Oxidative Stress Indexes for Diagnosis of Health or Disease in Humans

Oxidative stress (OS) is the imbalance between oxidant and antioxidant molecules, in favor of oxidants, that causes aging and disease. Many studies have been published that demonstrate the relationship between OS and human health and disease; however, the following questions arise: (i) how are we sure that the OS is present in a biological process? (ii) Is the OS reported in the different investigations equivalent? (iii) What are the best oxidant and antioxidant markers for OS diagnosis? (iv) Can we establish the types and the intensity of the OS? (v) Does OS index could be useful for research and/or application in clinical medicine? In this regard, several indexes have been proposed to measure OS in humans relative to the state of health and disease, among which the following can be highlighted: Oxidative Stress Index (OSI), Tiol Ratios (-SH/TT, -SS/-SH, and-SS/TT), Glutathione Ratio (GSSG/GSH), Oxidative Stress Score (OSS), and OXY-index. Therefore, the aim of this review is to present the state of the art of knowledge about OS indexes for diagnosis of health or disease in humans. We searched for articles in English or Spanish in the PubMed/MEDLINE and Scopus electronic databases published up until May 2019. The keywords used were “oxidative stress,” “index,” and “oxidative stress index.” It was identified 11479 records in both databases, and 490 articles were analyzed. Our review suggests that all indexes analyzed allow diagnose and differentiate the OS related to human health and disease. Also, the studies on OSI, Oxy-score, and OSS indexes have proven to be reliable, practical, and with clinical utility. However, it is necessary to continue with longitudinal studies, especially assess the usefulness of the indexes in the clinical prognosis, and make comparative studies between the different indexes.


Introduction
Since that Gerschman et al. and Harman in the 50s proposed the Free Radicals Theory to explain the disease and aging, respectively [1,2], this theoretical orientation has had a great impulse, mainly after the description of superoxide dismutase (SOD), an antioxidant enzyme, in 1969 [3]. During these years, many researchers have sought the relation among different pathological events and the aging, with the free radicals (FR), oxidized molecules, and antioxidants, based on Free Radicals Theory.
The term "oxidative stress" (OS) arises in 1985, as a proposal of Helmut Sies that expresses the imbalance between oxidant and antioxidant molecules, in favor of oxidants, that causes aging and disease [4]. This concept has revolutionized the field of knowledge of FR; thus, most of the research about the oxidants and antioxidants is referred as OS. Likewise, the studies of OS mechanisms had described that the reactions involved in this process are oxidation-reduction reactions (redox) [4,5].
In parallel, the knowledge about the homeostatic function of the FR, oxidants, and antioxidants has been emerging, highlighting its importance in the signaling and maintenance of cellular mechanisms, because FR and oxidants (called reactive species) are essential components in several processes such as phagocytosis and maintenance of the cell membrane, and physiological function such as immunological mechanisms and vascular function, among others [6,7]. In this sense, actually it is recognized that OS may be separated in oxidative eustress and oxidative distress. In the oxidative eustress, the physiological mechanisms of the reactive species (RS) intervene, mainly by H 2 O 2 reactions, and in the oxidative distress, there are the oxidative damage to biomolecules with disrupted redox signaling that may be causing disease and aging [4,8]. Although these terms are little used, most researchers use the term OS when they refer at the imbalance between oxidant and antioxidant components.
The central problem for the application of OS in biology and clinical medicine is its measurement, considering the complexity of the process and the number of components oxidants and antioxidants that make it up. In this sense, the OS has been measured using different markers, both biomolecules oxidized (lipids, proteins, and DNA) and antioxidants (enzymes and nonenzymatic antioxidants), besides of oxidants (reactive oxygen species), everything in biological samples (Table 1) [4,[9][10][11][12].
Another problem is the variety of methods used to measure the markers in biological samples due to the difference among them. Those methods have different sensibility and reliability and advantages and disadvantages, which hindering its interpretation, and some require very sophisticated techniques [9][10][11]13], which have been widely discussed by Forman et al. and Marrocco et al. [9,13].
On the other hand, the measurements of OS for diagnosis of health or disease in humans must be with noninvasive procedures, so that the employment of blood, urine, or other biological fluids is the commonest.
In this context, the following questions arise: (i) who are we sure that the OS is present in a biological process? (ii) Is the OS reported in the different investigations equivalent? (iii) What are the best oxidant and antioxidant markers for OS diagnosis? (iv) Can we establish the types and the intensity of the OS? (v) Does OS index could be useful for research and/or application in clinical medicine?
Thus, starting from the theory of OS proposed by Sies et al. [4], if exposure to exogenous or endogenous oxidants increases or is insufficiently balanced by antioxidants, oxidative damage occurs to biomolecules, which gives the option to select different oxidants, oxidized molecules, and antioxidants to be measured in a research, making the comparability of the results difficult and generating confusion, perhaps because with individual markers, it is not considered that OS is an integral and dynamic process.
With this view, several proposals have been published to evaluate the oxidant/oxidized components and the antioxidants integrally considering a global index to avoid the bias of each marker measurement, which are intended to demonstrate their ability to differentiate the health of the disease, as well the changes produced by the aging process, although the clinical value of the different proposals has not been systematized and/or analyzed.
Therefore, the aim of this review of the literature is to present the state of the art of knowledge about OS indexes for diagnosis of health or disease in humans.

Data Sources and Search Strategy.
We searched for articles in English or Spanish in the PubMed/MEDLINE and Table 1: Biomarkers used in the oxidative stress measurement in human fluids. Scopus electronic databases published up until May 2019, which included the measurement of oxidative stress using oxidative/antioxidant ratios or indexes. The keywords used were "oxidative stress," "index," and "oxidative stress index," and the specific name of the index was found such as "OSI," "glutathione ratio," "thiol ratio," "oxidative index," "oxidative balance score," "OBS," "OXY score," and "oxidative stress score." References in each article were searched to identify missed studies.

Eligibility Criteria and Study Selection.
We included the studies that involved human subjects, with experimental or observational design (cross-sectional or longitudinal), assessed any disease or health state as primary event, and used biological samples, no tissues, any age, and sex. We excluded animal and cell culture studies, conference abstracts without full text, case reports and case series, reviews of any kind, editorials, and opinion articles. Also, we excluded the studies that measured oxidants and antioxidants, but the authors did not relate them, that is, they did not calculate a sum or ratio between them.

Analysis of the Information.
After removing the duplicate records, we screened the relevant studies from the references retrieved from the databases, reviewing the titles and abstracts. We obtained the full texts of the relevant studies to assess for inclusion or exclusion in this study. The author name, publication year, study design, country, population included (age group, sex), sample size (subjects included in the study), biological sample, main outcome, results, and conclusion related to the ratio or index were extracted from each included study.
We found a great number of studies that use the oxidative stress index (OSI) and thiol ratios, so we analyzed this information in a different way to the rest of the ratios and indexes; in fact, the reference list of those indexes is in supplement files (available here).

Results and Discussion
It was identified 11479 records, in both databases, with the selected keywords. After removing the duplicate records and that did not meet the inclusion criteria, 501 full-text articles plus 14 records identified searching on articles references were eligible. After reviewing the full-text articles, 25 were excluded by different reasons. A total of 490 articles were analyzed ( Figure 1).
The information is ordered according to the name of the index, trying to maintain a chronological order and with an increase in the complexity of the calculation, except the first section because we included a series of different calculations that have in common the use of a ratio and they do not have a specific name.
3.1. Oxidative Balance. One simple form to measure the oxidative/antioxidant balance is calculating a ratio. The first proposal to use a ratio to measure oxidative balance was the oxidized/reduced glutathione (GSSG/GSH). Initial studies in animal models indicated that its increase is an index of OS in vivo and suggests that it can be used as a complement of other parameters [14]. Thus, this ratio has been used in many studies in humans over the past 10 years, so we analyzed its importance in a separate section.
The first reports of an attempt to find the relationship between the oxidant and antioxidant counterparts of a biological sample, and not the evaluation of individual markers, are two studies conducted on seminal samples. In these studies, reactive oxygen species (ROS) and total antioxidant capacity were measured, both by chemiluminescence, and a score called ROS-TAC score was calculated. The computation of the ROS-TAC score is complex, with mathematical transformations, standardizations, and a principal component analysis as part of the computation [15,16], which makes it an impractical score; thus, there is neither study with it.
To avoid complex calculations that are not easy to apply to the clinical setting, the researchers resorted to obtaining ratios between an oxidized molecule and an antioxidant. In this sense, the most used oxidation markers have been lipoperoxide level, measured as malondialdehyde (MDA or TBARS) or total lipid hydroperoxides (LOOH), but for the measurement of the antioxidants, a variety of parameters have been used, such as antioxidant enzymes (superoxide dismutase [SOD] [17][18][19][20][21][22][23][24]. The most commonly used biological samples are the components of the blood, although these reasons have also been applied in measurements obtained from saliva and follicular fluid (Table 2).
