Ghulam Abbas Kandhro. Received: 24 August 2007 /Accepted: 2 November 2007 /. Published online: 11 January 2008. © Humana Press Inc. 2007. Abstract ...
Biol Trace Elem Res (2008) 122:1–18 DOI 10.1007/s12011-007-8062-y
Copper, Chromium, Manganese, Iron, Nickel, and Zinc Levels in Biological Samples of Diabetes Mellitus Patients Tasneem Gul Kazi & Hassan Imran Afridi & Naveed Kazi & Mohammad Khan Jamali & Mohammad Bilal Arain & Nussarat Jalbani & Ghulam Abbas Kandhro
Received: 24 August 2007 / Accepted: 2 November 2007 / Published online: 11 January 2008 # Humana Press Inc. 2007
Abstract There is accumulating evidence that the metabolism of several trace elements is altered in diabetes mellitus and that these nutrients might have specific roles in the pathogenesis and progress of this disease. The aim of present study was to compare the level of essential trace elements, chromium (Cr), copper (Cu), iron (Fe), manganese (Mn), nickel (Ni), and zinc (Zn) in biological samples (whole blood, urine, and scalp hair) of patients who have diabetes mellitus type 2 (n=257), with those of nondiabetic control subjects (n=166), age ranged (45–75) of both genders. The element concentrations were measured by means of an atomic absorption spectrophotometer after microwave-induced acid digestion. The validity and accuracy was checked by conventional wet-acid-digestion method and using certified reference materials. The overall recoveries of all elements were found in the range of (97.60–99.49%) of certified values. The results of this study showed that the mean values of Zn, Mn, and Cr were significantly reduced in blood and scalp-hair samples of diabetic patients as compared to control subjects of both genders (p7.7 mmol/l) and postprandial blood glucose >200 mg/dl (>11.1 mmol/l), 2 h after 75 g of oral glucose. Only type-2 diabetic patients were included in the study. The duration of diabetes among the patients was 8–20 years. A questionnaire was administered to all patients to collect details of their physical data, ethnic origin, duration of diabetes, insulin therapy, dietary habits, age, and consent. Physical examinations were performed at the Basic Health Unit of Hyderabad City, Pakistan. At the
Table 3 Measurement Conditions of ETAAS for Determination of Mn, Cr and Ni
Parameters Lamp Current (mA) Mn and Cr Ni Sample volume (μl) Carrier gas (Argon; ml/min) Background correction (D2 lamp) Drying temperature (°C) Time (s) Hold (s) Ashing temperature for Mn and Cr (°C) Ashing time for Mn and Cr (s) Ramp/hold for Mn and Cr (s) Ashing temperature for Ni (°C) Ashing time for Ni (s) Ramp/hold for Ni (s) Atomization temperature (°C) For Mn For Cr For Ni Atomization time (s) Ramp/hold (s) Cleaning temperature (°C) Cleaning time (s) Ramp/hold (s)
Conditions
7.5 3.5 10 200 Cuvette (cup) 140 15 5 1,400 10 20 1,000 10 20 2,200 2,500 2,300 0 5.0 2600 1 3
Metals in Diabetes Mellitus
5
Table 4 Demographics of Controls and Diabetic Patients Age groups
46–60 61–75 Total
Control groups
Diabetic patients
Men
Women
Men
Women
43 37 80
47 39 86
72 65 137
63 71 120
start of the study, the participants’ weight, height, blood pressure, and biochemical data were measured and recorded. Forty per cent of patients in our survey had documented vascular disease, 30% having a history of cardiovascular disease, and 50% had hypertension (receiving antihypertensive therapy). The 60% of understudied patients were obese, while 40% of DM patients were insulin dependent. There were no statistically significant differences of height and weight between nondiabetic controls and diabetic patients. Venous blood (3–5 ml) was sampled by using metal-free Safety Vacutainer bloodcollecting tubes containing >1.5 mg K2EDTA obtained from Becton Dickinson, Rutherford, USA. The samples were stored at −20°C until required for analysis. Morning urine samples were collected in acid-washed, decontaminated 100 ml polyethylene tubes (Kartell1, Milan, Italy). In between sampling sessions, the container is wrapped in a clean polyethylene bag. Urine samples were acidified with ultrapure concentrated