- Email: [email protected]
Ferroportin Q248H mutation, hyperferritinemia and atypical type 2 diabetes mellitus in South Kivu
Diabetes & Metabolic Syndrome: Clinical Research & Reviews 7 (2013) 112–115
Contents lists available at SciVerse ScienceDirect
Diabetes & Metabolic Syndrome: Clinical Research & Reviews journal homepage: www.elsevier.com/locate/dsx
Original article
Ferroportin Q248H mutation, hyperferritinemia and atypical type 2 diabetes mellitus in South Kivu Philippe Bianga Katchunga a,*, Marius Baguma a, Jean-Rene´ M‘Buyamba-Kabangu a, Jan Philippe´ c, Michel P. Hermans b, Joris Delanghe a,c a b c
Faculty of Medicine NCD’s Observatory, Catholic University of Bukavu/VLIR-UOS, PO Box 235, Bukavu, The Democratic Republic of the Congo Department of Endocrinology, Saint Luc Academic Hospital, PO Box 1200, Brussels, Belgium Ghent University, Belgium
A R T I C L E I N F O
A B S T R A C T
Keywords: Ferroportin Hyperferritinemia Diabetes Mellitus South Kivu
Background: The ferroportin Q248H mutation is relatively common in sub-Saharan Africa. No previous study examined its relationship with atypical diabetes mellitus (DM) in this area. Objective: To determine the potential interactions between ferroportin Q248H mutation, hyperferritinemia and DM in South Kivu (RDC). Methodology: Presence of ferroportin Q248H mutation and iron status were investigated in diabetic patients (n = 179, age (mean) 57.7 years, CRP (median) 0.16 mg/L) and non-diabetic subjects (n = 86, age 44.5 years, CRP 0.07 mg/L) living in the city of Bukavu. Hyperferritinemia was considered for values greater than 200 and 300 mg/L in women and in men, respectively. Results: The prevalence of ferroportin Q248H mutation [12.1%] was non-significantly higher in diabetics than non-diabetics [14.0% vs. 8.1%, p = 0.17]. Similarly, hyperferritinemia frequency was higher in diabetic patients with Q248H mutation [44.0% vs. 14.3%, p = 0.16] and in mutation carriers [37.0% vs 16.5%, p = 0.001] than in the control groups, respectively. The association between Q248H mutation and DM was nevertheless not significant [adjusted OR 1.70 (95% CI: 0.52–5.58), p = 0.37], whereas hyperferritinemia [OR 2.72 (1.24–5.98), p = 0.01] showed an independent effect after adjustment for age and metabolic syndrome. Conclusions: The present work suggests a potential association between abnormal iron metabolism, ferroportin Q248H mutation and atypical DM in Africans, which may be modulated by environmental factors. ß 2013 Diabetes India. Published by Elsevier Ltd. All rights reserved.
1. Introduction Diabetes mellitus (DM) is atypical in a substantial subset of patients (>10%) in sub-Saharan Africa characterized by a lesser degree of insulin resistance and even preserved insulin sensitivity [1]. This frequency may be even higher in the DR of Congo, with less than 25% of diabetics obese [2]. The pathogenesis of atypical DM in Africa remains poorly documented despite its description dating back more than 20 years [1,3–5]. Sobngwi et al. recently noted that the phenotype of such DM was associated with type 8 human herpes virus infection (HHV-8) [6]. They hypothesized that HHV-8 infection may contribute to transient insulin deficiency [6]. HHV-8 is very common and endemic in sub-Saharan Africa, infecting 30–60% of
* Corresponding author. E-mail address: [email protected] (P.B. Katchunga).
its population [7]. There seems to exist an environmental epidemiological association between HHV-8 presence and high iron content of volcanic soils, which dominate this part of central Africa, particularly in the Middle and Eastern parts [8,9]. In this area, HHV-8 is also causally involved with high occurrence rates of endemic Kaposi’s sarcomas [8,9]. Interestingly, the tissues involved are known to contain high iron concentrations [9]. Recent studies describe an association between serum ferritin and risk of DM irrespective of insulin resistance [10]. Abnormalities in iron metabolism may hypothetically contribute partly to the pathogenesis of African DM, with a potentiating or triggering effect of HHV8 infection. In addition, ferroportin Q248H mutation is relatively common in Africa, with reported prevalence ranging from 2 to 10% across studies [11–16]. This mutation is almost nonexistent in Caucasian populations [12]. At present, there are no published studies having examined a candidate relationship between this mutation and DM phenotype in Central Africa. The objective of the present study was therefore
1871-4021/$ – see front matter ß 2013 Diabetes India. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dsx.2013.02.017
P.B. Katchunga et al. / Diabetes & Metabolic Syndrome: Clinical Research & Reviews 7 (2013) 112–115
to determine potential interactions between ferroportin Q248H mutation and hyperferritinemia in diabetic patients living in an iron-rich, volcanic area in South Kivu.
