Age‐ and sex‐specific reference intervals for superoxide dismutase enzyme and several minerals in a healthy adult cohort

Abstract Introduction The aim of this study was to establish RIs for clinically important markers including superoxide dismutase (SOD), serum copper, zinc, calcium, magnesium, and phosphate in a cohort of healthy Iranian adults. Materials A subsample from MASHAD cohort study was used to assess serum SOD, copper, zinc, calcium, magnesium and phosphate. Serum SOD was measured according to its inhibitory potential of pyrogallol oxidation. Micro‐ and macro‐minerals were measured using flame atomic absorption spectrometry and a BT3000 autoanalyzer, respectively. Sex‐ and age‐specific RIs were then calculated based on CLSI Ep28‐A3 guidelines. Results Reference value distributions for studied parameters did not demonstrate any age‐specific differences that were statistically significant. In addition, sex partitioning was not required for all parameters, apart from serum magnesium, which showed a wider range in females (0.81–1.26 mg/dl) compared with males (0.82–1.23 mg/dl). Conclusion The RIs established in this study can be expected to improve mineral assessment and clinical decision‐making in the Iranian adult population.


| INTRODUC TI ON
Accurate disease diagnosis and monitoring relies heavily on the accuracy of laboratory test results as well as the reference intervals (RIs) reported for test interpretation. 1 RIs can be defined as healthassociated benchmarks, commonly calculated as the central 95% of laboratory test values observed each in a healthy reference population. 2-4 While the same RIs are often widely applied across different geographical regions, genetic and environmental differences can influence laboratory test results. 5,6 In addition, age, sex, diet, and lifestyle can influence disease manifestation and also has a major impact on circulating biomarker levels, in both healthy and unhealthy subjects. For instance, serum concentrations of certain fatty acids or plasma phospholipids may reflect medium-term intake of various foods and can related to habitual intake of specific dietary fats or fatty acids in different populations. 2,3,7 Therefore, specific RIs for each population may be required.
The International Federation of Clinical Chemistry (IFCC) has published many reports recommending each laboratory establishes specific RIs based on the age, sex, and ethnicity of its population, or validates a pre-existing RI given common resource limitations. 8 These recommendations particularly relate to minerals which are likely to differ based on the nutritional status of different geographical regions. 9,10 Minerals can be divided into two categories: macrominerals and micro-minerals. 11 Calcium, magnesium, and phosphate are considered macro-minerals which are needed in greater amounts compared with micro-minerals like zinc and copper. 11 Minerals play important roles in both physiological and pathological processes, and evaluation of these factors is important to assess health status.
However, careful interpretation of test results is essential as several factors such as sex, age of population, geographical area, lifestyle, dietary intake, and diet/drinking water can influence serum levels of minerals. 9,10,12 The physiological roles of minerals are diverse. For example, zinc and copper have effects on growth, neuronal development, immunity, bone and blood health, and cell apoptosis, 13,14 while magnesium plays an important role in carbohydrate metabolism, acting as a cofactor in several enzymatic reactions. 15 Calcium and phosphate form the main contents of the bone structure 16 in the form of hydroxyapatite [Ca 10 (PO 4 ) 6 (OH) 2 ] crystals, 17 playing a role in the formation of the bone by their effects on parathyroid hormone (PTH) and 1,25-dihydroxyvitamin D. 18 Finally, copper and zinc are involved in the structure of SOD, an enzyme with the role of reducing oxidative stress by converting the superoxide to oxygen and hydrogen peroxide. 19 Copper and zinc form the part of the active site of this enzyme. 20 Besides the specific roles of each of these minerals, it has been reported that the concentration of calcium, phosphate, magnesium, zinc copper, and SOD can act as biomarkers for several clinical conditions. For example, these minerals have been shown to differ in sickle cell anemic patients when compared with healthy controls, as a result of altered hemoglobin and production of reactive oxygen species (ROS). 21 In addition, it has been established that SOD activity is enhanced in Hashimoto's thyroiditis due to elevated oxidative status. Hashimoto's thyroiditis has also been shown to be associated with hypomagnesaemia, hyperphosphatemia, and hypocalcemia due to the influence of thyroid hormones on the reabsorption of magnesium, phosphate, and calcium. 22 In addition, it has been shown that plasma levels of SOD could be a useful clinical biomarker of oxidative stress in patients with end-stage renal disease 23 and might be considered in the identification of gastric adenocarcinoma from healthy control tissue. 24 The metabolism of calcium, phosphate, and magnesium can also be altered in chronic kidney disease, 25 a disease which cause vascular calcification because of the oxidative stress process. 26 Also, concentration of calcium and phosphate is important for assessing skeletal function. 16 Previous findings from other regions suggest that populationspecific RIs for each of these parameters are needed. 2,27-31 This study aims to determine age-and sex-specific RIs for a specific region (northeastern of Iran) for calcium, phosphate, magnesium, zinc, copper, and SOD in middle-aged population (35-65 years).

| Study population
The MASHAD cohort study was designed in 2010 and is currently ongoing. In this study, 9,704 participants were recruited and provided informed written consent. In the current study, RIs for macroand micro-minerals were determined by using available data from | 3 of 8 GHAZIZADEH Et Al.