HNO3, (l% v/v) and kept at −4°C. Before subsampling for analysis, the sample should be shaken vigorously for 1 min to ensure a homogeneous suspension. The hair samples (approximately 0.5 g each) were taken from five different parts of the scalp (frontal, cranial, occipital, right, and left lateral). The scalp-hair samples were washed and treated as reported in previous studies [25]. Microwave-Assisted Acid Digestion (MD) For digestion of biological samples, duplicate samples of dried scalp hair and five replicate samples of certified reference material (CRM) BCR 397 (200 mg). For blood and urine, 0.5 ml of each sample and CRMs were directly taken into Teflon polytetrafluoroethylene (PTFE) flasks. Added to each flask was 2 ml of a freshly prepared mixture of concentrated HNO3–H2O2 (2:1, v/v), and this was kept for 10 min at room temperature, then the flasks were placed in covered PTFE container. This was then heated following a one-stage digestion program at 80% of total power (900 W). Complete digestion of blood and urine required 2–4 min, while 5–8 min was necessary for scalp-hair samples. After the required time intervals, the digestion flasks were cooled, and the resulting solutions were evaporated to semidried residue to remove excess acid. Approximately 5 ml of 0.1 M nitric acid were added to the residue and filtered through a Whatman no. 42 filter paper and then diluted to 10.0 ml. Blank extractions were carried through the complete procedure. Blanks and standard solutions were prepared in a similar acid matrix. Tables 1, 2 and 3 summarize the experimental conditions. The validity and efficiency of the microwave-assisted digestion method was checked with certified samples of human hair, urine, and whole blood with those obtained from conventional wet-acid digestion method [26]. The concentrations were obtained directly from calibration plots after correction of the signal obtained from blank. All experiments were conducted at room temperature (30–35°C) following well-established laboratory protocols.
6
Kazi et al.
Statistical Evaluations The statistical analyses were performed with the Excel X State computer program (Microsoft, Redmond, WA, USA) and Minitab 13.2 (Minitab, State College, PA, USA). The Student’s t test was used to assess the significance of the differences between concentrations of elements [27]. Calibration curves were constructed from known standards. The least-squares regression slope of the calibration curves represented the sensitivity. The equations used for the calibration curves are: Y ¼ 5:0 104 1:0 105 ðCrÞ þ 2:0 103 6:0 104 Y ¼ 7:0 102 2:0 103 ðCuÞ þ 5:0 103 2:0 104 Y ¼ 5:4 102 8:0 104 ðFeÞ þ 1:0 103 2:0 104 Y ¼ 0:152 0:46 103 ðMnÞ þ 9:0 103 1:0 104 Y ¼ 7 105 6:0 105 ðNiÞ þ 5:0 103 8:0 104 Y ¼ 0:248 4:2 103 ðZnÞ þ 4:2 103 5:0 104 where Y is the integrated absorbance, and Cr, Cu, Fe, Mn, Ni, and Zn are the masses deposited in the furnace (in nanograms and picograms, depending on the element). The linear range of the calibration curve used for all trace metals went from the quantification limit up to 100 μg/l. The limit of detection, was 30 pg/g for Cr, 0.17 pg/g for Cu, 0.45 ng/g for Fe, 0.13 ng/g for Mn, 10 pg/g for Ni, and 0.065 ng/g for Zn, defined as 3 sm−1. The quantification limits was 91.0 pg/g for Cr, 0. 51 pg/g for Cu, 1.35 ng/g for Fe, 0.4 ng/g, 31 pg/g Mn, 3.4 pg/g for Ni and 0.19 ng/g for Zn, defined as 10 sm−1. Analytical results of the certified samples were in agreement with the certified values, confirming the reliability of our methods. The percentage recovery of all elements in CRM samples obtained by conventional digestion varied between 95.2 and 99.3%. The microwave-assisted digestion method was less time-consuming, and it took less than 10 min to complete the digestion of samples and standards. The mean values for all elements differed less than 1–2% from the certified values. The coefficient of variation deviated less than 2% for all elements. Nonsignificant differences were observed when comparing the values obtained by both methods (paired t test, (p>0.05 (Table 5).
Results and Discussion The concentration of elements in the samples studied vary widely among individuals; thus, a significantly large number of samples from population need to be analyzed if the results were treated statistically for meaningful correlation. The mean concentrations with standard deviations for each element in samples are shown in Table 6. The concentrations of Zn in scalp-hair samples of controls were found in the range of 203.6–220.5 and 177.2–197.5 μg/g but in the male patients was in the range 131.9–163.5 and 140.7–169.3 μg/g in the age groups 45–60 and 61–75 years, respectively. The same trend was observed in female subjects. The concentration of Zn in blood samples of controls of both genders was in the range 6.0–12.1 and 6.9–12.3 mg/l, higher as compared to controls, 4.5–8.5 and 4.7–8.9 mg/l, in male and female subjects.
Metals in Diabetes Mellitus
7
Table 5 Determination of Cr, Ni, Fe, Cu, Zn, and Mn in Certified Samples by Conventional Digestion (CDM) and Microwave Digestion (MAD; N=10) Elements
CDM
MAD
Certified sample of whole blood (μg/l) Cr 1.98±0.11 (5.55) 1.94±0.09 (4.64) Cu 13.87±1.02 (7.35) 13.69±0.89 (6.5) Fec 14.4±0.98 (6.80) 14.06±1.02 (7.25) Mn 24.80±1.8 (7.26) 24.67±1.3 (5.27) Ni 7.52±0.62 (8.24) 7.39±0.49 (6.63) 2.25±0.09 (4.00) 2.20±0.08 (3.64) Znc Certified sample of urine (μg/l) Cr 14.57±0.78 (5.35) 14.31±0.93 (6.49) Cu 54.0±0.7 (1.29) 53.27±0.6 (1.13) Fe 35.5±0.6 (1.69) 33.8±0.4 (1.18) Mn 5.56±0.51 (9.17) 5.54±0.47 (8.48) Ni 11.88±0.87 (7.32) 11.72±0.69 (5.89) Zn 205±6.0 (2.93) 201±7.0 (3.48) Certified sample of human scalp hair (μg/g) Cr 90.94±5.99 (6.59) 89.23±6.53 (7.32) Cu 108.4±6.7 (6.1) 105.8±7.2 (7.3) Fe 575.0±22.0 (3.8) 565.8±24.0 (4.2) Mn 11.09±0.85 (7.66) 10.88±0.95 (8.73) Ni 46.07±1.41 (3.06) 45.75±1.38 (3.02) Zn 197.2±12.8 (6.2) 194.5±11.3 (5.7)
T valuea
% Recoveryb
Certified values
0.0039 0.72 0.037 0.0017 0.041 0.091
97.98 98.70 97.64 99.5 98.27 97.77
2.0±0.5 13.9±2.7 14.2±3.24 25±5c 7.5±1.8 2.27±0.68
0.044 0.41 0.997 0.006 0.0173 0.47
98.2 98.65 95.2 98.9 98.6 98.05
14.6±3.0 56±10 39.0±10 5.6±1.5 11.9±3.0 210±50
0.0023 0.00049 0.00112 0.0012 0.9242 0.000034
98.12 97.60 98.40 98.11 99.3 98.63
91.0±10d 110±5e 580.0±10d 11.2±0.3c 46.0±1.4e 199±5
Values in parentheses are RSD a
Paired t test between CDM and MWD; df=9, T (critical) at 95%, CL=2.262
b
Percent recovery, calculated according to: 100×MDM/CD
c
Milligrams per liter
d
Informative value
e
Indicative value
The excretion of Zn was higher in diabetic patients than that of the controls in both genders. It was also observed that the concentration of Zn in scalp-hair and blood samples was different according to gender and disease state of the subjects. Elevated iron was observed in scalp hair of male and female diabetics. There were no differences that could be attributed to age (Table 6). The mean values of iron in blood samples showed no differences but were lower for the older men, 436.9–543.9 mg/l, and women, 424.1–543.8 mg/l, than in their respective controls, 621.2–707.8 mg/g and 597.5–663.9 mg/l. The hair copper levels in controls and patients had no significant differences (p>0.05). The levels for healthy male and female subjects of both age groups were found in the 6.1– 13.3 and 5.65–13.8 μg/g range, while that of diabetics was 6.25–14.5 and 5.65–15.9 μg/g. In blood samples, the range of Cu values in healthy subjects of both age groups were found to be lower (0.9–1.78 and 0.99–1.99 mg/l) than those values obtained for Cu in blood samples of diabetic patients (1.21–2.34 and 1.25–2.66 mg/l) for male and female subjects, respectively (Table 6). Analysis for Cr in hair revealed a significant difference between controls and patients (p=0.001). The concentration of Cr in scalp-hair samples of male and female controls were in the 2.87–3.8 and 3.05–4.2 μg/g range, while the diabetic patients of both age groups were in the 0.96–2.72 and 1.2–2.82 μg/g range for male and female subjects, respectively.
8
Kazi et al.
Table 6 Trace Element Concentrations in Scalp hair, Blood and Urine Samples of Healthy Controls and Diabetes Mellitus Patients Age groups
Male n=217
Chromium scalp hair (μg/g) Normal 46–60
61–75
3.5±0.28 3.16–3.79 (3.41) 3.2±0.31 2.87–3.54 (3.18)
Blood (μg/l) 46–60 60.5±3.2 58.1–63.8 (61.3) 61–75 58.9±2.61 55.2–61.1(58.4) Urine (μg/l) 46–60 8.5±2.46 6.9–12.1 (8.69) 61–75
10.3±2.85 6.8–12.8 (9.85)
Copper Scalp Hair (μg/g) 46–60 11.5±0.9 10.5–13.3 (11.4) 61–75 7.3±1.37 6.1–8.7 (7.5) Blood (mg/l) 46–60 1.4±0.32 0.9–1.78 (1.35) 61–75
1.42±0.28 1.12–1.55 (1.48)
Urine (mg/l) 46–60 0.15±0.023 0.13–0.17 (0.16) 61–75 0.15±0.025 0.128–0.17 (0.14) Iron Scalp hair (μg/g) 46–60 31.9±1.4 30.5–33.3 (32.5)
Female n=206
Diabetic 2.3±0.43 1.92–2.72 (2.35) 1.5±0.51 0.96–2.13 1.47) 49.9±4.6 46.6–50.3 (49.25) 44.6±4.87 39.8–45.2 (44.2) 12.5±3.58 10.3–16.21 (12.68) 13.6±4.98 9.42–18.4 (14.0) 12.7±1.8 10.9–14.5 (12.6) 8.7±2.56 6.25–11.33 (8.5) 1.8±0.59 1.42–2.34 (1.76) 1.68±0.46 1.21–2.09 (1.72) 0.19±0.043 0.159–0.24 (0.18) 0.18±0.045 0.13–0.226 (0.18) 38.5±2.8 35.7–41.3 (38.6)
Normal DF (tExp)a 18.1 3.6±0.45 3.1–4.2 (3.6)
Diabetic 2.4±0.54 1.93–2.82 (2.49)
DF (tExp)a 12.7
14.8
3.5±0.43 3.05–4.06 (3.55)
1.8±0.68 1.2–2.45 (1.87)
16
14.5
59.2±2.94 57.1–63.5 (59.1)
20.7
19.3
54.2±2.78 51.92–57.9 (54.4)
44.2±4.62 37.8–46.6 (44.6) 44.7±4.3 41.9–47.9 (44.03)
7.09
9.2±2.54 7.5–11.4 9.23)
4.26
10.2±1.5 8.7–11.7 (10.2)
4.75
11.9±1.6 10.3–13.8 (11.8)
3.60
7.7±1.8 5.65–10.3 (7.67)
13.2±1.8 10.5–15.9 (13.2) 9.6±3.15 6.9–12.6 9.64)
4.71
1.5±0.38 1.12–1.99 (1.49)
1.98±0.61 1.39–2.66 (1.99)
5.07
3.55
1.43±0.41 0.99–1.84 (1.43)
1.79±0.68 1.25–2.43 (1.75)
3.46
6.49
0.14±0.04 0.12–0.21 (0.14)
3.74
4.33
0.15±0.042 0.92–0.175 (0.15)
0.17±0.03 0.147–0.21 (0.18) 0.19±0.067 0.13–0.234 (0.19)
11.0
29.2±2.4 25.6–30.4 (29.6)
13.2±2.98 10.9–17.52 (12.6) 13.8±3.2 10.9–17.5 (13.8)
40.5±3.8 36.7–44.3 (40.1)
14
7.58
8.01
3.99
4.03
3.84
19.1
Metals in Diabetes Mellitus
9
Table 6 (continued) Age groups 61–75
Male n=217
25.4±2.23 21.9–28.9 (25.9)
Blood (mg/l) 46–60 708.4±51.3 630.0–786.8 (705) 61–75 664.5±33.3 621.2–707.8 (663) Urine(mg/l) 46–60 2.4±0.5 1.5–3.0 (2.40) 61–75
2.6±0.45 1.3–3.2 (2.49)
Manganese Scalp Hair (μg/g) 46–60 3.8±0.63 3.05–4.55 (3.69) 61–75 3.2±0.41 2.6–3.58 (3.23) Blood (μg/l) 46–60 54.7±6.12 48.2–61.9 (54.6) 61–75 51.3±6.05 44.2–57.7 (50.97) Urine (μg/l) 46–60 1.4±0.17 1.25–1.65 (1.34) 61–75 1.5±0.25 1.22–1.78 (1.53) Nickel Scalp Hair (μg/g) 46–60 5.4±0.9 4.5–6.4 (5.4) 61–75 4.8±1.6 2.6–6.5 (4.83) Bood (μg/l) 46–60 2.2±0.56 1.62–2.64 (2.17) 61–75 1.8±0.48 1.3–2.5 (1.74)
Female n=206
28.4±3.98 25.1–31.8 (28.2)
4.88
23.8±3.7 19.8–29.5 (23.5)
27.9±5.9 22.7–33.5 (28.0)
4.47
656.9±72.4 624–726.2 (655) 489.6±59.8 437–543.9 (490)
4.45
702.5±62.2 618–786.1 (702)
4.05
18.9
625.7±35.6 597.5–664 (629)
651.3±69.8 646.6–720.0 (651) 478.9±56.9 24.1–543.8 (484)
5.34
2.30±0.45 1.82–2.77 (2.30)
1.85±0.68 1.43–2.4 (1.83)
4.17
5.43
2.56±0.43 2.14–2.98 (2.59)
1.94±0.69 1.36–2.39 (1.98)
5.79
2.7±0.9 1.82–3.67 (2.75) 2.5±0.7 1.8–3.2 (2.53)
7.69
4.3±1.09 3.05–5.62 (4.4)
4.82
6.37
3.8±0.51 3.2–4.4 (3.78)
3.2±1.3 1.87–4.52 (3.32) 2.8±0.78 2.1–3.5 (2.64)
43.4±8.8 36.6–52.4 (43.6) 41.9±9.9 32.8–51.6 (40.8)
8.1
58.2±7.7 50.5–66.9 (58.3)
6.7
5.95
52.9±6.2 46.2–60.5 (52.49)
47.2±9.5 39.3–55.7 (46.99) 43.3±9.9 33.4–54.5 (43.12)
8.74
1.59±0.57 1.03–2.24 (1.55)
6.12
1.6±0.45 1.15–2.14 (1.67)
6.9±1.4 5.5–7.4 (6.82) 6.35±2.7 4.42–8.5 (6.35)
6.99
5.4±1.15 4.4–6.3 (5.36) 4.85±2.3 1.8–7.7 (4.93)
7.2±1.4 5.9–8.6 (7.23) 7.15±3.8 2.6–9.3 (7.23)
7.39
2.7±0.86 1.94–3.52 (2.61) 2.3±0.82 1.7–3.1 (2.25)
3.77
2.02±0.7 1.35–2.82 (1.98)
2.5±0.6 1.5–3.47 (2.46)
3.8
3.88
1.42±0.5 1.1–1.99 (1.38)
2.1±1.3 1.2–3.56 (2.05)
3.4
1.8±0.7 1.12–2.47 (1.83) 1.9±0.85 1.09–2.82 (1.99)
1.8±0.32 1.45–2.2 (1.91) 1.9±0.41 1.5–2.45 (1.86)
3.64
2.0±0.69 1.31–2.72 (1.90) 2.1±0.71 1.33–2.93 (2.06)
16.6
8.1
6.24
3.41
4.51
3.95
10
Kazi et al.
