Inglés (pdf)
Articulo en XML
Referencias del artículo
Traducción automática
Enviar articulo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
INTRODUCTION
Sickle cell disease is a common hemoglobinopathy, and approximately 4,000 people in South Carolina, USA, live with this disease (SCD) 1. In the management of SCD, chronic transfusion therapy is used to prevent and treat complications 2. Each unit of transfused packed red blood cells (pRBCs) provides 200-250 mg of iron, and repeated blood transfusions will lead to iron overload (IO) in SCD patients. Previous studies have shown that the IO resulting from transfusional therapy in other patient populations is associated with significant morbidity and mortality 3. Iron overload has been considered a major cause of end-organ damage in multitransfused patients with hemoglobinopathies 4, and the source of excessive iron burden in SCD is primarily blood transfusions and intravascular hemolysis.
To assess the degree of IO, liver iron concentration (LIC) can be measured invasively via liver biopsy or noninvasively by magnetic resonance imaging (MRI). There is a very close correlation between the two methods; thus, MRI has largely eliminated the need for liver biopsies 5. Ferritin should be used as a measure of IO in SCD patients, but its values should be interpreted with caution for therapeutic decision-making 6. Iron accumulation depends on the age at which blood transfusions are started, the rate of transfusions, and the nature of the transfusion regimen7. Compared with thalassemia, iron deposition in cardiac, renal, or endocrine organs is lower in SCD because intravascular hemolysis promotes biliary and urinary elimination of iron as hemosiderin, heme, and hemoglobin, and the chronic inflammatory state reduces toxic accumulation of iron in macrophages 7-9. Additionally, there is a difference between multiple simple transfused SCD patients and patients receiving blood with exchange transfusions because there is less liver accumulation of transfused iron in the latter 8,9.
It is complicated to establish if organ damage is caused by iron from transfusion as a treatment for some SCD comorbidities or if damage is a consequence of the complications themselves 5,6. The non-transferrinbound iron free in plasma is toxic because it produces radicals with oxidation products that are responsible for a significant part of iron accumulation and cell injury associated with regular multiple transfusions. Additionally, inflammation, as a pathophysiologic mechanism in SCD, leads to increased hepcidin synthesis and decreased iron absorption, with increased iron retention in the reticulo-endothelial system 9.
IO is known to increase morbidity and mortality in SCD, although the exact mechanism is unclear 10-12. This risk of increased mortality could be associated with a high rate of comorbidities 13. Except for the comorbidities of stroke or the presence of an abnormal transcranial doppler (TCD), whose preventive treatment is based on chronic transfusions 14, it is difficult to determine whether IO induces the presence of comorbidities or whether these comorbidities are independent of IO. It has been shown that in groups of SCD patients with IO, there was a higher mortality rate than those without IO. Those deaths were attributable to sudden death or pneumonia associated with acute chest syndrome 13. The rate of admission for vasooclusive crisis related to the prevalence of organ damage and SCD complications only found a meaningful relationship with IO and acute chest syndrome (ACS) 15. Ferroptosis has also been recognized as a novel mechanism in SCD and an additional avenue for organ damage in SCD patients. In ferroptosis, elevated iron levels trigger cell death by increasing reactive oxygen species (ROS) and lipid peroxides, leading to organ impairment 16.
Iron chelation is the primary treatment for IO in SCD, and it can be initiated if more than 18 transfusions have been received (or >120 cc/kg of PRBC) within a defined period. Additionally, a serum ferritin level >1000 ng/mL on two separate measurements, or hepatic MRI-quantified iron>3 mg/g of dry weight, indicates IO treatment. Because SCD causes a degree of chronic inflammation, ferritin, an acute-phase reactant, is less reliable as a diagnostic tool than in other anemias.
There are three iron chelators currently approved for clinical use: (1) deferoxamine (DFO), (2) deferiprone (DFP), and (3) deferasirox (DFX). DFO is the oldest drug and is administered intravenously or subcutaneously, while DFP and DFX are taken orally 14,15.
The primary aim of this study was to determine the relationship between iron overload and comorbidities in SCD patients. The secondary aims included evaluating the association between liver iron concentration (LIC), serum ferritin levels, liver enzymes, and the characteristics of the iron chelation regimen in chronically transfused SCD patients.
MATERIAL AND METHODS
We conducted a retrospective chart review of the electronic medical records (EMRs) of 245 adults with SCD who receive care at MUSC in Charleston, South Carolina, and are enrolled in the Sickle Cell Disease Implementation Consortium (SCDIC) 17. The study participants had a diagnosis of SCD, were English-speaking, and aged 15-50. From that population, we identified 85 subjects with both SCD and iron overload (34.7%). EMR data were collected over seven years (2017-2024). The inclusion criteria for this study were a diagnosis of SCD, enrollment in the MUSC-SCDIC-II Registry and REAL Answers projects, and a diagnosis of IO using the previously mentioned parameters. Age, sex, genotype, blood ABO group, comorbidities, ferritin, LIC, liver enzymes, and iron chelation treatment were obtained from the EMR. Patients with iron overload who received a hematopoietic stem cell transplant were excluded from this study.
Institutional review board (IRB) approval was obtained from each of the eight study sites and a central IRB (CIRBI/Advarra) prior to data collection, and a written informed consent was obtained from each participant.
Statistical Analysis
Statistical analysis was performed using GraphPad InStat3 (GraphPad Software, Boston, MA 02110) to analyze frequency distributions for categorical variables and to perform linear regression and Pearson correlation for continuous variables. A 2 x 2 contingency table was used to calculate odds ratios with 95% confidence intervals by Fisher’s exact test and to calculate sensitivity and specificity parameters. The reference group for the odds ratio (OR) was adult SCD patients (160/245) without iron overload. A p<0.05 was considered statistically significant.
RESULTS
Table 1 describes the demographic characteristics of the 85 (34.7%) subjects with SCD and IO who met inclusion criteria. Seventy-three IO patients (85.9%) were between 18- 45 years old with a median age of 32.5 years (range 16-55), 61.2% were female, 95.3% had SS genotype, and 56.5% had blood group O. From 67.0% with IO criteria receiving transfusions (57/85), 51% (29/57) were treated under manual blood exchange regimen, and the rest of them by erythrocytapheresis regimen.
Table 1 Demographic characteristics of Iron overload in Sickle Cell Disease.
| Characteristics | Iron Overload (N=85) | No Iron Overload (N=160) | Total (N=245) |
|---|---|---|---|
| Age group (years) | |||
| <18 | 4 (4.7%) | 10 (6.3%) | 14 (5.7%) |
| 18-24 | 20 (23.5%) | 22 (13.7%) | 42 (17.2%) |
| 25-34 | 24 (28.2%) | 65 (40.6%) | 89 (36.3%) |
| 35-45 | 29(34.2%) | 43 (26.9%) | 72 (29.4%) |
| >45 | 8(9.4%) | 20 (12.5%) | 28 (11.4%) |
| Median | 32.5 | 31.0 | 32.0 |
| Gender | |||
| Male | 34 (40.0%) | 62 (38.8%) | 96 (39.2%) |
| Female | 51 (60.0%) | 98 (61.2%) | 149 (60.8%) |
| Age (N) | |||
| Male | 31.36±10.4 (34) | 34.36±11.0 (62) | 33.33±10.8(96) |
| Female | 33.43±9.4 (51) | 32.51±11.1 (98) | 32.75±10.7(149) |
| Mean±SD | 32.61±9.8 (85) | 33.26±11.2 (160) | 33.05±10.7 (245) |
| SCD Genotype | |||
| SS | 81 (95.3%) | 98 (61.3%) | 179 (73.1%) |
| SC | 1 (1.2%) | 40 (25.0%) | 41 (16.7%) |
| SThal | 3 (3.5%) | 19 (11.9%) | 22 (9.0%) |
| Other | 0 (0%) | 3 (1.8%) | 3 (1.2%) |
| Blood Group (ABO) | |||
| Group O | 48 (56.5%) | ||
| Group A | 20 (23.5%) | ||
| Group B | 10 (11.8%) | ||
| Group AB | 7 (8.2%) |
The prevalence of several comorbidities with SCD is described in Table 2. Patients with IO were noted to have a higher rate of four comorbidities: acute chest syndrome (OR= 2.46, p= 0.003), deep vein thrombosis (DVT, OR= 1.84, p= 0.04), stroke (OR= 14.67, p= 0.0001) and pulmonary hypertension (PH, OR= 4.75, p= 0.0006), seen in Table 2 and Fig. 1. Pulmonary hypertension (PH) was defined as a mean pulmonary arterial pressure of at least 25 mm Hg, and a TRV >2.5m/s.
Table 2 Sickle Cell Disease Iron overload and comorbidities relationships.
| Comorbidity | IO (N=85) | No IO (160) | Total (N=245) | OR (CI95%) | OR p-value |
|---|---|---|---|---|---|
| ACS | 65 (26.5%) | 91 (37.1%) | 156(63.7%) | 2.46 (1.37-4.45) | 0.0032 |
| DVT | 33 (12.2%) | 41 (18.0%) | 74(30.2%) | 1.84 (1.05-3.23) | 0.0406 |
| Retinopathy | 13 (5.3%) | 42 (17.1%) | 55 (22.4%) | 0.50 (0.26-1.01) | 0.0548 |
| Stroke* | 46(18.8%) | 13 (5.3%) | 59(24.1%) | 13.34 (6.56-27.12) | 0.0001 |
| CKD | 20 (7.8%) | 25 (10.2%) | 45 (17.9%) | 1.66 (0.86-3.21) | 0.1650 |
| Anxiety/Depression | 39 (16.7%) | 61 (23.3%) | 100(40.0%) | 1.38 (0.80-2.34) | 0.2752 |
| AVN | 42 (17.1%) | 62(25.3%) | 104 (42.4%) | 1.54 (0.91-2.63) | 0.1351 |
| Pulmonary Htn | 17 (7.8%) | 8(2.9%) | 25 (10.7%) | 4.75 (1.96-11.54) | 0.0006 |
ACS: Acute Chest Syndrome; DVT: Deep Vein Thrombosis; CKD: Chronic Kidney Disease; AVN: Avascular Necrosis; Htn: Hypertension. *Include overt stroke and abnormal transcranial doppler (TCD).
MRI studies, liver biopsy, LIC, and ferritin
Seventy-four percent of subjects with IO initially had MRI/LIC records available (63/85) and 26% had also liver biopsy data available (22/85). Twenty-four percent (20/85) of them did not have a subsequent record of LIC, as measured by liver MRI or liver biopsy, and were followed only with ferritin determination. Cardiac MRI data were available for 25% (21/85) of subjects, and only one patient showed myocardial iron deposition.
The average values at the start and end of the study for LIC were 11.71±8.46 mg/g dry weight (n=59) and 12.88±7.86 mg/g dry weight (n=44), respectively; the average values for ferritin were 3779.5±2852 ng/mL (n=85) and 5055.1±5567.4 ng/mL (n=85), respectively; p=0.06 for both.
There was a strong correlation between LIC (mg/g dry weight) and ferritin level (ng/mL) (r=0.5148, p=0.0001), as seen in Fig. 2. We also observed a strong correlation between serum ferritin and LIC in individuals with serum ferritin levels > 2500 ng/mL (r=0.5220, p=0.0026, n= 31). There was no correlation in patients with serum ferritin levels below 2500 ng/mL (r=0.3133, p= 0.3213; n=12).
Fig. 2 Linear regression with 95%CI between LIC (mg/g dry weight) and Ferritin (ng/mL) in SCD participants (n=59) diagnosed as Iron overload, r= 0.5148, p<0.0001.
An increasing number of comorbidities was associated with rising LIC and serum ferritin (Table 3). The median number of co-morbidities per patient was 3 (mean 3.12 ± 1.5).
Table 3 Values of LIC and ferritin associated with the number of comorbidities in subjects with Iron overload.
| No comorbidities | LIC (N=43) | Ferritin (N=85) | p |
|---|---|---|---|
| 1 | 11.58±5.2 (7) | 5538.28±7793.6 (9) | 0.0661 |
| 2 | 14.82±7.6 (17) | 5748.89±5336.7 (29) | 0.0001 |
| 3 | 12.83±9.6 (6) | 7026.32±7608.4 (15) | 0.0031 |
| 4 | 7.74±4.37 (5) | 2588.65±2552.1 (14) | 0.0023 |
| >5 | 11.58±8.17 (8) | 3936.66±3800.9 (18) | 0.0004 |
LIC: liver iron concentration (mg/g dry weight), ferritin (ng/mL).
Accuracy estimation of ferritin as a marker of iron overload in SCD participants
Among 69 IO subjects with paired LIC and ferritin levels, 36 had a LIC≥ 10 mg/g dw and 23 had a LIC < 10 mg/g dw. Using a ROC curve and a serum ferritin cut-off of ≥2500 ng/mL, we obtained 78% sensitivity, 74% specificity, 86% positive predictive value, and 61% negative predictive value. In our patient sample, 46 subjects were correctly diagnosed with IO based on their serum ferritin levels (TP), and 13 patients were incorrectly identified as not having IO (FN). The accuracy of the serum ferritin level to diagnose IO was 76% and the area under the curve (AUC) was 76% (Table 4), (Garcia-Casal et al. 18).
Table 4 Sensitivity and specificity of ferritin as a marker in Sickle Cell Disease to assess iron overload (N=69).
| Variable | Value | 95% Confidence Interval |
|---|---|---|
| Sensitivity | 0.7750 | 0.6159 to 0.8917 |
| Specificity | 0.7368 | 0.4879 to 0.9085 |
| Positive Predictive Value | 0.8611 | 0.7052 to 0.9533 |
| Negative Predictive Value | 0.6087 | 0.3856 to 0.8027 |
| Likelihood Ratio | 2.945 | |
| DOR | 9.65 |
DOR: diagnostic odds ratio, calculated as the effectiveness as an index of iron overload 18. DOR= (sens x spec) / (1-sens) x (1-spec) = (sens x spec) / (FNs) x (FPs).
Iron overload and liver enzymes
In SCD individuals with IO, liver enzymes showed a good correlation with LIC and ferritin levels (Supplemental Table 5), except for alkaline phosphatase and LIC. The mean serum values ± SD were AST: 47.8±26.32 U/L, ALT: 33.86±27.36 U/L, and AP: 113.66±55.12 U/L.
Table 5 Correlations between LIC and ferritin versus liver enzymes.
| Variable (N) | r | p |
|---|---|---|
| LIC vs AST (43) | 0.5551 | 0.0001 |
| LIC vs ALT (43) | 0.5248 | 0.0003 |
| LIC vs AP (43) | 0.1393 | 0.3730 |
| Ferritin vs AST (85) | 0.5794 | 0.0001 |
| Ferritin vs ALT (85) | 0.6015 | 0.0001 |
| Ferritin vs AP (85) | 0.3644 | 0.0006 |
LIC: liver iron concentration; AST: aspartate transferase; ALT: alanine aminotransferase; AP: alkaline phosphatase. The values of liver enzymes were expressed in U/L, LIC in mg/g dry weight, and ferritin in ng/mL. The reference ranges for liver enzymes were AST (5-34 U/L), ALT (5-45 U/L) and AP (35-150 U/L).