2. Methodology 2.1. Study population and methodology Diabetic patients and non-diabetic controls without signs of chronic systemic inflammation (CRP < 10 mg/L) from South Kivu and living in the city of Bukavu were recruited after obtaining informed consent. For each participant, demographic parameters (age, gender) and medical histories of T2DM and/or arterial hypertension were investigated. Body weight was measured to the nearest 100 g, with the subject dressed in light-weight clothing only. A Teraillon electronic balance scale was used for weighings, a folding meter ruler for measuring height, and a measuring tape for measuring abdominal circumference (AC) to the nearest 0.5 cm. Blood pressure (BP) measurement was performed using an OMRON M6 Comfort digital blood pressure monitor (OMRON Corp., Kyoto, Japan), applied to the right wrist, with the subject relaxed for at least 5 min in a sitting position. Three consecutive measures were obtained, with the lowest value considered. Fasting glycaemia levels was determined from capillary blood, using a portable glucometer (ACCU-CHEK Aviva, Roche Diagnostics, Mannheim, Germany). Finally, 4 mL of blood were obtained in the fasting state from an antecubital venopuncture for genetic analyzes, and 4 mL of blood were centrifuged to obtain serum for iron status and lipids determinations. These samples were stored in a freezer at minus 20 8C before being transported to the laboratory of the University of Ghent in Belgium for analysis. For the detection of the ferroportin 744 G ! T mutation, DNA was isolated from whole blood using a Qiagen QIAamp Blood Kit. Exon 6 of ferroportin was amplified by using a set of primers encompassing portions of the introns that flank the exon (forward primer: 50 -CTG TGG CATCGC CTTTATTT-30 reverse primer: 50 GCTCACATCAAGGAAGAGGG-30 ). After initial denaturation at near 94 8C for 10 min, a PCR was performed for 5 cycles of heating at 94 8C for 45 s, cooling at 56 8C for 45 s, and heating at 68 8C for 45 s, which were followed by 25 cycles of heating at 94 8C for 45 s, cooling at 52 8C for 45 s, and heating at 68 8C for 45 s and a final cycle of 15 min at 68 8C in a thermocycler (PTC-100, MJ Research Inc., Waltham, MA, USA). The 392-base pair (bp) product was digested with PvuII enzyme (MBI Fermentas, Hannover, 140 bp) were fractioned in a 3% agarose gel with ethidium bromide and detected [8]. Serum ferritin was assayed using electrochemiluminescence on a Modular instrument (Roche Diagnostics Gmbh, Mannheim, Germany). Concentrations of triglycerides, serum high-density lipoprotein (HDL), total cholesterol and serum iron were determined using colorimetric methods on the Modular P instrument (Roche), and serum transferrin (TF) concentration was determined using an immunoturbidimetric method on the same analyzer. LDL-cholesterol was calculated according to Friedewald’s formula [17]. 2.2. Study outcomes T2DM was diagnosed in subjects whose fasting glucose was 126 mg/dL [18] and/or previously diagnosed with T2DM. Patients with T1DM or secondary forms of diabetes were excluded. A subject was considered non-diabetic in the absence of a history of
113
DM or treatment with glucose-lowering medications, if fasting glucose was <126 mg/dL [18]. Hyperferritinemia was considered when the value of ferritin was 200 mg/L in women and 300 mg/L in men, respectively [19]. The transferrin saturation was considered high if 45%. Body mass index (BMI) was calculated as weight (kg) divided by the square of height (m). BMI scores greater or equal to 30.0 kg/m2 were defined as obesity. An abdominal circumference (AC) greater or equal to 80 cm (women), and greater or equal to 94 cm (men) defined the cut-off for abnormal waist circumference, as surrogate measure for central obesity [20]. Hypertension (HT) was defined as a personal history of the disorder and/or a BP measurement greater or equal to 140/ 90 mmHg [21]. In the present study, the presence of the metabolic syndrome (MetS) was defined according to the 2009 harmonised criteria for sub-Saharan Africans, based on the presence of 3 out of 5 of the following metabolic abnormalities: fasting glycemia 100 mg/dL or proven DM, blood pressure 130/85 mmHg and/or treated hypertension, HDL-cholesterol <50 mg/dL and <40 mg/dL, respectively, in women and men, fasting triglycerides >150 mg/dL in both sexes, and AC 80 cm and 94 cm in women and men, respectively [22]. 2.3. Statistics The Epi Info 2000 (version 3.3.2) software (Centre for Disease Control and Prevention, 204 Atlanta, GA, USA) was used for statistical analyzes. Data are presented, as appropriate, by median (interquartile range) and frequency. The Student’s t-test and the chi square were used to assess statistical significance of observed differences. The relative contribution of various risk factors of DM was assessed by multiple logistic regression method. We also investigated whether the Hardy–Weinberg equilibrium applied to the study population. A p value <0.05 defined the threshold for statistical significance. 3. Results 3.1. General characteristics of the study population In total, two hundred sixty-five subjects were selected for the study. Among them 179 (67.5%) were diabetic, 86 (32.5%) were non-diabetic, 138 (52.1%) were women and 127 (47.9%) were men. For the entire group, mean age was, respectively, 54.2 17.1 years (diabetics vs non-diabetics: 57.7 15.5 years vs 44.5 18.3 years, p < 0.0001), and BMI 23.9 4.5 kg/m2 (23.8 4.4 kg/m2 vs 24.4 4.6 kg/m2, p = 0.40). Median CRP was 0.15 (0.04–0.33) mg/L (0.16 mg/L vs 0.07 mg/L, p = 0.23), confirming the absence, at a cohort level, of subclinical chronic inflammation. 3.2. Frequency of ferroportin Q248H mutation Table 1 shows the frequency of the ferroportin Q248H mutation. In the entire study population (n = 265), the Q248H mutation was present in 32 (12.1%) subjects. The frequency of the Q248H mutation mutation was similar in both genders, without differences between older or younger patients or subjects [p > 0.05]. A Q248H mutation was relatively, albeit not significantly, more frequent in diabetic patients than in non-diabetics [14.0% vs. 8.1%, p = 0.17]. The homozygous form [n = 4 (1.5%)] was only found in diabetics [p = 0.16]. Similarly, the Q248H mutation was exclusively observed in non-obese subjects [p = 0.051]. The control group of the present study was to Hardy– Weinberg equilibrium (df = 1, x2 = 1.82 to 3.14).
P.B. Katchunga et al. / Diabetes & Metabolic Syndrome: Clinical Research & Reviews 7 (2013) 112–115
114
Table 1 Frequency of the ferroportin Q248H mutation. Q248H mutation n = 32 (12.1%) Diabetics Non-diabetics p
25 (14.0%) 7 (8.1%) 0.17
Male Female p
14 (11.0%) 18 (13.0%) 0.61
Age 60 years Age < 60 years p
14 (13.3) 18 (11.6) 0.68
BMI 30 kg/m2 BMI < 30 kg/m2 p
0 (0.0) 32 (13.4) 0.051
BMI = body mass index.
