R., V.T., B.T. and P.C. analyzed the results; N.L., Y.L., M.R., M.M.-M., and P.C. interpreted the data; N.L., Y.L., M.R., and P.C. wrote the article; and N.L., Y.L., M.R., M.M.-M., M.-D.H.-D., V.T., D.M., X.W., B.T., M.-L.L.-M., M.E.-J., and P.C. read and approved the final version of the manuscript.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
Correspondence: Philippe Connes, INSERM UMR 665, Hôpital Ricou, CHU de Pointe-à-Pitre, 97157 Pointe-à-Pitre, Guadeloupe; e-mail: pconnes@yahoo.fr.
References
Author notes
Y.L. and M.R. contributed equally to this study.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal