- Email: [email protected]
Association of haptoglobin phenotypes with markers of diabetic nephropathy in Type 2 diabetes mellitus
Journal of Diabetes and Its Complications 22 (2008) 384 – 388 WWW.JDCJOURNAL.COM
Association of haptoglobin phenotypes with markers of diabetic nephropathy in Type 2 diabetes mellitus☆ Samir M. Awadallah a,⁎, Suleiman A. Saleh a , Qasem M. Abu Shaqra b , Hisham Hilow c a
Department of Medical Laboratory Sciences, The Hashemite University, PO Box 330077, Zarqa 13133, Jordan b Diagnostic Medical Laboratories, Zarqa 13111, Jordan c Department of Mathematics, University of Jordan, Amman 11942, Jordan Received 4 February 2007; received in revised form 10 June 2007; accepted 22 July 2007
Abstract Background: Recent studies have suggested an inheritable influence of haptoglobin phenotypes on the development of microalbuminuria in diabetic patients. In this study, we aimed to investigate whether microalbuminuria and serum levels of Cystatin C (Cys C) were correlated with haptoglobin phenotypes in patients with Type 2 diabetes mellitus. Methods: Fasting blood samples and first morning urine specimens were collected from 216 patients with type 2 diabetes (mean age 60.8±6.8) and from 108 healthy controls (mean age 59.3±6.5 years). Serum was used for the determination of Cys C, creatinine, and haptoglobin (Hp) phenotypes, while urine specimens were used for the determination of microalbuminuria. Microalbuminuria was expressed as albumin (mg)/creatinine (g) ratio [ACR (mg/g)], and patients were divided into three groups according to the values of ACR: normoalbuminuria (b30), microalbuminuria (30–300) and macroalbuminuria (N300). Results: Irrespective of Hp phenotypes, Cys C, but not creatinine, was significantly higher in patients than that in controls. According to haptoglobin phenotypes, however, the levels of Cys C, duration of diabetes, and age, were significantly higher in patients with Hp1-1 phenotype than that in patients with the other phenotypes. Moreover, micro- and/or macroalbuminuria were present in all diabetic patients, with no significant difference in frequency of occurrence among the three Hp phenotypes. Multiple linear regression revealed that duration of diabetes and age of patient were significant predictors of Cys C (Pb.0001), whereas creatinine associated with age only (P=.016). Logistic regression analysis showed that duration of diabetes [P≤.0001; odds ratio (OR)=1.678, 95% CI=1.43–1.97], and age (P=.0004; OR=0.847, 95% CI=0.77–0.93), significantly associated with micro- and/or macroalbuminuria. Hp phenotypes however, were not associated with any marker of kidney function. Conclusion: Findings from this study demonstrate no association between Hp phenotypes and markers of diabetic nephropathy. Markers of diabetic nephropathy however, are greatly influenced by duration of diabetes and age of patients. © 2008 Elsevier Inc. All rights reserved. Keywords: Cystatin C; Microalbuminuria; Diabetes; Haptoglobin
1. Introduction Microalbuminuria is considered one of the earliest manifestations of diabetic nephropathy and is routinely conducted on patients with diabetes (American Diabetes ☆
This work was supported by a research grant from the Hashemite University, Jordan. ⁎ Corresponding author. Department of Medical Laboratory Sciences, The Hashemite University, PO Box 330077, Zarqa 13133, Jordan. Tel.: +962 5 3903333; fax: +962 5 3826613. E-mail address: [email protected] (S.M. Awadallah). 1056-8727/08/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jdiacomp.2007.07.004
Association, 2003). According to the American Diabetes Association, the gold standard method for measuring albuminuria is a 24-h urine collection (American Diabetes Association, 2001). However, in clinical practice, a more convenient method to detect microalbuminuria is the albumin (mg)/creatinine (g) ratio (ACR) measured in a random urine specimen (American Diabetes Association 2001; Eknoyan et al., 2003). In this case, microalbuminuria is defined as an ACR between 30 and 300 mg/g. Accordingly, values below and above this ACR range are considered as normoalbuminuria and macroalbuminuria, respectively (American Diabetes Association 2001; Eknoyan et al., 2003).
S.M. Awadallah et al. / Journal of Diabetes and Its Complications 22 (2008) 384–388
In addition to microalbuminuria, recent studies have demonstrated that plasma Cystatin C (Cys C) is another useful marker for early detection of renal dysfunction (Coll et al., 2000; Fliser & Ritz, 2001). Serum Cys C is a lowmolecular-weight protein that is steadily produced by all human nucleated cells. It is freely filtered by the glomeruli, reabsorbed, and metabolized in the proximal tubules (Grubb, 2001). Serum levels are independent of muscle mass or sex and are affected mainly by the filtration rate of the kidney. The use of serum Cys C as a marker of altered renal function is well documented and have been claimed to be superior to creatinine for this purpose (Fliser & Ritz, 2001; Dharnidharka, Kwon, & Stevens, 2002). Haptoglobin (Hp) is a polymorphic protein that binds with free hemoglobin (Hb) in circulation. The formation of Hb–Hp complex prevents the loss of iron and iron-driven oxidative damage. Hp exists in three major phenotypes: Hp1-1, 2-1, and 2-2, exhibiting distinct structural and functional properties with significant biological and clinical implications (Asleh & Levy, 2005; Langlois & Delanghe, 1996). Many reports have established a strong association between Hp phenotypes and the occurrence of diseases, namely, complications of diabetes (Asleh & Levy, 2005). In this regard, recent studies on patients with Type 1 and Type 2 diabetes mellitus have demonstrated conflicting results regarding the association between diabetic nephropathy and the inheritance patterns of haptoglobin phenotypes (Asleh & Levy, 2005; Khan, Khan, Ishaq, Sastry, & Rani, 2002; Moczulski, Rogus, & Krolewski, 2001; Nakhoul et al., 2001;). One group of investigators claimed that the inheritance of Hp1-1 phenotype protects against the development of diabetic nephropathy (Nakhoul et al., 2001). They reported microalbuminuria and macroalbuminuria only in diabetic individuals with the Hp 2-1 and 2-2 phenotypes but not in those with Hp1-1 phenotype. On the contrary, another group of investigators observed that diabetic patients with Hp1-1 phenotype suffer from more complications, including diabetic nephropathy, as compared to those with the other phenotypes (Khan et al., 2002). Others, however, reported no influence of the inheritance pattern of Hp on the presence or absence of microalbuminuria (Moczulski et al., 2001). To the best of our knowledge, the genetic predisposition of Hp phenotypes on the development of renal dysfunction in diabetic patients was only correlated with microalbuminuria as an early marker of diabetic nephropathy. No previous studies, however, investigated Cys C in this regard. The aim of this study was to investigate the correlation between Cys C and microalbuminuria on one hand and the inheritance pattern of Hp phenotypes in Type 2 diabetic patients, on the other.