3.3. Iron metabolism, DM and ferroportin Q248H mutation Table 2 shows the biological parameters of diabetic vs. control groups according to the presence or absence of Q248H mutation. In the group without Q248H mutation (n = 233), compared to non-diabetics, diabetics had significantly higher serum iron [96.9 36.6 mg/L vs. 82.2 36.0 mg/L, p = 0.004], transferrin [2.6 0.4 g/L vs. 2.5 0.5 g/L, p = 0.03], transferrin saturation [36.7 13.1% vs. 28.2 11.7%, p < 0.0001], and a higher frequency of hyperferritinemia [37.0% vs. 16.5%, p = 0.001], respectively. On the other hand, in the group with Q248H mutation (n = 32), mean serum iron, transferrin and transferrin saturation were similar in diabetics and non-diabetics [p > 0.05]. However, 44.0% of diabetics (vs. 14.3% of non-diabetics) exhibited hyperferritinemia [p = 0.16]. In multiple logistic regression (Table 3), ferritin [adjusted OR = 2.72 (95% CI: 1.24–5.98), p = 0.01] was independently associated with DM after adjustment for age and metabolic syndrome. The association between Q248H mutation and DM was not significant [1.70 (0.52–5.58), p = 0.37]. 4. Discussion This study found a high frequency of Q248H mutation in this South Kivu population, with Q248H mutation relatively, albeit non-significantly more frequent in diabetics than non-diabetics. Similarly, hyperferritinemia frequency was higher in diabetic patients with mutation Q248H and non-mutation carriers than in the control groups, respectively. Hyperferritinemia showed an independent effect on the probability of DM risk after adjustment for age and metabolic syndrome. The frequent presence of a ferroportin Q248H mutation among natives from sub-Saharan Africa was previously reported by several authors, with reported rates identified in the literature for sub-Saharan Africa ranging from 2.2 to 10% [11–16].
However, the prevalence found in the diabetic group studied in this work was even higher. Thus, a comparable study in HIV-positive patients in Rwanda showed a 6% prevalence of Q248H mutation [16]. The exclusive presence of the homozygous form of the mutation in diabetics, and the absence of obesity in carriers of the Q248H mutation versus controls suggest a possible association between this abnormality and the pathogenesis of atypical African DM. However, we could not demonstrate statistically significant differences, partly as a result of the limited size of studied population. It is worth mentioning the very high frequency of hyperferritinemia in diabetic subjects, either carriers or non-carriers of the Q248H mutation. It is possible that the Q248H mutation is an emerging determinant of abnormal glucose homeostasis, which may be modulated by environmental iron exposure. In a metaanalysis, Mayr et al. noted a high frequency of hyperferritinemia in subjects with ferroportin disease who also had co-morbidities such as DM compared with a group without co-morbidity [23]. This observation matches our results. Nevertheless, this high presence of hyperferritinemia in diabetics non-carriers of the Q248H mutation calls for other risk factors. Such very high frequency of hyperferritinemia in the diabetic group is at odd with the usual association between hepatosiderosis, non-alcoholic fatty liver, central obesity, insulin resistance and/or compensatory hyperinsulinemia [24–28]. On the other end, ferritin showed an independent association with the presence of DM. This observation was also noted by Sun et al. who found an association between serum ferritin and risk of T2DM, irrespective of the metabolic syndrome [10]. However, their pioneering observations may have gone largely unnoticed. One cannot rule out that high ferritinemia in diabetic patients from Kivu may be linked to sub-clinical inflammation associated with chronic communicable diseases, since infectious diseases are very common in this region. It is also possible that the iron-rich environment of the South Kivu biosphere may modulate some risk factors for (a)typical DM. Thus, acquired abnormalities in iron metabolism apart from insulin resistance, probably from genetic and environmental causes, and associated with hyperferritinemia in diabetics from sub-Saharan Africa may contribute to modulate insulin sensitivity in the liver and/or impair insulin secretion. Such hyperferritinemia, unrelated to insulin resistance, might as such contribute to the atypical phenotype of T2DM in sub-Saharan Africans. Finally, our study should be interpreted within the limits of its methodology. The relatively small size of the sample surely impacted the statistical robustness of the associations observed. Similarly, our study was cross-sectional and could not elicit causal links between hyperferritinemia and DM. In addition, other causes of hyperferritinemia have not been ruled out, such as excess ethanol intake, abnormal liver function, haemolysis, or HFE mutations. However, a very high frequency of hyperferritinemia associated with normal transferrin saturation (>80%) minimized the involvement of such confounders. Finally, cardiometabolic parameters and quantitative measurement (such as HOMA tests)
Table 2 Biological parameters of diabetic vs. control group. Q248H mutation n = 32
Iron (mg/L) Transferrin (g/L) TF saturation (%) Ferritin (mg/L) Hyperferritinemia (%) TF = transferrin.
Q248H mutation absent n = 233
DM (n = 25)
Controls (n = 7)
p
DM (n = 154)
Controls (n = 79)
p
97.5 32.3 2.5 0.4 38.9 13.5 214.4 (121.2–309.1) 44.0
78.7 46.3 2.5 0.3 28.9 14.8 105.3 (92.3–186.7) 14.3
0.24 0.87 0.10 0.058 0.16
96.9 36.6 2.6 0.4 36.7 13.1 183.6 (97.1–292.6) 37.0
82.2 36.0 2.4 0.5 28.2 11.7 128.3 (65.1–228.7) 16.5
0.004 0.03 <0.0001 0.004 0.001
P.B. Katchunga et al. / Diabetes & Metabolic Syndrome: Clinical Research & Reviews 7 (2013) 112–115
115
Table 3 Multivariate-adjusted odds ratios (OR) for diabetes mellitus (DM).
Hyperferritinemia vs. normal ferritinemia Age > 60 years vs. <60 years MetS 3/5 vs. MetS < 3/5 Q248H mutation vs. absent
DM frequency (%)
Unadjusted OR (95% CI)
p
Adjusted OR (95% CI)
p
82.9 84.8 91.7 78.1
3.15 4.58 5.84 1.83
0.0005 0.0001 0.0001 0.17
2.72 2.63 4.68 1.70
0.01 0.008 0.0005 0.37
vs. vs. vs. vs.
60.7 54.8 65.3 66.1
(1.64–6.01) (2.46–8.50) (2.51–13.59) (0.75–4.42)
(1.24–5.98) (1.27–5.43) (1.95–11.21) (0.52–5.58)
MetS = metabolic syndrome.