Regarding the name used to refer the ratio, some authors called it as "oxidative stress index" (OSI) [22,23], although this term is most used for other relation that we discuss later. Becatti et al. called their ratio as "redox index," and they define it as a reliable estimation of redox status in humans [24], but neither is it a good option because they measure the ORAC/MDA ratio which is not necessarily a measure of redox status. Therefore, we think it is better to express only the calculated ratio to refer it.
The ratio can be calculated from the oxidized molecule to antioxidant, since the shift of the balance toward the oxidative side is considered to represent OS. The lipids oxidized, measured as MDA, or proteins oxidized, measured as advanced oxidation protein products (AOPP), are the main used oxidized molecules, related to total antioxidant status (TAS) or TRAP as antioxidant measure; thus, the MDA/TAS and AOPP/TRAP ratios are recommended indicators of the OS [17,21]. The use of the ratio toward the antioxidant side indicates only the antioxidant-oxidant imbalance [20,24].
Another way to get the oxidizing/antioxidant ratio is to measure the oxidants like superoxide anion (O 2 −⋅ ) and nitric oxide (NO), and glutathione (GSH) or melatonin as antioxidants, although their use should not be frequent, because it exists only one report about melatonin in which it was calculated the NO/melatonin and MDA/melatonin ratios; the authors suggest that these ratios reflect an oxidantantioxidant balance and could be used to determine OS homeostasis [25], but we should consider that within the OS homeostasis, melatonin is not one of the essential markers. On the other hand, if we calculate the ratio with the oxidants ) and GSH as the antioxidant, a problem may arise because the measurement of the oxidants is not easy to apply at patient samples; thus, there remains doubt about its possible usefulness [26].
3.2. Glutathione Ratio. As we noted above, the GSSG/GSH ratio is the first attempt to assess the OS as a dynamic way, since this coupler represents the oxidant/antioxidant homeostasis. After a series of animal studies, Curello et al. in 1987 improved the assay of total and oxidized glutathione in blood samples, which has been useful in monitoring of clinical status [27]. From this, different procedures have been developed with various technologies to carry out the measurement not only in blood but also in other biological samples and tissues. In this sense, pitfalls in the measurement of both reduced and oxidized glutathione with those technologies have recently been discussed, pointed an alert in the interpretation of the results [28]; however, glutathione ratio is an indicator of OS widely used.
The ratio can be analyzed from GSH to GSSG or inversely GSSG to GSH; in either direction, it is considered as a marker of intracellular OS. If the ratio is obtained from reduced to oxidized glutathione, low values indicate OS; in the opposite direction, a high value denotes OS. Many studies have been realized using the ratio in both ways; thus, we revised them by separation. It is important to keep in mind that most of the time, the glutathione ratio is used as another marker of the OS, as was suggested from the beginning of its use [14], not as an individual indicator, like the ratios pointed above.
Furthermore, it is possible to express the glutathione ratio as redox potential (E h GSH/GSSG), which is a way to 1147     represent the redox environment in biological fluids [29]. The redox potential is calculated by the Nernst equation as follows [30]: where E 0 = −264 mV (GSH/GSSG couple standard potential for pH 7.4), R is the gas constant (8.314 J/°Kmol), T is the absolute temperature (°K), F is the Faraday constant (9:6485 × 10 4 C/mol), and GSH and GSSG are the concentrations in moles/liter. In the interpretation, higher E h values indicate a greater potential for oxidation or OS.
The first studies in humans that use this ratio have begun in the 90s. In this review, we found three reports in this decade. Two who obtained the GSSG/GSH ratio of measurements made in the blood of newborns and whose results suggest its use as a noninvasive method to evaluate the OS [31,32]. The other research is an experimental study with free fatty acid (FFA) infusions in which the lipoperoxide level was measured through the MDA and LOOH, in addition to the GSH/GSSG ratio. In this study, the role of the ratio in the effect sought is not directly concluded, but an inference is made when it considers that the OS intensifies with the increase of FFA [33].
Subsequently, in this century, an avalanche of information has emerged. Twenty-four articles located inform the use of the GSH/GSSG ratio to evaluate the antioxidant/oxidant balance, which is on the Table 3  . The sample mainly used is whole blood and its derivatives (erythrocytes, polymorphonuclear cells, or plasma), and other markers are added, such as oxidation products, oxidants, and enzymatic and nonenzymatic antioxidants, to obtain a more comprehensive measurement of the OS, although it is possible to use exclusively the glutathione ratio, but remember that this ratio measures only the intracellular OS [35,36,49,51,55]. Regarding to the GSSG/GSH ratio, fifteen studies were localized, and blood is the biological sample used mainly, but also it is possible to measure glutathione in saliva [56,57]. As in the previous ratio, it is common to add more OS markers [30,31,[56][57][58][59][60][61][62][63][64][65][66][67][68], and the measurement of the ratio from oxidized to reduced form is a good marker of oxidant/antioxidant imbalance in different clinical situations ( Table 4).
The redox potential (E h ) is a complement of the GSH/GSSG ratio to evaluate the redox status [30,34,37,43,49,51,55], but it can be used as an independent measurement since it provides indirect information about the redox state of all cell types, although the measurement has been performed only in erythrocytes, because the cells have similar essential functions [69].
Nevertheless, the glutathione ratio seems to be a promise OS marker, it is limited because it only measures intracellular environment without considering other OS components as antioxidant enzymes and oxidized biomolecules.
3.3. Thiol Ratios. The thiols are compounds with high vulnerability to the oxidation by reactive species, mainly of oxygen, due to their -SH group, which is oxidized to their disulfide (-SS-). As this is an oxidation reaction, the imbalance between both forms in favor of the oxidized molecule is considered as OS, being a dynamic homeostasis involved in many diseases and alterations of the physiological state [70,71]. Many proteins, nonproteins, and low molecular mass thiols have -SH group, but only cysteine and glutathione are considerate in the redox balance, where cysteine represents the extracellular environment and glutathione the intracellular. Then, the term "thiol ratio" is used to refer the measurement of these aminothiols together, both in its oxidized and reduced forms. There are some techniques to their measurement, less complex and automated what has facilitated its use in different investigations.
After the proposal of an automated procedure to evaluate native thiol (-SH), total thiol (TT), and thiol-disulfide (-SS) [70], -SH/TT, -SS/-SH, and -SS/TT ratios have been calculated; where if the value of -SH/TT ratio is low and both ratios of -SS are high, the OS is present. Also, it is possible to present the result as a percentage, multiplying the total obtained by 100.
Its use has spread widely to become the second most popular indicator of OS so far. In this review, we found sixty-six studies that refer them. All the studies, except one [72], were carried out in Turkey in a variety of age groups and clinical situations since 2015. The thiols can be measured in serum/plasma, but semen sample [72] and aqueous humor [73] also have been used, and they are applicable to evaluate the oxidative state in a variety of clinical events, from alterations during pregnancy, cardiovascular, psychiatric, neurological, hereditary, infectious diseases, etc. Although the information is abundant, the analysis result has not shown substantial differences among the studies, because all agree in recommending the thiol ratios to their use in clinical events; therefore, we only did a descriptive analysis of the information (Table 5) and presented references found in a supplementary file in case of any reader is interested in consulting the information.
It is important to note that, contrary to glutathione ratio, the thiol ratios generally do not need other markers to interpret the OS, although antioxidant enzymes, FRAP, TAC, total oxidant capacity (TOC), and ischemia-modified albumin (IMA), have been some of the other markers included to complement the interpretation.
The thiol ratios are the OS markers used more recently, but like the glutathione ratio, to make a complete interpretation of the OS, it is necessary to include antioxidant enzymes and oxidized biomolecules.

Oxidative Stress Index (OSI).
All the previous proposals are recognizing that OS is a dynamic process, because they consider both the oxidizing and antioxidant components in a single calculation; however, only one type of oxidant and antioxidant is related, which is a disadvantage because it is necessary to evaluate many components to establish more effective if the OS is present. Hence, other proposals have emerged that consider the measurement of oxidant and antioxidant activity whose effect is produced by the action of several components of OS measured as one.