2. Materials and methods Patients with an established diagnosis of Type 2 diabetes mellitus were recruited during their regular attendance of
385
primary health care centers in Amman and Zarqa districts. History of the disease and other relevant data were collected from all participants with prior permission of attending physician. Patients with a history of hypertension or with an evidence of hematuria, urinary tract infection, fever, or other illness contributing to albuminuria were excluded from the study. Routine urinalysis was used as a screening tool to exclude subjects with hematuria or urinary tract infection. The final study group involved 216 patients (112 males and 104 females) with a mean age of 60.8+6.8 years. The control group was randomly selected from age–sex-matched healthy volunteers and was subject to the same exclusion criteria applied to patients. The control group constituted of 108 healthy subjects (64 males and 44 females) with a mean age of 59.3+6.5 years. This study was approved by the Ethics Review Board of our institutions, and all patients and/or controls gave informed written consent prior to the study. Fasting blood samples and a midstream first morning urine specimens were collected from all participating subjects. Microalbuminuria was determined using immunoturbidimetric method (COBAS Integra 400, Roche Diagnostics, Modrice, Germany). The same autoanalyzer was used to measure urinary creatinine based on Jaffe reaction. The ACR was calculated and expressed as milligrams albumin per gram of creatinine (mg/g). Accordingly, patients were divided into three groups according to the ACR results: normoalbuminuria (ACRb30 mg/g), microalbuminuria (ACR 30–300 mg/g), and macroalbuminuria (ACR N300 mg/g). Serum was separated from plain tubes and kept at −20°C till analysis. Serum creatinine and glucose were determined using conventional methods (COBAS Integra 400, Roche Diagnostics). Cys C was determined by ELISA technique (BioVendor Laboratory Medicine, Indianapolis, IN). Hp phenotypes were determined by vertical polyacrylamide gel electrophoresis, and the bands were visualized by staining with benzedine solution (Awadallah & Hamad, 2000). All statistical analyses were performed using SAS under Windows, Version 9 (SAS, Cary, NC, USA). Student's t test and Kruskal–Wallis test were used (as appropriate) to test for significant differences among the studied groups. chi-Square or Fisher's Exact test (whichever appropriate) was used to compare frequencies and/or proportions of categorical variables. The association between markers of nephropathy (Cys C, creatinine, and microalbuminuria) and other factors including: age, duration of diabetes, and Hp phenotypes was investigated using multiple regression analysis. On one hand, the association between each of Cys C and creatinine with age, duration of diabetes, and Hp phenotype was investigated by multiple linear regression analysis. On the other hand, the risk of developing albuminuria (micro- and/ or macroalbuminuria) in association with age, duration of diabetes, and Hp phenotypes was assessed by logistic regression analysis and odds ratio (OR). Results were presented either as mean±S.D. or frequencies. Statistical significance was set at Pb.05.
386
S.M. Awadallah et al. / Journal of Diabetes and Its Complications 22 (2008) 384–388 Table 2 Microalbuminuria and Cys C vs. Hp phenotypes in diabetic patients
3. Results Initial data analysis revealed no significant differences between patients and controls with respect to sex, except for serum creatinine. The levels, as expected, were significantly higher in men than in women within each of the patient and control groups. Because of this, data and results are presented regardless of sex. Table 1 demonstrates the general characteristics of investigated subjects. As expected, the mean levels of glucose were significantly higher in patients than that in controls (P=.000). No significant differences were observed between patients and controls in relation to age and serum creatinine. Cys C, however, was significantly higher in patients than in controls (P=.016). The frequency of haptoglobin phenotypes in both patients and controls were closely similar and were in agreement with Hp frequency of this country (Awadallah & Hamad, 2000). Also, the table shows that 17.1% of patients had macroalbuminuria, compared to 41.7% with microalbuminuria. Table 2 illustrates results of age, duration of diabetes, Cys C, serum creatinine, and types of albuminuria according to Hp phenotypes in the patients group. As shown from the table, the mean age of individuals with Hp1-1 phenotype was significantly higher than that of Hp2-1 and 2-2 (P=.014 and .019, respectively). Similarly, duration of diabetes among patients with Hp1-1 phenotype was significantly higher than that of Hp2-1 and 2-2 (P=.023 and 0.029, respectively). The levels of Cys C were also significantly higher in patients with Hp1-1 than that of Hp2-1 and Hp2-2 (P=.039 and .022, respectively). No significant differences were observed between Hp2-1 and Hp2-2 regarding age, duration of diabetes, and/or Cys C. With respect to serum creatinine, no significant differences were observed between the three Hp phenotypes. The table also demonstrates that both micro- and macroalbuminuria were present in the three Hp phenotypes. Although statistically not significant, the Table 1 General characteristics of investigated patients and controls Parameter
Diabetic patients (n=216)
Controls (n=108)
Age, years Males, n (%) Females, n (%) Glucose, mg/dl Duration of DM, years Serum creatinine, mg/dl Cys C, mg/l Normoalbuminuria, n (%) Microalbuminuria, n (%) Macroalbuminuria, n (%) Hp 1-1, n (%) Hp 2-1, n (%) Hp 2-2, n (%)
60.8±6.8 112 (52%) 104 (48%) 207±49 ⁎⁎ 11.4±7.5 0.91±0.33 1.16±0.28 ⁎ 89 (41.2) 90 (41.7) 37 (17.1) 19 (8.8) 89 (41.2) 108 (50.0)
59.3±6.5 64 (59.3) 44 (40.7) 94±12 NA 0.90±0.22 0.92±0.13 108 (100) NA NA 10 (9.3) 42 (38.9) 56 (51.8)
Results are expressed as mean±S.D. and as percentage. NA, not applicable. ⁎ P=.016 for patients against controls. ⁎⁎ P=.000 for patients against controls.
Hp phenotypes
Age, years Duration of diabetes, years Cys C, mg/l Serum creatinine, mg/dl Normoalbuminuria, n(%) Microalbuminuria, n(%) Macroalbuminuria, n(%)
Hp 1-1 (n=19)
Hp 2-1 (n=89)
Hp 2-2 (n=108)
65.6±9.0 ⁎⁎⁎ 14.6+6.1 ⁎⁎
60.7±6.3 10.9+7.4
60.8±6.9 11.7+7.5
1.30±0.46 ⁎ 0.92±0.25
1.14±0.24 0.91±0.30
1.13±0.27 0.92±0.33
4 (21.1)
40 (44.9)
45 (41.7)
8 (42.1)
37 (41.6)
45 (41.7)
7 (36.8)
12 (13.5)
18 (16.7)
Results are expressed as mean±S.D. and as percentage. ⁎ P=.039 and .022 for Hp1-1 vs. Hp2-1 and Hp2-2 respectively. ⁎⁎ P=.023 and .029 for Hp1-1 vs. Hp2-1 and Hp2-2 respectively. ⁎⁎⁎ P=.014 and .019 for Hp1-1 vs. Hp2-1 and Hp2-2 respectively.
frequency of macroalbuminuria among patients with Hp 1-1 (36.8%) was higher than that in the other phenotypes. It is worth mentioning here that a similar pattern was observed concerning the mean age of subject within the control group. The mean age for individuals with Hp 1-1 (63.7±8.2) was significantly higher than that of Hp 2-1 (59.2±4.8) and Hp 2-2 (58.5±6.4) (P=.029 and 0.016, respectively). Table 3 contains 2 multiple linear regression analyses. In one (Cys C), age (P=.0170) and duration of diabetes (P=.0004) were significantly associated with levels of Cys C. In the other (creatinine), creatinine levels were associated only with age (P=.0165). Both regression analyses (Cys C and creatinine), however, demonstrate no significant association with Hp phenotypes. In addition, logistic regression analysis (Table 4) revealed that the risk of developing microand/or macroalbuminuria was significantly associated with age and duration of diabetes and not with Hp phenotypes. With respect to the control group, simple Pearson's correlation analysis was also performed to elucidate possible associations between age and serum levels of Cys C and creatinine. No significant correlation was observed between age and each of Cys C and creatinine in the control group (r=.136, P=.162 and r=0.108, P=.342, respectively).
4. Discussion The main purpose of this study was to investigate whether microalbuminuria and serum levels of Cys C were correlated with Hp phenotypes in patients with Type 2 diabetes mellitus. In addition, the correlation of age and duration of diabetes with markers of diabetic nephropathy was also assessed by this study. Concerning the association of Hp phenotype with markers of diabetic nephropathy, three major findings were
S.M. Awadallah et al. / Journal of Diabetes and Its Complications 22 (2008) 384–388
387
Table 3 Multiple linear regression analysis for each of Cys C and creatinine (dependent variables) in diabetic patients Independent variables Age Duration of diabetes Hp phenotypes
Cys C
Serum creatinine
Mean square
F value
P value
Mean square
F value
P value
0.3412 0.7774 0.0386
5.79 13.19 0.66
.0170 .0004 .5202
0.5185 0.1302 0.0462
5.84 1.47 0.52
.0165 .2271 .5944
observed. First, we observed that Hp phenotypes had no influence on the occurrence of microalbuminuria in the studied group of patients (Tables 2 and 4). Previous studies on patients with Type 2 diabetes have reported that microand macroalbuminuria occurred only in individuals with Hp2-1 and Hp2-2 phenotypes and not in those with Hp1-1 type (Nakhoul et al., 2001). Accordingly, this group of investigators suggested that the inheritance of the Hp1-1 phenotype, which possesses greater antioxidant properties than the other Hp phenotypes, plays a role in the protection against the development of diabetic nephropathy. Results reported here and by others (Khan et al., 2002; Moczulski et al., 2001), however, observed micro- and macroalbuminuria in all Hp phenotypes, including Hp1-1. In fact, we observed that the frequency of macroalbuminuria (although not significant), was greater among individuals with Hp1-1 than that in other types. Furthermore, this group of patients (Hp1-1) is probably at a greater risk of developing diabetic nephropathy, findings that have been also claimed by other investigators (Khan et al., 2002). Therefore, results from this study demonstrate that Hp1-1 phenotype provides no protection against the development of diabetic nephropathy. Second, we observed that the mean age and duration of diabetes among patients with the Hp1-1 phenotype was significantly higher than that of the other phenotypes (Table 2). Interestingly, similar pattern concerning the association of age with Hp 1-1 was also observed among the control group. This finding might be attributed, in part, to a previous report on the suggested prolonged life expectancy among individuals with Hp1-1 phenotype (Hamad & Awadallah, 2000). Previous studies have demonstrated that the antioxidative property of Hp1-1 phenotype is more powerful than that of the other phenotypes, suggesting a more effective protection against oxidative stress-related diseases, including complications of diabetes (Asleh & Levy, 2005; Tseng, Lin, Huang, Liu, Mao, 2004). Therefore, it would be expected that individuals with Hp1-1 phenotype who are protected against diseases and oxidative stress would live longer than those with the other Hp phenotypes. It is likely then that this group of individuals happened to be among the diabetic patients investigated by this study. The third observation from this study is related to the association of higher levels of Cys C in patients with Hp1-1 phenotype than that in the other phenotypes. Since this finding was observed only in the patient group (diabetics) and not in the control group, it would be concluded then
that duration of diabetes and not age is the major factor contributing to higher levels of Cys C observed in this group of patients (Hp 1-1). Moreover, results from multiple regression analysis (Table 3) support this conclusion by showing that duration of diabetes was associated more significantly with Cys C than age. In general, and irrespective of Hp phenotypes, results from this study were in agreement with previous reports on the levels of serum creatinine, Cys C, and the occurrence of microalbuminuria among patients with Type 2 diabetes (Aksun et al., 2004; Coll et al., 2000; Dharnidharka et al., 2002; Grubb, 2001; Wasen et al., 2004). In this study population, the finding that the levels of Cys C were higher in patients than that in controls (Table 1) supports the concept that Cys C is a more sensitive marker than creatinine in detection of diabetic nephropathy (Dharnidharka et al., 2002). Furthermore, results reported by this study and by others (Fliser & Ritz, 2001; Mohan et al., 2000; Mussap et al., 2002) demonstrate that markers of diabetic nephropathy are greatly influenced by duration of diabetes and by age of patients. In addition, the observation that the levels of Cys C were influenced more significantly than creatinine by duration of diabetes and age (Table 3) also suggests that Cys C is better than creatinine as an indicator of diabetic nephropathy. Similar findings were also observed regarding the influence of duration of diabetes and age on the occurrence of micro- and macroalbuminuria in the studied group of patients (Table 4). Collectively, these results demonstrate that duration of diabetes and/or age are major factors contributing to alteration of markers related to kidney function. In conclusion, results from this study observed no protective role of Hp1-1 phenotype in the development of diabetic nephropathy. Additionally, findings related to this studied population of Type 2 diabetic patients suggest that the association between markers of diabetic nephropathy and haptoglobin phenotypes is greatly influenced by duration of diabetes and/or age of patients. Conclusions from this
Table 4 Predictive values (OR) of risk factors for micro- and/or macroalbuminuria using multiple logistic regression analysis Independent variables
Wald chi-square
P value
OR
95% CI
Age Duration of diabetes Hp phenotypes
12.355 40.382 1.438
.0004 b.0001 .4871
0.847 1.678 1.032
0.77–0.93 1.43–1.97 0.32–5.26
388
S.M. Awadallah et al. / Journal of Diabetes and Its Complications 22 (2008) 384–388
study, however, may have major limitations that need to be acknowledged. Since all patients were Jordanian in the present study, whether these conclusions can be extended to other population awaits confirmation in racematched studies. Acknowledgments The authors wish to thank Sultan Medical Laboratories in Amman, Jordan, for their assistance in using their autoanalyzer (Cobas Integra). References Aksun, S. A., Ozmen, D., Ozmen, B., Parildar, Z., Mutaf, I., Turgan, N., Habif, S., Kumanliogluc, K., & Bayindir, O. (2004). Beta2-microglobulin and Cystatin C in Type 2 diabetes: Assessment of diabetic nephropathy. Experimental and Clinical Endocrinology & Diabetes, 112, 1195−1200. American Diabetes Association. (2001). Clinical practice recommendations 2001: Diabetic nephropathy. Diabetes Care, 24 (Suppl 1), S69−S72. American Diabetes Association. (2003). Diabetic nephropathy (position statement). Diabetes Care, 26 (Suppl 1), S94−S98. Asleh, R., & Levy, A. P. (2005). In vivo and in vitro studies establishing haptoglobin as a major susceptibility gene for diabetic vascular disease. Vascular Health and Risk Management, 1, 19−28. Awadallah, S., & Hamad, M. (2000). The prevalence of type II diabetes mellitus is haptoglobin phenotype-independent. Cytobios, 101, 145−150. Coll, E., Botey, A., Alvarez, L., et al. (2000). Serum Cystatin C as a new marker for noninvasive estimation of glomerular filtration rate and as a marker for early renal impairment. American Journal of Kidney Diseases, 36, 29−34. Dharnidharka, V. R., Kwon, C., & Stevens, G. (2002). Serum Cystatin C is superior to serum creatinine as a marker of kidney function: A metaanalysis. American Journal of Kidney Diseases, 40, 221−226.
Eknoyan, G., Hostetter, T., Bakris, G. L., Hebert, L., Levey, A. S., Parving, H. H., Steffes, M. W., & Toto, R. (2003). Proteinuria and other markers of chronic kidney disease: A position statement of the National Kidney Foundation (NKF) and the National Institute of Diabetes and digestive and Kidney diseases (NIDDK). American Journal of Kidney Diseases, 42, 617−622. Fliser, D., & Ritz, E. (2001). Serum Cystatin C concentration as a marker of renal dysfunction in the elderly. American Journal of Kidney Diseases, 37, 79−83. Grubb, A. O. (2001). Cystatin C: Properties and use as diagnostic marker. Advances in Clinical Chemistry, 35, 63−99. Hamad, M., & Awadallah, S. (2000). Age group-associated variations in the pattern of Hp type distribution in Jordanians. Clinica Chimica Acta, 300, 75−81. Khan, A. G., Khan, N., Ishaq, M., Sastry, E. P., & Rani, T. S. H. (2002). Increased frequency of Haptoglobin phenotype Hp 1-1 in diabetic patients with complications. International Journal of Diabetes in Developing Countries, 22, 100−102. Langlois, M. R., & Delanghe, J. R. (1996). Biological and clinical significance of haptoglobin polymorphism in humans. Clinical Chemistry, 42, 1589−1600. Moczulski, D. K., Rogus, J. J., & Krolewski, A. S. (2001). Comment to: FM Nakhoul et al (2001) Haptoglobin phenotype and diabetes Diabetologia, 44:602-604. Diabetologia, 44, 2237−2239. Mohan, V., Meera, R., Premalatha, G., Deepa, R., Miranda, P., & Rema, M. (2000). Frequency of proteinuria in Type 2 diabetes mellitus seen at a diabetes centre in south India. Postgraduate Medical Journal, 76, 569−573. Mussap, M., Dalla Vestra, M., Fioretto, P., Saller, A., Varagnolo, M., Nosadini, R., & Plebani, M. (2002). Cystatin C is a more sensitive marker than creatinine for estimation of GFR in Type 2 diabetic patients. Kidney International, 61, 1453−1461. Nakhoul, F. M., Zoabi, R., Kanter, Y., Zoabi, M., Skorecki, K., Hochberg, I., Leibu, R., Miller, B., & Levy, A. P. (2001). Haptoglobin phenotypes and diabetic nephropathy. Diabetologia, 44, 602−604. Tseng, C. F., Lin, C. C., Huang, H. Y., Liu, H. C., & Mao, S. J. (2004). Antioxidant role of human haptoglobin. Proteomics, 4, 2221−2228. Wasen, E., Isoaho, R., Mattila, K., Vahlberg, T., Kivela, S. L., & Irjala, K. (2004). Renal impairment associated with diabetes in the elderly. Diabetes Care, 27, 2648−2653.