of insulin resistance were not available in the present study, although previous studies carried out in diabetics in this region have showed a very low prevalence of insulin resistance [29]. In conclusion, the present work suggests a potential association between abnormal iron metabolism and the atypical phenotype of T2DM in Africans, possibly through genetic and environmental factors. Further studies are needed to refute or confirm this hypothesis. Conflict of interest The authors declare that they have no conflicts of interest concerning this article. Acknowledgements This study was supported by the Flemish Inter-University Council (VLIR-UOS grant ZIUS2012AP024). References [1] Sobngwi E, Mauvais-Jarvis F, Vexiau P, Mbanya JC, Gautier JF. Diabetes in Africans. Part 1. Epidemiological and clinical specificities. Diabetes and Metabolism 2001;27:628–34. [2] Katchunga P, Masumbuko B, Belma M, Kashongwe Munogolo Z, Hermans MP, M’Buyamba-Kabangu JR. Age and living in an urban environment as major determinants of prevalent diabetes mellitus among South-Kivu Congolese adults. Diabetes and Metabolism 2012;38:324–31. [3] Mauvais-Jarvis F, Sobngwi E, Porcher R, Riveline JP, Kevorkian JP, Vaisse C, et al. Ketosis-prone type 2 diabetes in patients of sub-Saharan African origin: clinical pathophysiology and natural history of beta-cell dysfunction and insulin resistance. Diabetes 2004;53:645–53. [4] Sobngwi E, Mauvais-Jarvis FF, Vexiau P, Mbanya JC, Gautier JF. Diabetes in Africans. Part 2. Ketosis prone atypical diabetes mellitus. Diabetes and Metabolism 2002;28:5–12. [5] Umpierrez GE, Smiley D, Kitabchi AE. Narrative review: ketosis-prone type 2 diabetes mellitus. Annals of Internal Medicine 2006;144:350–7. [6] Sobngwi E, Choukem SP, Agbalika F, Blondeau B, Fetita LS, Lebbe C, et al. Ketosis prone type 2 diabetes and human herpes virus 8 infection in sub-Saharan Africans. Journal of the American Medical Association 2008;299:2770–6. [7] Schulz TF. Kaposi’s sarcoma-associated herpesvirus (human herpesvirus 8): epidemiology and pathogenesis. Journal of Antimicrobial Chemotherapy 2000;45(Suppl. 3):15–27. [8] Simonart T. Iron: a target for the management of Kaposi’s sarcoma? BMC Cancer 2004;4:1–8. [9] Ziegler JL. Endemic Kaposi’s sarcoma in Africa and local volcanic soils. Lancet 1993;342:1348–51. [10] Sun L, Franco OH, Hu FB, Cai L, Yu Z, Li H. Ferritin concentrations, metabolic syndrome, and type 2 diabetes in middle-aged and elderly Chinese. Journal of Clinical Endocrinology and Metabolism 2008;93(December (12)):4690–6. [11] Albuquerque D, Manco L, Loua KM, Arez AP, Trovoada MJ, Relvas L, et al. SLC40A1 Q248H allele frequencies and associated SLC40A1 haplotypes in three West African population samples. Annals of Human Biology 2011; 38(3):378–81.
[12] Barton JC, Acton RT, Lee PL, West C. SLC40A1 Q248H allele frequencies and Q248H-associated risk of non-HFE iron overload in persons of sub-Saharan African descent. Blood Cells Molecules and Diseases 2007;39:206–11. [13] Barton JC, Acton RT, Rivers CA, Bertoli LF, Gelbart T, West C, Beutler E. Genotypic and phenotypic heterogeneity of African Americans with primary iron overload. Blood Cells Molecules and Diseases 2003;31:310–9. [14] Beutler E, Barton JC, Felitti VJ, Gelbart T, West C, Lee PL, Waalen J, Vulpe C. Ferroportin 1 (SCL40A1) variant associated with iron overload in AfricanAmericans. Blood Cells Molecules and Diseases 2003;31:305–9. [15] Gordeuk VR, Caleffi A, Corradini E, Ferrara F, Jones RA, Castro O, et al. Iron overload in Africans and African-Americans and a common mutation in the SCL40A1 (ferroportin 1) gene. Blood Cells Molecules and Diseases 2003; 31:299–304. [16] Masaisa F, Breman C, Gahutu JB, Mukiibi J, Delanghe J, Philippe´ J. Ferroportin (SLC40A1) Q248H mutation is associated with lower circulating serum hepcidin levels in Rwandese HIV-positive women. Annals of Hematology 2012;91:911–6. [17] Friedewald WT, Levy RI, Frederickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clinical Chemistry 1972;18:499–502. [18] American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2012;35(Suppl.):S64–71. [19] World Health Organization/United Nations Children’s Fund/United Nations University 2001 Iron deficiency anaemia. Assessment, prevention and control: a guide for programme managers, WHO/NHD/01.3. Geneva: World Health Organization. [20] Alberti KG, Zimmet P, Shaw J. IDF Epidemiology Task Force Consensus Group. The metabolic syndrome: a new worldwide definition. Lancet 2005;366: 1059–62. [21] Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL, et al. The seventh report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood Pressure: the JNC 7 report. Journal of the American Medical Association 2003;289:2560–72. [22] Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al. Harmonizing the Metabolic Syndrome. A joint interim statement of the I.D.F. Task Force on Epidemiology and Prevention; N.H.L. and Blood Institute; A.H.A; W.H.F; I.A.S; and I.A. for the study of obesity. Circulation 2009;120:1640–5. [23] Mayr R, Janecke AR, Schranz M, Griffiths WJH, Vogel W, Pietrangelo A, et al. Ferroportin disease: a systematic meta-analysis of clinical and molecular findings. Journal of Hepatology 2010;53:941–9. [24] Jouanolle AM, Douabin-Gicquel V, Halimi C, Lore´al O, Fergelot P, Delacour T, et al. Novel mutation in ferroportin 1 gene is associated with autosomal dominant iron overload. Journal of Hepatology 2003;39:286–9. [25] Beaton M, Guyader D, Deugnier Y, Moirand R, Chakrabarti S, Adams P. Noninvasive prediction of cirrhosis in C282Y linked hemochromatosis. Hepatology 2002;36:673–8. [26] Morrison ED, Brandhagen DJ, Phatak PD, Barton JC, Krawitt EL, El-Serag HB, et al. Serum ferritin level predicts advanced hepatic fibrosis among U.S. patients with phenotypic hemochromatosis. Annals of Internal Medicine 2003;138:627–33. [27] Adams PC, Reboussin DM, Barton JC, Christine E, McLaren CE, John H, et al. Hemochromatosis and iron-overload screening in a racially diverse population. New England Journal of Medicine 2005;352:1769–78. [28] Damade R, Rosenthal E, Cacoub P. Les hyperferritine´mies. Ann Med Interne 2000;151:169–77. [29] Katchunga P, Hermans MP, Manwa B, Lepira F, Kashongwe Z, M’BuyambaKabangu JR. Hypertension, insulin resistance and chronic kidney disease in type 2 diabetes patients from South Kivu, DR Congo. Nephrologie & Therapeutique 2010;6:520–5.
Contents lists available at SciVerse ScienceDirect
Diabetes & Metabolic Syndrome: Clinical Research & Reviews journal homepage: www.elsevier.com/locate/dsx
Original article
Ferroportin Q248H mutation, hyperferritinemia and atypical type 2 diabetes mellitus in South Kivu Philippe Bianga Katchunga a,*, Marius Baguma a, Jean-Rene´ M‘Buyamba-Kabangu a, Jan Philippe´ c, Michel P. Hermans b, Joris Delanghe a,c a b c
Faculty of Medicine NCD’s Observatory, Catholic University of Bukavu/VLIR-UOS, PO Box 235, Bukavu, The Democratic Republic of the Congo Department of Endocrinology, Saint Luc Academic Hospital, PO Box 1200, Brussels, Belgium Ghent University, Belgium
A R T I C L E I N F O
A B S T R A C T
Keywords: Ferroportin Hyperferritinemia Diabetes Mellitus South Kivu
Background: The ferroportin Q248H mutation is relatively common in sub-Saharan Africa. No previous study examined its relationship with atypical diabetes mellitus (DM) in this area. Objective: To determine the potential interactions between ferroportin Q248H mutation, hyperferritinemia and DM in South Kivu (RDC). Methodology: Presence of ferroportin Q248H mutation and iron status were investigated in diabetic patients (n = 179, age (mean) 57.7 years, CRP (median) 0.16 mg/L) and non-diabetic subjects (n = 86, age 44.5 years, CRP 0.07 mg/L) living in the city of Bukavu. Hyperferritinemia was considered for values greater than 200 and 300 mg/L in women and in men, respectively. Results: The prevalence of ferroportin Q248H mutation [12.1%] was non-significantly higher in diabetics than non-diabetics [14.0% vs. 8.1%, p = 0.17]. Similarly, hyperferritinemia frequency was higher in diabetic patients with Q248H mutation [44.0% vs. 14.3%, p = 0.16] and in mutation carriers [37.0% vs 16.5%, p = 0.001] than in the control groups, respectively. The association between Q248H mutation and DM was nevertheless not significant [adjusted OR 1.70 (95% CI: 0.52–5.58), p = 0.37], whereas hyperferritinemia [OR 2.72 (1.24–5.98), p = 0.01] showed an independent effect after adjustment for age and metabolic syndrome. Conclusions: The present work suggests a potential association between abnormal iron metabolism, ferroportin Q248H mutation and atypical DM in Africans, which may be modulated by environmental factors. ß 2013 Diabetes India. Published by Elsevier Ltd. All rights reserved.
1. Introduction Diabetes mellitus (DM) is atypical in a substantial subset of patients (>10%) in sub-Saharan Africa characterized by a lesser degree of insulin resistance and even preserved insulin sensitivity [1]. This frequency may be even higher in the DR of Congo, with less than 25% of diabetics obese [2]. The pathogenesis of atypical DM in Africa remains poorly documented despite its description dating back more than 20 years [1,3–5]. Sobngwi et al. recently noted that the phenotype of such DM was associated with type 8 human herpes virus infection (HHV-8) [6]. They hypothesized that HHV-8 infection may contribute to transient insulin deficiency [6]. HHV-8 is very common and endemic in sub-Saharan Africa, infecting 30–60% of
* Corresponding author. E-mail address: [email protected] (P.B. Katchunga).
its population [7]. There seems to exist an environmental epidemiological association between HHV-8 presence and high iron content of volcanic soils, which dominate this part of central Africa, particularly in the Middle and Eastern parts [8,9]. In this area, HHV-8 is also causally involved with high occurrence rates of endemic Kaposi’s sarcomas [8,9]. Interestingly, the tissues involved are known to contain high iron concentrations [9]. Recent studies describe an association between serum ferritin and risk of DM irrespective of insulin resistance [10]. Abnormalities in iron metabolism may hypothetically contribute partly to the pathogenesis of African DM, with a potentiating or triggering effect of HHV8 infection. In addition, ferroportin Q248H mutation is relatively common in Africa, with reported prevalence ranging from 2 to 10% across studies [11–16]. This mutation is almost nonexistent in Caucasian populations [12]. At present, there are no published studies having examined a candidate relationship between this mutation and DM phenotype in Central Africa. The objective of the present study was therefore
1871-4021/$ – see front matter ß 2013 Diabetes India. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dsx.2013.02.017
P.B. Katchunga et al. / Diabetes & Metabolic Syndrome: Clinical Research & Reviews 7 (2013) 112–115
to determine potential interactions between ferroportin Q248H mutation and hyperferritinemia in diabetic patients living in an iron-rich, volcanic area in South Kivu.
2. Methodology 2.1. Study population and methodology Diabetic patients and non-diabetic controls without signs of chronic systemic inflammation (CRP < 10 mg/L) from South Kivu and living in the city of Bukavu were recruited after obtaining informed consent. For each participant, demographic parameters (age, gender) and medical histories of T2DM and/or arterial hypertension were investigated. Body weight was measured to the nearest 100 g, with the subject dressed in light-weight clothing only. A Teraillon electronic balance scale was used for weighings, a folding meter ruler for measuring height, and a measuring tape for measuring abdominal circumference (AC) to the nearest 0.5 cm. Blood pressure (BP) measurement was performed using an OMRON M6 Comfort digital blood pressure monitor (OMRON Corp., Kyoto, Japan), applied to the right wrist, with the subject relaxed for at least 5 min in a sitting position. Three consecutive measures were obtained, with the lowest value considered. Fasting glycaemia levels was determined from capillary blood, using a portable glucometer (ACCU-CHEK Aviva, Roche Diagnostics, Mannheim, Germany). Finally, 4 mL of blood were obtained in the fasting state from an antecubital venopuncture for genetic analyzes, and 4 mL of blood were centrifuged to obtain serum for iron status and lipids determinations. These samples were stored in a freezer at minus 20 8C before being transported to the laboratory of the University of Ghent in Belgium for analysis. For the detection of the ferroportin 744 G ! T mutation, DNA was isolated from whole blood using a Qiagen QIAamp Blood Kit. Exon 6 of ferroportin was amplified by using a set of primers encompassing portions of the introns that flank the exon (forward primer: 50 -CTG TGG CATCGC CTTTATTT-30 reverse primer: 50 GCTCACATCAAGGAAGAGGG-30 ). After initial denaturation at near 94 8C for 10 min, a PCR was performed for 5 cycles of heating at 94 8C for 45 s, cooling at 56 8C for 45 s, and heating at 68 8C for 45 s, which were followed by 25 cycles of heating at 94 8C for 45 s, cooling at 52 8C for 45 s, and heating at 68 8C for 45 s and a final cycle of 15 min at 68 8C in a thermocycler (PTC-100, MJ Research Inc., Waltham, MA, USA). The 392-base pair (bp) product was digested with PvuII enzyme (MBI Fermentas, Hannover, 140 bp) were fractioned in a 3% agarose gel with ethidium bromide and detected [8]. Serum ferritin was assayed using electrochemiluminescence on a Modular instrument (Roche Diagnostics Gmbh, Mannheim, Germany). Concentrations of triglycerides, serum high-density lipoprotein (HDL), total cholesterol and serum iron were determined using colorimetric methods on the Modular P instrument (Roche), and serum transferrin (TF) concentration was determined using an immunoturbidimetric method on the same analyzer. LDL-cholesterol was calculated according to Friedewald’s formula [17]. 2.2. Study outcomes T2DM was diagnosed in subjects whose fasting glucose was 126 mg/dL [18] and/or previously diagnosed with T2DM. Patients with T1DM or secondary forms of diabetes were excluded. A subject was considered non-diabetic in the absence of a history of
113
DM or treatment with glucose-lowering medications, if fasting glucose was <126 mg/dL [18]. Hyperferritinemia was considered when the value of ferritin was 200 mg/L in women and 300 mg/L in men, respectively [19]. The transferrin saturation was considered high if 45%. Body mass index (BMI) was calculated as weight (kg) divided by the square of height (m). BMI scores greater or equal to 30.0 kg/m2 were defined as obesity. An abdominal circumference (AC) greater or equal to 80 cm (women), and greater or equal to 94 cm (men) defined the cut-off for abnormal waist circumference, as surrogate measure for central obesity [20]. Hypertension (HT) was defined as a personal history of the disorder and/or a BP measurement greater or equal to 140/ 90 mmHg [21]. In the present study, the presence of the metabolic syndrome (MetS) was defined according to the 2009 harmonised criteria for sub-Saharan Africans, based on the presence of 3 out of 5 of the following metabolic abnormalities: fasting glycemia 100 mg/dL or proven DM, blood pressure 130/85 mmHg and/or treated hypertension, HDL-cholesterol <50 mg/dL and <40 mg/dL, respectively, in women and men, fasting triglycerides >150 mg/dL in both sexes, and AC 80 cm and 94 cm in women and men, respectively [22]. 2.3. Statistics The Epi Info 2000 (version 3.3.2) software (Centre for Disease Control and Prevention, 204 Atlanta, GA, USA) was used for statistical analyzes. Data are presented, as appropriate, by median (interquartile range) and frequency. The Student’s t-test and the chi square were used to assess statistical significance of observed differences. The relative contribution of various risk factors of DM was assessed by multiple logistic regression method. We also investigated whether the Hardy–Weinberg equilibrium applied to the study population. A p value <0.05 defined the threshold for statistical significance. 3. Results 3.1. General characteristics of the study population In total, two hundred sixty-five subjects were selected for the study. Among them 179 (67.5%) were diabetic, 86 (32.5%) were non-diabetic, 138 (52.1%) were women and 127 (47.9%) were men. For the entire group, mean age was, respectively, 54.2 17.1 years (diabetics vs non-diabetics: 57.7 15.5 years vs 44.5 18.3 years, p < 0.0001), and BMI 23.9 4.5 kg/m2 (23.8 4.4 kg/m2 vs 24.4 4.6 kg/m2, p = 0.40). Median CRP was 0.15 (0.04–0.33) mg/L (0.16 mg/L vs 0.07 mg/L, p = 0.23), confirming the absence, at a cohort level, of subclinical chronic inflammation. 3.2. Frequency of ferroportin Q248H mutation Table 1 shows the frequency of the ferroportin Q248H mutation. In the entire study population (n = 265), the Q248H mutation was present in 32 (12.1%) subjects. The frequency of the Q248H mutation mutation was similar in both genders, without differences between older or younger patients or subjects [p > 0.05]. A Q248H mutation was relatively, albeit not significantly, more frequent in diabetic patients than in non-diabetics [14.0% vs. 8.1%, p = 0.17]. The homozygous form [n = 4 (1.5%)] was only found in diabetics [p = 0.16]. Similarly, the Q248H mutation was exclusively observed in non-obese subjects [p = 0.051]. The control group of the present study was to Hardy– Weinberg equilibrium (df = 1, x2 = 1.82 to 3.14).
P.B. Katchunga et al. / Diabetes & Metabolic Syndrome: Clinical Research & Reviews 7 (2013) 112–115
114
Table 1 Frequency of the ferroportin Q248H mutation. Q248H mutation n = 32 (12.1%) Diabetics Non-diabetics p
25 (14.0%) 7 (8.1%) 0.17
Male Female p
14 (11.0%) 18 (13.0%) 0.61
Age 60 years Age < 60 years p
14 (13.3) 18 (11.6) 0.68
BMI 30 kg/m2 BMI < 30 kg/m2 p
0 (0.0) 32 (13.4) 0.051
BMI = body mass index.
3.3. Iron metabolism, DM and ferroportin Q248H mutation Table 2 shows the biological parameters of diabetic vs. control groups according to the presence or absence of Q248H mutation. In the group without Q248H mutation (n = 233), compared to non-diabetics, diabetics had significantly higher serum iron [96.9 36.6 mg/L vs. 82.2 36.0 mg/L, p = 0.004], transferrin [2.6 0.4 g/L vs. 2.5 0.5 g/L, p = 0.03], transferrin saturation [36.7 13.1% vs. 28.2 11.7%, p < 0.0001], and a higher frequency of hyperferritinemia [37.0% vs. 16.5%, p = 0.001], respectively. On the other hand, in the group with Q248H mutation (n = 32), mean serum iron, transferrin and transferrin saturation were similar in diabetics and non-diabetics [p > 0.05]. However, 44.0% of diabetics (vs. 14.3% of non-diabetics) exhibited hyperferritinemia [p = 0.16]. In multiple logistic regression (Table 3), ferritin [adjusted OR = 2.72 (95% CI: 1.24–5.98), p = 0.01] was independently associated with DM after adjustment for age and metabolic syndrome. The association between Q248H mutation and DM was not significant [1.70 (0.52–5.58), p = 0.37]. 4. Discussion This study found a high frequency of Q248H mutation in this South Kivu population, with Q248H mutation relatively, albeit non-significantly more frequent in diabetics than non-diabetics. Similarly, hyperferritinemia frequency was higher in diabetic patients with mutation Q248H and non-mutation carriers than in the control groups, respectively. Hyperferritinemia showed an independent effect on the probability of DM risk after adjustment for age and metabolic syndrome. The frequent presence of a ferroportin Q248H mutation among natives from sub-Saharan Africa was previously reported by several authors, with reported rates identified in the literature for sub-Saharan Africa ranging from 2.2 to 10% [11–16].