Those are based in fact that the measurement of different oxidants and antioxidant molecules separately is not Behçet's disease (BD) The GSH/GSSG ratio was reduced in BD patients. Exist an OS in BD as shown by the diminution of the ratio.      practical because they are costly and time-consuming, and frequently, the procedures are complex; moreover, their oxidant or antioxidant effects are additive [74,75]; therefore, it is possible to measure the total oxidant or antioxidant capacities of a biological sample and related them in a ratio, called index. The antioxidant capacity is measured by different procedures identified by their acronyms: (i) Ferric reducing antioxidant potential (FRAP) or total antioxidant potential (TAOP): measurement of the ability of the sample to reduce Fe 3+ to Fe 2+ as a result of antioxidant activity. The formation of a complex between Fe 2+ and 2,4,6-tripyridyl-1,3,5triazine (TPTZ) is measured [20,76] (ii) Oxygen radical absorbance capacity (ORAC): inhibition measurement of the peroxyl radical-induced oxidation initiated by thermal decomposition of 2,20-azobis(2-amidinopropane) dihydrochloride (AAPH) [24] (iii) Total antioxidant activity (TAA): it is the assessment of the ability of a sample to inhibit linolenic acid peroxidation. The results are expressed as percentage of inhibition of linolenic acid peroxidation produced by samples [58] (iv) Total antioxidant capacity (TAC): the method is based on the hydroxyl radical (OH ⋅ ) produced via the Fenton reaction using a mixing ferrous ion solution and H 2 O 2 . The OH ⋅ generated reacts with o-dianisidine molecule to form dianisidyl radicals [75] (v) Total antioxidant status (TAS): the method is based on the production of the radical cation (ABTS ⋅+ ) by the incubation of ABTS (2,20-azino-bis[3-ethylbenzothiazoline-6-sulphonic acid]) with peroxidase (metmyoglobin) and H 2 O 2 [17] (vi) Total radical-trapping antioxidant parameter (TRAP): measurement of hydrosoluble and/or liposoluble plasma antioxidants using chemiluminescence inhibition time induced by 2,2-azobis(2amidinopropane) [21] Except in TAA, the results in all the quoted methods are expressed as Trolox equivalents (μmol/L), which is the antioxidant standard.
Likewise, the measurement of the oxidant capacity can be carried out by different methods: (i) Total oxidative capacity (TOC): assay based on the oxidation of 3,5,3 ′ ,5 ′ -tetramethylbenzidine (TMB) to colored radical cations in the presence of peroxides by peroxidase. It used to determine total peroxides in μmol/L [77,78] (ii) Total oxidant status (TOS): it is the measurement of the oxidation of Fe 2+ to Fe 3+ by peroxides of the sample to produce a Fe 3+ -xylenol orange complex. This assay quantifies total peroxides; the results are expressed as H 2 O 2 μmol Eq/L [75] Some indexes are in the literature, but the most used is called "oxidative stress index" (OSI) initially proposed in 2003 [18]. This index has had few modifications across the time, but all the studies agree in that it is a ratio between TOS and TAS expressed in arbitrary units: where OSI is in arbitrary units (AU), TOS is in μmol H 2 O 2 Eq/L, and TAS is in μmol Trolox Eq/L. Sometimes, the result is multiplied by 100 to represent the percentage ratio. TAC or TAOP can also be used as antioxidant component because being in the same units (μmol Trolox Eq/L), and TOS may be replaced by TOC.
As we pointed above, OSI is the index most frequently reported in the literature. In this review, we found 302 studies in humans since 2003 that use this index, the majority carried out in Turkey, mainly in adult events such as cardiovascular diseases, alterations in the pregnancy, and psychiatric disorders. As the thiol ratio information, after analyzing the studies, we observe that most research recommends the use of OSI without other important contributions to this review; thus, we present only a descriptive analysis and the references in the supplementary file. However, we want to note that the measurement of oxidant and antioxidant activities for the  [96,97] that have also been used in the measurement of thiols (Table 6). Like the other calculations, OSI also has been accompanied of several OS markers such as MDA, DNA damage, hydroperoxides, protein carbonyls, IMA, thiols, antioxidant enzymes, prolidase, paraoxonase, and antioxidant vitamins, many times in order to show the relationship between the different markers and prove their usefulness as an independent indicator of OS. All this information makes this index a very promising way to measure the OS; however, there are still doubts regarding its interpretation. Actually, for its interpretation, the increase in the value of the index indicates a greater intensity of OS, but it has not been clearly established which is the cut-off value for the OS, necessary value to better clinical application, because in clinical practice, it is important to be able to differentiate oxidative eustress and oxidative distress, as we have pointed early in this review.

Oxidative INDEX. This is a proposal of Vassalle et al.
(2008) that also related the serum oxidative/antioxidative capacities; only that for this index, the oxidative part is the measurement of total hydroperoxides by dROMs and to the serum antioxidant activity has used a procedure called OXY Adsorbent Test [98]; both methods are automated and commercial.