Association of haptoglobin phenotypes with markers of diabetic nephropathy in Type 2 diabetes mellitus☆ Samir M. Awadallah a,⁎, Suleiman A. Saleh a , Qasem M. Abu Shaqra b , Hisham Hilow c a
Department of Medical Laboratory Sciences, The Hashemite University, PO Box 330077, Zarqa 13133, Jordan b Diagnostic Medical Laboratories, Zarqa 13111, Jordan c Department of Mathematics, University of Jordan, Amman 11942, Jordan Received 4 February 2007; received in revised form 10 June 2007; accepted 22 July 2007
Abstract Background: Recent studies have suggested an inheritable influence of haptoglobin phenotypes on the development of microalbuminuria in diabetic patients. In this study, we aimed to investigate whether microalbuminuria and serum levels of Cystatin C (Cys C) were correlated with haptoglobin phenotypes in patients with Type 2 diabetes mellitus. Methods: Fasting blood samples and first morning urine specimens were collected from 216 patients with type 2 diabetes (mean age 60.8±6.8) and from 108 healthy controls (mean age 59.3±6.5 years). Serum was used for the determination of Cys C, creatinine, and haptoglobin (Hp) phenotypes, while urine specimens were used for the determination of microalbuminuria. Microalbuminuria was expressed as albumin (mg)/creatinine (g) ratio [ACR (mg/g)], and patients were divided into three groups according to the values of ACR: normoalbuminuria (b30), microalbuminuria (30–300) and macroalbuminuria (N300). Results: Irrespective of Hp phenotypes, Cys C, but not creatinine, was significantly higher in patients than that in controls. According to haptoglobin phenotypes, however, the levels of Cys C, duration of diabetes, and age, were significantly higher in patients with Hp1-1 phenotype than that in patients with the other phenotypes. Moreover, micro- and/or macroalbuminuria were present in all diabetic patients, with no significant difference in frequency of occurrence among the three Hp phenotypes. Multiple linear regression revealed that duration of diabetes and age of patient were significant predictors of Cys C (Pb.0001), whereas creatinine associated with age only (P=.016). Logistic regression analysis showed that duration of diabetes [P≤.0001; odds ratio (OR)=1.678, 95% CI=1.43–1.97], and age (P=.0004; OR=0.847, 95% CI=0.77–0.93), significantly associated with micro- and/or macroalbuminuria. Hp phenotypes however, were not associated with any marker of kidney function. Conclusion: Findings from this study demonstrate no association between Hp phenotypes and markers of diabetic nephropathy. Markers of diabetic nephropathy however, are greatly influenced by duration of diabetes and age of patients. © 2008 Elsevier Inc. All rights reserved. Keywords: Cystatin C; Microalbuminuria; Diabetes; Haptoglobin
1. Introduction Microalbuminuria is considered one of the earliest manifestations of diabetic nephropathy and is routinely conducted on patients with diabetes (American Diabetes ☆
This work was supported by a research grant from the Hashemite University, Jordan. ⁎ Corresponding author. Department of Medical Laboratory Sciences, The Hashemite University, PO Box 330077, Zarqa 13133, Jordan. Tel.: +962 5 3903333; fax: +962 5 3826613. E-mail address: [email protected] (S.M. Awadallah). 1056-8727/08/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jdiacomp.2007.07.004
Association, 2003). According to the American Diabetes Association, the gold standard method for measuring albuminuria is a 24-h urine collection (American Diabetes Association, 2001). However, in clinical practice, a more convenient method to detect microalbuminuria is the albumin (mg)/creatinine (g) ratio (ACR) measured in a random urine specimen (American Diabetes Association 2001; Eknoyan et al., 2003). In this case, microalbuminuria is defined as an ACR between 30 and 300 mg/g. Accordingly, values below and above this ACR range are considered as normoalbuminuria and macroalbuminuria, respectively (American Diabetes Association 2001; Eknoyan et al., 2003).
S.M. Awadallah et al. / Journal of Diabetes and Its Complications 22 (2008) 384–388
In addition to microalbuminuria, recent studies have demonstrated that plasma Cystatin C (Cys C) is another useful marker for early detection of renal dysfunction (Coll et al., 2000; Fliser & Ritz, 2001). Serum Cys C is a lowmolecular-weight protein that is steadily produced by all human nucleated cells. It is freely filtered by the glomeruli, reabsorbed, and metabolized in the proximal tubules (Grubb, 2001). Serum levels are independent of muscle mass or sex and are affected mainly by the filtration rate of the kidney. The use of serum Cys C as a marker of altered renal function is well documented and have been claimed to be superior to creatinine for this purpose (Fliser & Ritz, 2001; Dharnidharka, Kwon, & Stevens, 2002). Haptoglobin (Hp) is a polymorphic protein that binds with free hemoglobin (Hb) in circulation. The formation of Hb–Hp complex prevents the loss of iron and iron-driven oxidative damage. Hp exists in three major phenotypes: Hp1-1, 2-1, and 2-2, exhibiting distinct structural and functional properties with significant biological and clinical implications (Asleh & Levy, 2005; Langlois & Delanghe, 1996). Many reports have established a strong association between Hp phenotypes and the occurrence of diseases, namely, complications of diabetes (Asleh & Levy, 2005). In this regard, recent studies on patients with Type 1 and Type 2 diabetes mellitus have demonstrated conflicting results regarding the association between diabetic nephropathy and the inheritance patterns of haptoglobin phenotypes (Asleh & Levy, 2005; Khan, Khan, Ishaq, Sastry, & Rani, 2002; Moczulski, Rogus, & Krolewski, 2001; Nakhoul et al., 2001;). One group of investigators claimed that the inheritance of Hp1-1 phenotype protects against the development of diabetic nephropathy (Nakhoul et al., 2001). They reported microalbuminuria and macroalbuminuria only in diabetic individuals with the Hp 2-1 and 2-2 phenotypes but not in those with Hp1-1 phenotype. On the contrary, another group of investigators observed that diabetic patients with Hp1-1 phenotype suffer from more complications, including diabetic nephropathy, as compared to those with the other phenotypes (Khan et al., 2002). Others, however, reported no influence of the inheritance pattern of Hp on the presence or absence of microalbuminuria (Moczulski et al., 2001). To the best of our knowledge, the genetic predisposition of Hp phenotypes on the development of renal dysfunction in diabetic patients was only correlated with microalbuminuria as an early marker of diabetic nephropathy. No previous studies, however, investigated Cys C in this regard. The aim of this study was to investigate the correlation between Cys C and microalbuminuria on one hand and the inheritance pattern of Hp phenotypes in Type 2 diabetic patients, on the other.
2. Materials and methods Patients with an established diagnosis of Type 2 diabetes mellitus were recruited during their regular attendance of
385
primary health care centers in Amman and Zarqa districts. History of the disease and other relevant data were collected from all participants with prior permission of attending physician. Patients with a history of hypertension or with an evidence of hematuria, urinary tract infection, fever, or other illness contributing to albuminuria were excluded from the study. Routine urinalysis was used as a screening tool to exclude subjects with hematuria or urinary tract infection. The final study group involved 216 patients (112 males and 104 females) with a mean age of 60.8+6.8 years. The control group was randomly selected from age–sex-matched healthy volunteers and was subject to the same exclusion criteria applied to patients. The control group constituted of 108 healthy subjects (64 males and 44 females) with a mean age of 59.3+6.5 years. This study was approved by the Ethics Review Board of our institutions, and all patients and/or controls gave informed written consent prior to the study. Fasting blood samples and a midstream first morning urine specimens were collected from all participating subjects. Microalbuminuria was determined using immunoturbidimetric method (COBAS Integra 400, Roche Diagnostics, Modrice, Germany). The same autoanalyzer was used to measure urinary creatinine based on Jaffe reaction. The ACR was calculated and expressed as milligrams albumin per gram of creatinine (mg/g). Accordingly, patients were divided into three groups according to the ACR results: normoalbuminuria (ACRb30 mg/g), microalbuminuria (ACR 30–300 mg/g), and macroalbuminuria (ACR N300 mg/g). Serum was separated from plain tubes and kept at −20°C till analysis. Serum creatinine and glucose were determined using conventional methods (COBAS Integra 400, Roche Diagnostics). Cys C was determined by ELISA technique (BioVendor Laboratory Medicine, Indianapolis, IN). Hp phenotypes were determined by vertical polyacrylamide gel electrophoresis, and the bands were visualized by staining with benzedine solution (Awadallah & Hamad, 2000). All statistical analyses were performed using SAS under Windows, Version 9 (SAS, Cary, NC, USA). Student's t test and Kruskal–Wallis test were used (as appropriate) to test for significant differences among the studied groups. chi-Square or Fisher's Exact test (whichever appropriate) was used to compare frequencies and/or proportions of categorical variables. The association between markers of nephropathy (Cys C, creatinine, and microalbuminuria) and other factors including: age, duration of diabetes, and Hp phenotypes was investigated using multiple regression analysis. On one hand, the association between each of Cys C and creatinine with age, duration of diabetes, and Hp phenotype was investigated by multiple linear regression analysis. On the other hand, the risk of developing albuminuria (micro- and/ or macroalbuminuria) in association with age, duration of diabetes, and Hp phenotypes was assessed by logistic regression analysis and odds ratio (OR). Results were presented either as mean±S.D. or frequencies. Statistical significance was set at Pb.05.
386
S.M. Awadallah et al. / Journal of Diabetes and Its Complications 22 (2008) 384–388 Table 2 Microalbuminuria and Cys C vs. Hp phenotypes in diabetic patients
3. Results Initial data analysis revealed no significant differences between patients and controls with respect to sex, except for serum creatinine. The levels, as expected, were significantly higher in men than in women within each of the patient and control groups. Because of this, data and results are presented regardless of sex. Table 1 demonstrates the general characteristics of investigated subjects. As expected, the mean levels of glucose were significantly higher in patients than that in controls (P=.000). No significant differences were observed between patients and controls in relation to age and serum creatinine. Cys C, however, was significantly higher in patients than in controls (P=.016). The frequency of haptoglobin phenotypes in both patients and controls were closely similar and were in agreement with Hp frequency of this country (Awadallah & Hamad, 2000). Also, the table shows that 17.1% of patients had macroalbuminuria, compared to 41.7% with microalbuminuria. Table 2 illustrates results of age, duration of diabetes, Cys C, serum creatinine, and types of albuminuria according to Hp phenotypes in the patients group. As shown from the table, the mean age of individuals with Hp1-1 phenotype was significantly higher than that of Hp2-1 and 2-2 (P=.014 and .019, respectively). Similarly, duration of diabetes among patients with Hp1-1 phenotype was significantly higher than that of Hp2-1 and 2-2 (P=.023 and 0.029, respectively). The levels of Cys C were also significantly higher in patients with Hp1-1 than that of Hp2-1 and Hp2-2 (P=.039 and .022, respectively). No significant differences were observed between Hp2-1 and Hp2-2 regarding age, duration of diabetes, and/or Cys C. With respect to serum creatinine, no significant differences were observed between the three Hp phenotypes. The table also demonstrates that both micro- and macroalbuminuria were present in the three Hp phenotypes. Although statistically not significant, the Table 1 General characteristics of investigated patients and controls Parameter
Diabetic patients (n=216)
Controls (n=108)
Age, years Males, n (%) Females, n (%) Glucose, mg/dl Duration of DM, years Serum creatinine, mg/dl Cys C, mg/l Normoalbuminuria, n (%) Microalbuminuria, n (%) Macroalbuminuria, n (%) Hp 1-1, n (%) Hp 2-1, n (%) Hp 2-2, n (%)
60.8±6.8 112 (52%) 104 (48%) 207±49 ⁎⁎ 11.4±7.5 0.91±0.33 1.16±0.28 ⁎ 89 (41.2) 90 (41.7) 37 (17.1) 19 (8.8) 89 (41.2) 108 (50.0)
59.3±6.5 64 (59.3) 44 (40.7) 94±12 NA 0.90±0.22 0.92±0.13 108 (100) NA NA 10 (9.3) 42 (38.9) 56 (51.8)
Results are expressed as mean±S.D. and as percentage. NA, not applicable. ⁎ P=.016 for patients against controls. ⁎⁎ P=.000 for patients against controls.