However, the prevalence found in the diabetic group studied in this work was even higher. Thus, a comparable study in HIV-positive patients in Rwanda showed a 6% prevalence of Q248H mutation [16]. The exclusive presence of the homozygous form of the mutation in diabetics, and the absence of obesity in carriers of the Q248H mutation versus controls suggest a possible association between this abnormality and the pathogenesis of atypical African DM. However, we could not demonstrate statistically significant differences, partly as a result of the limited size of studied population. It is worth mentioning the very high frequency of hyperferritinemia in diabetic subjects, either carriers or non-carriers of the Q248H mutation. It is possible that the Q248H mutation is an emerging determinant of abnormal glucose homeostasis, which may be modulated by environmental iron exposure. In a metaanalysis, Mayr et al. noted a high frequency of hyperferritinemia in subjects with ferroportin disease who also had co-morbidities such as DM compared with a group without co-morbidity [23]. This observation matches our results. Nevertheless, this high presence of hyperferritinemia in diabetics non-carriers of the Q248H mutation calls for other risk factors. Such very high frequency of hyperferritinemia in the diabetic group is at odd with the usual association between hepatosiderosis, non-alcoholic fatty liver, central obesity, insulin resistance and/or compensatory hyperinsulinemia [24–28]. On the other end, ferritin showed an independent association with the presence of DM. This observation was also noted by Sun et al. who found an association between serum ferritin and risk of T2DM, irrespective of the metabolic syndrome [10]. However, their pioneering observations may have gone largely unnoticed. One cannot rule out that high ferritinemia in diabetic patients from Kivu may be linked to sub-clinical inflammation associated with chronic communicable diseases, since infectious diseases are very common in this region. It is also possible that the iron-rich environment of the South Kivu biosphere may modulate some risk factors for (a)typical DM. Thus, acquired abnormalities in iron metabolism apart from insulin resistance, probably from genetic and environmental causes, and associated with hyperferritinemia in diabetics from sub-Saharan Africa may contribute to modulate insulin sensitivity in the liver and/or impair insulin secretion. Such hyperferritinemia, unrelated to insulin resistance, might as such contribute to the atypical phenotype of T2DM in sub-Saharan Africans. Finally, our study should be interpreted within the limits of its methodology. The relatively small size of the sample surely impacted the statistical robustness of the associations observed. Similarly, our study was cross-sectional and could not elicit causal links between hyperferritinemia and DM. In addition, other causes of hyperferritinemia have not been ruled out, such as excess ethanol intake, abnormal liver function, haemolysis, or HFE mutations. However, a very high frequency of hyperferritinemia associated with normal transferrin saturation (>80%) minimized the involvement of such confounders. Finally, cardiometabolic parameters and quantitative measurement (such as HOMA tests)
Table 2 Biological parameters of diabetic vs. control group. Q248H mutation n = 32
Iron (mg/L) Transferrin (g/L) TF saturation (%) Ferritin (mg/L) Hyperferritinemia (%) TF = transferrin.
Q248H mutation absent n = 233
DM (n = 25)
Controls (n = 7)
p
DM (n = 154)
Controls (n = 79)
p
97.5 32.3 2.5 0.4 38.9 13.5 214.4 (121.2–309.1) 44.0
78.7 46.3 2.5 0.3 28.9 14.8 105.3 (92.3–186.7) 14.3
0.24 0.87 0.10 0.058 0.16
96.9 36.6 2.6 0.4 36.7 13.1 183.6 (97.1–292.6) 37.0
82.2 36.0 2.4 0.5 28.2 11.7 128.3 (65.1–228.7) 16.5
0.004 0.03 <0.0001 0.004 0.001
P.B. Katchunga et al. / Diabetes & Metabolic Syndrome: Clinical Research & Reviews 7 (2013) 112–115
115
Table 3 Multivariate-adjusted odds ratios (OR) for diabetes mellitus (DM).
Hyperferritinemia vs. normal ferritinemia Age > 60 years vs. <60 years MetS 3/5 vs. MetS < 3/5 Q248H mutation vs. absent
DM frequency (%)
Unadjusted OR (95% CI)
p
Adjusted OR (95% CI)
p
82.9 84.8 91.7 78.1
3.15 4.58 5.84 1.83
0.0005 0.0001 0.0001 0.17
2.72 2.63 4.68 1.70
0.01 0.008 0.0005 0.37
vs. vs. vs. vs.
60.7 54.8 65.3 66.1
(1.64–6.01) (2.46–8.50) (2.51–13.59) (0.75–4.42)
(1.24–5.98) (1.27–5.43) (1.95–11.21) (0.52–5.58)
MetS = metabolic syndrome.