Table 6 (continued) Age groups
Male n=217
Urine (μg/l) 46–60 7.7±1.9 4.8–9.7 (7.62) 61–75
7.8±0.98 6.5–9.1 (7.76)
Zinc Scalp Hair (μg/g) 46–60 212.1±8.5 203.6–220.5 (212.19) 61–75 189.0±9.4 177.2–197.5 (189.0) Blood (mg/l) 46–60 10.4±1.35 8.9–12.1 (10.23) 61–75 8.9±1.2 6.0–10.9 (8.87) Urine (mg/l) 46–60 0.8±0.3 0.63–1.05 (0.72) 61–75 0.7±0.3 0.51–1.09 (0.70)
Female n=206
10.2±2.8 7.4–13.2 (10.14) 11.7±1.75 10.6–12.58 (11.63) 169.8±15.8 131.9–163.5 (171.5) 155.7±16.2 141–169.3 (157.3)
5.69
7.1±1.42 5.63–8.67 (7.04)
14.4
7.5±2.07 5.3–9.58 (7.46)
18.6
226.8±7.3 216.0–234.7 (224.44) 220.5±6.2 213.9–227.1 (221.4)
13.1
9.7±2.1 7.56–11.82 (9.64) 10.6±3.4 7.2–13.5 (10.53) 198.8±8.7 186.2–203.9 (197.79) 192.3±9.7 181.1–207.7 (191.3)
7.74
5.94
18.3
18.6
6.9±2.06 5.2–8.5 (7.12)
10.99 10.9±1.2 9.67–12.3 (10.81)
7.2±1.9 5.7–8.9 (7.32)
12.5
5.4±2.1 4.5–7.5 (5.39)
10.71 9.2±1.5 6.9–11.2 (9.21)
6.4±2.3 4.7–8.78 (6.63)
7.7
1.4±0.6 1.03–1.92 (1.37) 1.2±0.5 0.82–1.76 (1.1)
7.12
0.8±0.2 0.6–1.1 (0.77)
1.04±0.3 0.78–1.4 (1.06)
5.03
6.31
0.75±0.27 0.55–1.03 (0.71)
1.26±0.53 0.62–1.91 (1.22)
6.68
tcritical at 95% confidence limit for corresponding degree of freedom at 102, 107, 112 are 1.983, 1,981 and 1.982, respectively. DF degree of freedom a
Median and tabulated t value at corresponding degree of freedom
In blood samples, the Cr concentration was in the range of 55.2–63.8 and 51.92–63.5 μg/l in healthy male and female subjects, respectively, significantly higher than the corresponding values of diabetic patients, p=0.005–0.011 and p=0.003–0.012, respectively. Analysis for Mn in scalp hair is in the range of 1.82–3.67 μg/g for both age groups of patients, while that of healthy subjects was higher, 3.05–5.62 μg/g. The same is true for blood Mn in blood, 48.2–61.9 μg/g for healthy men and 50.5–66.9 μg/l for women compared to 36.6–52.4 μg/g and 39.3–55.7 μg/l, respectively, for male and female patients (Table 6). Elevated Ni was observed in scalp hair of male and female diabetics. There were significant differences that could be attributed to age (Table 6). The mean values of Ni in blood showed no differences but were higher for the older men, 1.7–3.1 μg/l, and women, 1.2–3.56 μg/l, than in the corresponding controls, 1.3–2.5 and 1.11–1.99 μg/l. Fig. 1 a Determination of zinc in scalp-hair samples of normal and diabetic subjects of both genders. b b Determination of zinc in human whole-blood samples of normal and diabetic subjects of both genders. c Determination of zinc in urine samples of normal and diabetic subjects of both genders
Metals in Diabetes Mellitus
11
a 260 240 220 200 180 160 140 120
Min-Max 25%-75%
100 MN1
MD1
MN2
MD2
FN1
FD1
FN2
FD2
Median value
b 16 14 12 10 8 6 4 2 Min-Max
0
25%-75% -2
MN1
MD1
MN2
MD2
FN1
FD1
FN2
FD2
Median value
c 3.2
2.6
2.0
1.4
0.8
0.2 Min-Max 25%-75% -0.4 MN1
MD1
MN2
MD2
FN1
FD1
FN2
FD2
Median value
12
Kazi et al.
a 5
4
3
2
1
0 Min-Max 25%-75% -1 MN1
MD1
MN2
MD2
FN1
FD1
FN2
FD2
Median value
b
70 65 60 55 50 45 40 35
Min-Max
30
25%-75% 25 MN1
MD1
MN2
MD2
FN1
FD1
FN2
FD2
Median value
c
28
24
20
16
12
8
4
Min-Max 25%-75%
0 MN1
MD1
MN2
MD2
FN1
FD1
FN2
FD2
Median value
Metals in Diabetes Mellitus
13
R Fig. 2
a Determination of chromium in scalp-hair samples of normal and diabetic subjects of both genders. b Determination of chromium in human blood samples of normal and diabetic subjects of both genders. c Determination of chromium in urine samples of normal and diabetic subjects of both genders
Significant differences were observed (unpaired t test, p