Hp phenotypes
Age, years Duration of diabetes, years Cys C, mg/l Serum creatinine, mg/dl Normoalbuminuria, n(%) Microalbuminuria, n(%) Macroalbuminuria, n(%)
Hp 1-1 (n=19)
Hp 2-1 (n=89)
Hp 2-2 (n=108)
65.6±9.0 ⁎⁎⁎ 14.6+6.1 ⁎⁎
60.7±6.3 10.9+7.4
60.8±6.9 11.7+7.5
1.30±0.46 ⁎ 0.92±0.25
1.14±0.24 0.91±0.30
1.13±0.27 0.92±0.33
4 (21.1)
40 (44.9)
45 (41.7)
8 (42.1)
37 (41.6)
45 (41.7)
7 (36.8)
12 (13.5)
18 (16.7)
Results are expressed as mean±S.D. and as percentage. ⁎ P=.039 and .022 for Hp1-1 vs. Hp2-1 and Hp2-2 respectively. ⁎⁎ P=.023 and .029 for Hp1-1 vs. Hp2-1 and Hp2-2 respectively. ⁎⁎⁎ P=.014 and .019 for Hp1-1 vs. Hp2-1 and Hp2-2 respectively.
frequency of macroalbuminuria among patients with Hp 1-1 (36.8%) was higher than that in the other phenotypes. It is worth mentioning here that a similar pattern was observed concerning the mean age of subject within the control group. The mean age for individuals with Hp 1-1 (63.7±8.2) was significantly higher than that of Hp 2-1 (59.2±4.8) and Hp 2-2 (58.5±6.4) (P=.029 and 0.016, respectively). Table 3 contains 2 multiple linear regression analyses. In one (Cys C), age (P=.0170) and duration of diabetes (P=.0004) were significantly associated with levels of Cys C. In the other (creatinine), creatinine levels were associated only with age (P=.0165). Both regression analyses (Cys C and creatinine), however, demonstrate no significant association with Hp phenotypes. In addition, logistic regression analysis (Table 4) revealed that the risk of developing microand/or macroalbuminuria was significantly associated with age and duration of diabetes and not with Hp phenotypes. With respect to the control group, simple Pearson's correlation analysis was also performed to elucidate possible associations between age and serum levels of Cys C and creatinine. No significant correlation was observed between age and each of Cys C and creatinine in the control group (r=.136, P=.162 and r=0.108, P=.342, respectively).
4. Discussion The main purpose of this study was to investigate whether microalbuminuria and serum levels of Cys C were correlated with Hp phenotypes in patients with Type 2 diabetes mellitus. In addition, the correlation of age and duration of diabetes with markers of diabetic nephropathy was also assessed by this study. Concerning the association of Hp phenotype with markers of diabetic nephropathy, three major findings were
S.M. Awadallah et al. / Journal of Diabetes and Its Complications 22 (2008) 384–388
387
Table 3 Multiple linear regression analysis for each of Cys C and creatinine (dependent variables) in diabetic patients Independent variables Age Duration of diabetes Hp phenotypes
Cys C
Serum creatinine
Mean square
F value
P value
Mean square
F value
P value
0.3412 0.7774 0.0386
5.79 13.19 0.66
.0170 .0004 .5202
0.5185 0.1302 0.0462
5.84 1.47 0.52
.0165 .2271 .5944
observed. First, we observed that Hp phenotypes had no influence on the occurrence of microalbuminuria in the studied group of patients (Tables 2 and 4). Previous studies on patients with Type 2 diabetes have reported that microand macroalbuminuria occurred only in individuals with Hp2-1 and Hp2-2 phenotypes and not in those with Hp1-1 type (Nakhoul et al., 2001). Accordingly, this group of investigators suggested that the inheritance of the Hp1-1 phenotype, which possesses greater antioxidant properties than the other Hp phenotypes, plays a role in the protection against the development of diabetic nephropathy. Results reported here and by others (Khan et al., 2002; Moczulski et al., 2001), however, observed micro- and macroalbuminuria in all Hp phenotypes, including Hp1-1. In fact, we observed that the frequency of macroalbuminuria (although not significant), was greater among individuals with Hp1-1 than that in other types. Furthermore, this group of patients (Hp1-1) is probably at a greater risk of developing diabetic nephropathy, findings that have been also claimed by other investigators (Khan et al., 2002). Therefore, results from this study demonstrate that Hp1-1 phenotype provides no protection against the development of diabetic nephropathy. Second, we observed that the mean age and duration of diabetes among patients with the Hp1-1 phenotype was significantly higher than that of the other phenotypes (Table 2). Interestingly, similar pattern concerning the association of age with Hp 1-1 was also observed among the control group. This finding might be attributed, in part, to a previous report on the suggested prolonged life expectancy among individuals with Hp1-1 phenotype (Hamad & Awadallah, 2000). Previous studies have demonstrated that the antioxidative property of Hp1-1 phenotype is more powerful than that of the other phenotypes, suggesting a more effective protection against oxidative stress-related diseases, including complications of diabetes (Asleh & Levy, 2005; Tseng, Lin, Huang, Liu, Mao, 2004). Therefore, it would be expected that individuals with Hp1-1 phenotype who are protected against diseases and oxidative stress would live longer than those with the other Hp phenotypes. It is likely then that this group of individuals happened to be among the diabetic patients investigated by this study. The third observation from this study is related to the association of higher levels of Cys C in patients with Hp1-1 phenotype than that in the other phenotypes. Since this finding was observed only in the patient group (diabetics) and not in the control group, it would be concluded then