Iron chelation treatment
Transfused patients with IO received an average of 22 pRBCs per year with a median of 12 transfusions. Eighty-six percent (74/85) were prescribed chelation therapy with deferasirox (DFX), and two patients were prescribed deferiprone (DFP) chelation therapy. Of those on chelation therapy with deferasirox, only 19% (14/74) obtained a good response with serum ferritin <1000 ng/mL. Providers followed 54% of these patients with periodic LIC determinations and serum ferritin (40/74), and the remaining patients were followed with serum ferritin alone. Information about the side effects of DFX treatment was scarce and difficult to collect from the data records. Fifty-four percent (40/74) of the subjects were on disease-modifying treatment.
DISCUSSION
These findings show a significant association of IO with several comorbidities in a cohort of SCD patients. Most of the patients were SS genotype young adults receiving frequent blood transfusions. Parameters such as LIC, measured by MRI and/or liver biopsy, were used to diagnose IO in many of these patients, and serum ferritin levels were used as a marker of IO. There was a strong correlation between LIC and serum ferritin when the serum ferritin levels were above 2500 ng/mL. Liver enzymes showed a good correlation with LIC and serum ferritin levels, and chelation therapy with DFX was unsuccessful in most cases.
Previous research has demonstrated that SCD patients with IO have an increased risk of mortality 11,12, and this risk could be associated with a high rate of comorbidities13. Moreover, liver injury is associated with mortality in SCD, with increased ferritin and direct bilirubin as predictors of mortality. All patients with advanced liver fibrosis had IO, but not all patients with IO had fibrosis 19. We have seen an increased number of comorbidities in our study patients with IO, and so this condition may be a predisposing factor for increased morbidity and mortality in SCD patients. In our study, five (2.0%) patients with SCD died, and three of those had IO.
As demonstrated previously 20, our findings showed a stroke or at-risk-of-stroke prevalence of 24% (59/245) in SCD patients. Of the patients with stroke or at risk of stroke, 78% of them had IO (70% stroke, 30% abnormal TCD), making stroke the most common comorbidity that we observed with IO. This high rate of IO could be explained by the use of frequent blood transfusion as primary or secondary prevention in SCD individuals with abnormal TCD or cerebrovascular event14. In our study population, none of these patients had recurrent strokes after the initial event.
Sixty-four percent of our study population had ACS, and 41.6% of patients with ACS also had IO. In this population, patients with IO were at a significantly higher risk of ACS. This relationship, at least in part, may be related to the use of simple blood transfusions to improve oxygen-carrying capacity in persons with symptomatic ACS, and to the urgent exchange transfusion performed when there is rapid progression of ACS 21,22.
The prevalence of DVT was 30.2% (74/245) in individuals with SCD, with a predominance in the female sex (66.2%), as has also been described in one other study23. A significant relationship between IO and DVT could be explained by the occurrence of a hypercoagulable state in individuals with SCD24, driven by inflammation, in part due to oxidative stress and iron deposition-induced ferroptosis, leading to thrombogenesis and activation of the coagulation cascade in SCD individuals 25.
Epidemiological studies reveal that PH development is associated with iron overload and other comorbidities 26. Chronic anemia in sickle cell disease results in cardiac chamber dilation and a compensatory increase in left ventricular mass. Elevated TRV as well as ferritin and the number of red blood cell units transfused, were also found to be associated with a higher risk of death 27. In the current study, we found that the prevalence of IO associated with PH, as diagnosed by echocardiography, was 7.8%. Additionally, patients with IO were statistically more likely to have PH, with 68% of patients with PH also having IO.
In this study, there was no significant association between IO and other comorbidities such as anxiety and depression. However, we found a 40% prevalence of these mental health issues in the entire cohort of SCD patients. Depression is prevalent in adult patients with SCD and is associated with worse health-related quality of life, so an assessment of anxiety and depression in persons living with SCD is vitally important, given the prevalence of this comorbidity in the general SCD population 12,28.