of insulin resistance were not available in the present study, although previous studies carried out in diabetics in this region have showed a very low prevalence of insulin resistance [29]. In conclusion, the present work suggests a potential association between abnormal iron metabolism and the atypical phenotype of T2DM in Africans, possibly through genetic and environmental factors. Further studies are needed to refute or confirm this hypothesis. Conflict of interest The authors declare that they have no conflicts of interest concerning this article. Acknowledgements This study was supported by the Flemish Inter-University Council (VLIR-UOS grant ZIUS2012AP024). References [1] Sobngwi E, Mauvais-Jarvis F, Vexiau P, Mbanya JC, Gautier JF. Diabetes in Africans. Part 1. Epidemiological and clinical specificities. Diabetes and Metabolism 2001;27:628–34. [2] Katchunga P, Masumbuko B, Belma M, Kashongwe Munogolo Z, Hermans MP, M’Buyamba-Kabangu JR. Age and living in an urban environment as major determinants of prevalent diabetes mellitus among South-Kivu Congolese adults. Diabetes and Metabolism 2012;38:324–31. [3] Mauvais-Jarvis F, Sobngwi E, Porcher R, Riveline JP, Kevorkian JP, Vaisse C, et al. Ketosis-prone type 2 diabetes in patients of sub-Saharan African origin: clinical pathophysiology and natural history of beta-cell dysfunction and insulin resistance. Diabetes 2004;53:645–53. [4] Sobngwi E, Mauvais-Jarvis FF, Vexiau P, Mbanya JC, Gautier JF. Diabetes in Africans. Part 2. Ketosis prone atypical diabetes mellitus. Diabetes and Metabolism 2002;28:5–12. [5] Umpierrez GE, Smiley D, Kitabchi AE. Narrative review: ketosis-prone type 2 diabetes mellitus. Annals of Internal Medicine 2006;144:350–7. [6] Sobngwi E, Choukem SP, Agbalika F, Blondeau B, Fetita LS, Lebbe C, et al. Ketosis prone type 2 diabetes and human herpes virus 8 infection in sub-Saharan Africans. Journal of the American Medical Association 2008;299:2770–6. [7] Schulz TF. Kaposi’s sarcoma-associated herpesvirus (human herpesvirus 8): epidemiology and pathogenesis. Journal of Antimicrobial Chemotherapy 2000;45(Suppl. 3):15–27. [8] Simonart T. Iron: a target for the management of Kaposi’s sarcoma? BMC Cancer 2004;4:1–8. [9] Ziegler JL. Endemic Kaposi’s sarcoma in Africa and local volcanic soils. Lancet 1993;342:1348–51. [10] Sun L, Franco OH, Hu FB, Cai L, Yu Z, Li H. Ferritin concentrations, metabolic syndrome, and type 2 diabetes in middle-aged and elderly Chinese. Journal of Clinical Endocrinology and Metabolism 2008;93(December (12)):4690–6. [11] Albuquerque D, Manco L, Loua KM, Arez AP, Trovoada MJ, Relvas L, et al. SLC40A1 Q248H allele frequencies and associated SLC40A1 haplotypes in three West African population samples. Annals of Human Biology 2011; 38(3):378–81.
[12] Barton JC, Acton RT, Lee PL, West C. SLC40A1 Q248H allele frequencies and Q248H-associated risk of non-HFE iron overload in persons of sub-Saharan African descent. Blood Cells Molecules and Diseases 2007;39:206–11. [13] Barton JC, Acton RT, Rivers CA, Bertoli LF, Gelbart T, West C, Beutler E. Genotypic and phenotypic heterogeneity of African Americans with primary iron overload. Blood Cells Molecules and Diseases 2003;31:310–9. [14] Beutler E, Barton JC, Felitti VJ, Gelbart T, West C, Lee PL, Waalen J, Vulpe C. Ferroportin 1 (SCL40A1) variant associated with iron overload in AfricanAmericans. Blood Cells Molecules and Diseases 2003;31:305–9. [15] Gordeuk VR, Caleffi A, Corradini E, Ferrara F, Jones RA, Castro O, et al. Iron overload in Africans and African-Americans and a common mutation in the SCL40A1 (ferroportin 1) gene. Blood Cells Molecules and Diseases 2003; 31:299–304. [16] Masaisa F, Breman C, Gahutu JB, Mukiibi J, Delanghe J, Philippe´ J. Ferroportin (SLC40A1) Q248H mutation is associated with lower circulating serum hepcidin levels in Rwandese HIV-positive women. Annals of Hematology 2012;91:911–6. [17] Friedewald WT, Levy RI, Frederickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clinical Chemistry 1972;18:499–502. [18] American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2012;35(Suppl.):S64–71. [19] World Health Organization/United Nations Children’s Fund/United Nations University 2001 Iron deficiency anaemia. Assessment, prevention and control: a guide for programme managers, WHO/NHD/01.3. Geneva: World Health Organization. [20] Alberti KG, Zimmet P, Shaw J. IDF Epidemiology Task Force Consensus Group. The metabolic syndrome: a new worldwide definition. Lancet 2005;366: 1059–62. [21] Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL, et al. The seventh report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood Pressure: the JNC 7 report. Journal of the American Medical Association 2003;289:2560–72. [22] Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al. Harmonizing the Metabolic Syndrome. A joint interim statement of the I.D.F. Task Force on Epidemiology and Prevention; N.H.L. and Blood Institute; A.H.A; W.H.F; I.A.S; and I.A. for the study of obesity. Circulation 2009;120:1640–5. [23] Mayr R, Janecke AR, Schranz M, Griffiths WJH, Vogel W, Pietrangelo A, et al. Ferroportin disease: a systematic meta-analysis of clinical and molecular findings. Journal of Hepatology 2010;53:941–9. [24] Jouanolle AM, Douabin-Gicquel V, Halimi C, Lore´al O, Fergelot P, Delacour T, et al. Novel mutation in ferroportin 1 gene is associated with autosomal dominant iron overload. Journal of Hepatology 2003;39:286–9. [25] Beaton M, Guyader D, Deugnier Y, Moirand R, Chakrabarti S, Adams P. Noninvasive prediction of cirrhosis in C282Y linked hemochromatosis. Hepatology 2002;36:673–8. [26] Morrison ED, Brandhagen DJ, Phatak PD, Barton JC, Krawitt EL, El-Serag HB, et al. Serum ferritin level predicts advanced hepatic fibrosis among U.S. patients with phenotypic hemochromatosis. Annals of Internal Medicine 2003;138:627–33. [27] Adams PC, Reboussin DM, Barton JC, Christine E, McLaren CE, John H, et al. Hemochromatosis and iron-overload screening in a racially diverse population. New England Journal of Medicine 2005;352:1769–78. [28] Damade R, Rosenthal E, Cacoub P. Les hyperferritine´mies. Ann Med Interne 2000;151:169–77. [29] Katchunga P, Hermans MP, Manwa B, Lepira F, Kashongwe Z, M’BuyambaKabangu JR. Hypertension, insulin resistance and chronic kidney disease in type 2 diabetes patients from South Kivu, DR Congo. Nephrologie & Therapeutique 2010;6:520–5.