| Laboratory evaluation
Blood samples for all subjects were collected in Vacutainer ® tubes after 14-h overnight fast. Following collection, samples were centrifuged at 5,000 g for 15 min at 4°C and then serum samples were frozen at −80°C for future analysis. Flame atomic absorption spectrometry was used to measure serum copper and zinc. 32 The accuracy of this method was measured as 93% ± 4.8% and 95% ± 3.75%.
The accuracy was evaluated based on the zinc and copper standard curves by utilizing zinc and copper standards (Merck and Co. Pharmaceutical Company) through evaluating the confirmed reference material (Merck KGaA 64271 Darmstadt, Germany) comprising known values (1,000 ± 2 mg/L). 32 The activity of SOD was determined as described by Torkanlou. 33 Briefly, a 96-well-plate microassay was utilized according to the inhibitory potential of pyrogallol oxidation, which was adapted for SOD, and the absorption was measured at 405 nm. Calcium and phosphate were measured by using a photometric method and Pars Azmoon kits (Tehran, Iran), on an autoanalyzer BT3000. 34 Specifically, the amount of calcium was measured based on the intensity of the purple color generated by the interaction between calcium and Cresolphthalein Complexone.
Phosphate was measured based on the intensity of the color induced by the interaction between phosphate and ammonium molybdate and sulfuric acid. 34 Magnesium also was assessed by using the xylidyl blue photometric method and the BT3000 autoanalyzer (Biotechnica, Rome, Italy) (Pars Azmoon kit [Tehran, Iran]).

| Statistical analysis
Age-and sex-specific partitioning was assessed visually and confirmed statistically by using the Harris and Boyd method. 35 Extreme outliers were manually removed based on visually inspection. Additional statistical outliers were then identified using Tukey's method, 36 if normally distributed, and the adjusted Tukey method, if skewed. 37 Following outlier removal, the nonparametric rank method was utilized to calculate the RIs, defined as the central 95% (2.5-97.5%), in alignment with CLSI Ep28-A3 guidelines.
Corresponding 90% confidence intervals (CIs) were also calculated for both the upper and lower reference limit intervals (CIs). 3

| RE SULTS
Reference intervals for magnesium, calcium, and phosphate were determined by using available data on 1,000 men (445) and women (433) and for zinc, copper, and SOD in a population sample of 4,040 men (2,112) and women (1,928) aged 35-65 years from the MASHAD cohort study. After applying the exclusion criteria and removing outliers and missing data, the sample size for each biomarker differed (Table 1). Established RIs for each parameter are defined in Table 1 and associated scatterplots are shown in Figure 1, depicting reference value distributions by age and sex.
For all studied parameters, no statistically significant age-specific differences were noted. Thus, only one age partition was required for all parameters (35-65 years). Magnesium was the only parameter that demonstrated statistically significant sex-specific differences, although the observed difference was minimal. Specifically, the established RIs for magnesium were wider in women (0.81-1.26 mg/dl) compared with males (0.82-1.23 mg/dl).

| DISCUSS ION
We have determined age-and sex-specific RIs for five serum minerals and SOD enzyme levels in a healthy adult sample of the Iranian population from Mashhad city. RIs were calculated by applying a statistical methodology that has been previously described by the Canadian Laboratory Initiative on Pediatric Reference Intervals (CALIPER) 2 and recommended by CLSI Ep28-A3 guidelines. 3

| RIs for macro-minerals
In this study, RIs for macro-minerals (ie, calcium, phosphate, and magnesium) were evaluated. Previous studies have recommended age-and sex-specific RIs for macro-minerals and are discussed below.
Previously established RIs for calcium are reported in Table 2 for men and women in Canada, 38 India, 31 and the Caxias do Sul population. 4 As seen, partitioning based on sex was applied in the Canadian and Caxias do Sul studies, though it was not necessary for the Indian population, as was the case in our Iranian population. Statistically significant age differences were also found in calcium concentrations in the Canadian study, likely due to robust sample size. The quantitative reference limits established for calcium across studies were fairly concordant, although an increased lower reference limit was reported in the Canadian and Indian studies when compared with our results.
Reference intervals reported for phosphate are also shown in Table 2 and compared with studies in men and women living in Canada, 38 Botswana, 30 and India. 31

| RIs for micro-minerals
Reference intervals for micro-minerals of zinc and copper were evaluated. Previous RIs established for zinc are shown in Table 2 for men and women of Tehran (Iran), 28 Nigeria, 48 Greek, 49 and Germany. 29 Method of zinc measurement was flame atomic absorption spectrometry in studies from Tehran, 28 Nigeria, 48 and Greece, 49 though the method applied by the German group was electrothermal atomic absorption spectrophotometry. 29

| RI for SOD enzyme
Tamai et al 53 demonstrated that the polymorphism of SOD gene is associated with an increased susceptibility to insulin resistance and type 2 diabetes mellitus. As well, lower SOD activity is considered an independent risk factor for intimal thickening of carotid artery. 54

| CON CLUS ION
The current study established sex-and age-specific RIs for macrominerals (calcium, phosphorus, and magnesium), micro-minerals

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to disclose.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.