that duration of diabetes and not age is the major factor contributing to higher levels of Cys C observed in this group of patients (Hp 1-1). Moreover, results from multiple regression analysis (Table 3) support this conclusion by showing that duration of diabetes was associated more significantly with Cys C than age. In general, and irrespective of Hp phenotypes, results from this study were in agreement with previous reports on the levels of serum creatinine, Cys C, and the occurrence of microalbuminuria among patients with Type 2 diabetes (Aksun et al., 2004; Coll et al., 2000; Dharnidharka et al., 2002; Grubb, 2001; Wasen et al., 2004). In this study population, the finding that the levels of Cys C were higher in patients than that in controls (Table 1) supports the concept that Cys C is a more sensitive marker than creatinine in detection of diabetic nephropathy (Dharnidharka et al., 2002). Furthermore, results reported by this study and by others (Fliser & Ritz, 2001; Mohan et al., 2000; Mussap et al., 2002) demonstrate that markers of diabetic nephropathy are greatly influenced by duration of diabetes and by age of patients. In addition, the observation that the levels of Cys C were influenced more significantly than creatinine by duration of diabetes and age (Table 3) also suggests that Cys C is better than creatinine as an indicator of diabetic nephropathy. Similar findings were also observed regarding the influence of duration of diabetes and age on the occurrence of micro- and macroalbuminuria in the studied group of patients (Table 4). Collectively, these results demonstrate that duration of diabetes and/or age are major factors contributing to alteration of markers related to kidney function. In conclusion, results from this study observed no protective role of Hp1-1 phenotype in the development of diabetic nephropathy. Additionally, findings related to this studied population of Type 2 diabetic patients suggest that the association between markers of diabetic nephropathy and haptoglobin phenotypes is greatly influenced by duration of diabetes and/or age of patients. Conclusions from this
Table 4 Predictive values (OR) of risk factors for micro- and/or macroalbuminuria using multiple logistic regression analysis Independent variables
Wald chi-square
P value
OR
95% CI
Age Duration of diabetes Hp phenotypes
12.355 40.382 1.438
.0004 b.0001 .4871
0.847 1.678 1.032
0.77–0.93 1.43–1.97 0.32–5.26
388
S.M. Awadallah et al. / Journal of Diabetes and Its Complications 22 (2008) 384–388
study, however, may have major limitations that need to be acknowledged. Since all patients were Jordanian in the present study, whether these conclusions can be extended to other population awaits confirmation in racematched studies. Acknowledgments The authors wish to thank Sultan Medical Laboratories in Amman, Jordan, for their assistance in using their autoanalyzer (Cobas Integra). References Aksun, S. A., Ozmen, D., Ozmen, B., Parildar, Z., Mutaf, I., Turgan, N., Habif, S., Kumanliogluc, K., & Bayindir, O. (2004). Beta2-microglobulin and Cystatin C in Type 2 diabetes: Assessment of diabetic nephropathy. Experimental and Clinical Endocrinology & Diabetes, 112, 1195−1200. American Diabetes Association. (2001). Clinical practice recommendations 2001: Diabetic nephropathy. Diabetes Care, 24 (Suppl 1), S69−S72. American Diabetes Association. (2003). Diabetic nephropathy (position statement). Diabetes Care, 26 (Suppl 1), S94−S98. Asleh, R., & Levy, A. P. (2005). In vivo and in vitro studies establishing haptoglobin as a major susceptibility gene for diabetic vascular disease. Vascular Health and Risk Management, 1, 19−28. Awadallah, S., & Hamad, M. (2000). The prevalence of type II diabetes mellitus is haptoglobin phenotype-independent. Cytobios, 101, 145−150. Coll, E., Botey, A., Alvarez, L., et al. (2000). Serum Cystatin C as a new marker for noninvasive estimation of glomerular filtration rate and as a marker for early renal impairment. American Journal of Kidney Diseases, 36, 29−34. Dharnidharka, V. R., Kwon, C., & Stevens, G. (2002). Serum Cystatin C is superior to serum creatinine as a marker of kidney function: A metaanalysis. American Journal of Kidney Diseases, 40, 221−226.
Eknoyan, G., Hostetter, T., Bakris, G. L., Hebert, L., Levey, A. S., Parving, H. H., Steffes, M. W., & Toto, R. (2003). Proteinuria and other markers of chronic kidney disease: A position statement of the National Kidney Foundation (NKF) and the National Institute of Diabetes and digestive and Kidney diseases (NIDDK). American Journal of Kidney Diseases, 42, 617−622. Fliser, D., & Ritz, E. (2001). Serum Cystatin C concentration as a marker of renal dysfunction in the elderly. American Journal of Kidney Diseases, 37, 79−83. Grubb, A. O. (2001). Cystatin C: Properties and use as diagnostic marker. Advances in Clinical Chemistry, 35, 63−99. Hamad, M., & Awadallah, S. (2000). Age group-associated variations in the pattern of Hp type distribution in Jordanians. Clinica Chimica Acta, 300, 75−81. Khan, A. G., Khan, N., Ishaq, M., Sastry, E. P., & Rani, T. S. H. (2002). Increased frequency of Haptoglobin phenotype Hp 1-1 in diabetic patients with complications. International Journal of Diabetes in Developing Countries, 22, 100−102. Langlois, M. R., & Delanghe, J. R. (1996). Biological and clinical significance of haptoglobin polymorphism in humans. Clinical Chemistry, 42, 1589−1600. Moczulski, D. K., Rogus, J. J., & Krolewski, A. S. (2001). Comment to: FM Nakhoul et al (2001) Haptoglobin phenotype and diabetes Diabetologia, 44:602-604. Diabetologia, 44, 2237−2239. Mohan, V., Meera, R., Premalatha, G., Deepa, R., Miranda, P., & Rema, M. (2000). Frequency of proteinuria in Type 2 diabetes mellitus seen at a diabetes centre in south India. Postgraduate Medical Journal, 76, 569−573. Mussap, M., Dalla Vestra, M., Fioretto, P., Saller, A., Varagnolo, M., Nosadini, R., & Plebani, M. (2002). Cystatin C is a more sensitive marker than creatinine for estimation of GFR in Type 2 diabetic patients. Kidney International, 61, 1453−1461. Nakhoul, F. M., Zoabi, R., Kanter, Y., Zoabi, M., Skorecki, K., Hochberg, I., Leibu, R., Miller, B., & Levy, A. P. (2001). Haptoglobin phenotypes and diabetic nephropathy. Diabetologia, 44, 602−604. Tseng, C. F., Lin, C. C., Huang, H. Y., Liu, H. C., & Mao, S. J. (2004). Antioxidant role of human haptoglobin. Proteomics, 4, 2221−2228. Wasen, E., Isoaho, R., Mattila, K., Vahlberg, T., Kivela, S. L., & Irjala, K. (2004). Renal impairment associated with diabetes in the elderly. Diabetes Care, 27, 2648−2653.