The relationship between IO and acute kidney injury (AKI) and chronic kidney disease (CKD), although it was not significant in our study, may be a subject of further studies due to the risk of toxicity by iron chelation treatment with an elevated risk of mortality in SCD individuals 29. Recently, ferroptosis has been implicated in the development of AKI/CKD, as renal cells are particularly vulnerable to IO 30.
Usually, follow-up of patients with SCD and IO is based on periodic MRI measurements of LIC and serum ferritin levels. The current study has demonstrated a strong correlation between LIC and ferritin levels above 2500 ng/mL, with a sensitivity of 78% and a specificity of 74%. This correlation is compatible with other studies 6. Additionally, this threshold could indicate the need for chelation treatment if LIC cannot be performed 31. However, when the serum ferritin level is less than 2500 ng/mL, it is less reliable as a marker of LIC. This may be related to serum ferritin levels being affected by chronic inflammation in SCD. Especially during an acute exacerbation of a comorbidity or a vasoocclusive event, serum ferritin may be less reliable for estimating LIC 32. Additionally, it is important to note that only one-third of the subjects in our study with SCD and elevated serum ferritin levels underwent MRI for LIC measurement, reflecting a lack of appropriate follow-up for screening, diagnosis, and treatment of IO. This observation is in agreement with other studies 31.
Liver enzymes, as a measure of liver function, correlated well with serum ferritin and LIC levels in the SCD cohort with IO. Alkaline phosphatase has been associated with mortality 19, and some studies have reported a correlation between serum ferritin and AST as an estimator of LIC. A serum ferritin/AST ratio >17 µg/U has been highly predictive of IO 33. Nevertheless, in the population, routine screening of liver enzymes has poor sensitivity and specificity for detecting liver damage 34.
As shown in our study, IO is associated with a greater number of comorbidities. Thus, treating IO intuitively makes sense to reduce the risks associated with this condition. We found that DFX was the most commonly prescribed treatment for IO. Although 86% of IO patients in our study population were prescribed DFX, only 19% achieved ferritin levels <1000 ng/mL during the study period. According to other references 8,15,35, one factor contributing to a limited response to this treatment is poor adherence. While we did not assess compliance with chelation treatment as a cause of poor response in our study, treatment compliance is an important factor and should be evaluated when treating a patient for IO.
This study has several limitations: its observational, retrospective, and cross-sectional design, along with the reliance on medical record abstraction and the patient’s self-report of symptoms, could introduce bias and affect the statistical power of the results. In some cases, the evolution of IO is represented solely by serial determinations of serum ferritin, without a matched LIC, and by periodic MRIs. Additionally, it was not distinguished whether the serum ferritin values were reported during vasoocclusive crisis or in the steady state of the patients, thus there is a risk that the vasooclusive event could confound elevated serum ferritin levels. It is well known that ferritin levels increase significantly during a vaso-occlusive crisis 9. Information on toxicities related to DFX treatment was scarce and difficult to extract from the data records.
In conclusion, iron overload in SCD is a life-threatening condition, often associated with an increased number of comorbidities leading to a declining clinical course with an inadequate quality of life, hepatic complications and premature death. Management suggests that MRI screening for liver iron concentration should be performed every one to two years in patients with SCD receiving chronic transfusion therapy. Serum ferritin levels should be measured after each transfusion; this inexpensive blood test broadly correlates with total body iron burden and, due to its ease of acquisition, is a valuable tool for monitoring trends in iron burden over time 31. A significant limitation for using ferritin levels as indicator of IO in SCD is that inflammation can raise ferritin levels irrespective of iron burden 2. Providers should monitor patients and assess adherence with chelation therapy in the treatment of IO. Iron overload in SCD requires preventive care and close monitoring to avoid irreversible organ damage.