The dROMs or derivatives of Reactive Oxygen Metabolites, also called diacron ROM, are based on the measurement of sample alkoxyl (RO ⋅ ) and peroxyl (ROO ⋅ ) radicals (sample peroxides) formed by iron, according to Fenton's reaction. These radicals are able to oxidize an alkyl-substituted aromatic amine (A-NH 2 ), such as N,N-diethyl-para-phenyle-

nediamine (DEPPD), transforming it into a derivative ([A-NH 2
⋅ ] + ). With some modifications in the procedure, this method also is called as free oxygen radical test (FORT), which reflects levels of organic hydroperoxides. Results are expressed in arbitrary units, called Carratelli units (Carr U), where one Carr U corresponds to 0.08 mg/dL H 2 O 2 . Values greater than 300 Carr U suggest OS [99][100][101].
The OXY Adsorbent Test evaluates the antioxidant ability of each sample to oppose the oxidant action of hypochlorous acid (HClO) added in excess. The results are expressed as μmol HClO/mL, and low values indicate a reduced antioxidant capacity [102].
Two aspects that we should highlight about this index is that it has only been used in serum as a biological sample and its calculation method, which makes its daily use difficult.
In this sense, due to dROMs and OXY are in different measurement units, it is necessary to have a standardization according to the following formula [101,102]: where sv var is the standard value of each measurement which is equivalent to z, v var is the original value of each subject, x var is the mean of the parameter, and s var is the standard deviation of the parameter. After this calculation, a simple subtraction between standardized dROM and standardized OXY must be calculated: This index has proven its usefulness in comparisons between young vs. old people [103], men vs. women [104], and in several clinical settings, such as cardiovascular diseases [  disorders [108], hepatic diseases [109,112], and cancer [110], showing that it has the capacity to discriminate between the groups (Table 7). It is important to note that since this index evaluates the oxidative balance, no other OS marker was included in any study. In the interpretation, being a standardized value, there can be negative and positive values in a range of -3 to +3, with 0 being the ideal balance between oxidants and antioxidants; therefore, positive values indicate predominant oxidized damage and negative values an imbalance towards antioxidants. Moreover, it has been shown that if the 75th percentile of the data is calculated, a cut-off value can be obtained to establish the intensity of the OS, at least by deciding whether it is low or high [102,106], although further studies need be carried out to confirm their usefulness as a diagnostic tool [108].
Another index with total hydroperoxides measured by dROMs method has been described. For this index, the measurement of the biological antioxidant potential (BAP) is used as the antioxidant component.
In the BAP method, the ability of plasma antioxidant components to give reducing equivalents to reactive species is determinate using Fe 3+ that are reduced to Fe 2+ . The values are expressed in terms of iron-reducing activity by vitamin C as standard, considering as cut-off values: >2200 μmol/L as optimum status, <1600 μmol/L as deficient status, and <1400 μmol/L as high deficiency status [99,113].
An arbitrary index is calculated as a ratio. This index has been used in two directions (i) Oxidant to antioxidant (dROMs/BAP) ratio, if it considers that the balance is disrupted by excessive production of dROMs or by low BAP (ii) Antioxidant to oxidant (BAP/dROMs) ratio, when it wants to measure the relative antioxidant capacity In addition, two modifications to the BAP/dROMs ratio have emerged. First a proposal of Yamamoto et al. (2015) called mOA that is an adjustment of the ratio by the constant 7.541 [114]: In the other modifications, each BAP/dROMs value is divided by mean BAP/dROMs ratio; in its interpretation, a cut-off value <1.0 indicates an antioxidant potential decreased [115]. This is the only study that establishes a cut-off value; others use the ratios as quantitative values. Both indexes aim to show more clearly the antioxidant potential.
These indexes have been used in research mainly in Japan; seven studies calculated the ratio in the oxidant direction to the antioxidant [99,[116][117][118][119][120][121], and six in the reverse direction [113,[122][123][124], including the investigations that calculated the modified ratios [106,107] (Table 8). Only in one report the follicular fluid was used as a biological sample [116]; in the others, serum or plasma was used, and neither markers of the OS were included in the studies.
Also, it draws attention that some researchers call the dROMs/BAP ratio "oxidative stress index" (OSI) [99,116,[119][120][121] as the first index that we included in this review, which is logical because this index is an oxidant/antioxidant ratio; the difference is the form of evaluation of both components. In both OSI, the oxidative part is the measurement of total peroxides or hydroperoxides, which are the oxidation products of several molecules, such as lipids, proteins, and amino acids; therefore, this is a measure of oxidative damage, what does not happen with the antioxidant capacity. The TAC and BAP have different principles, as well as their measurement units; both determinate the total nonenzymatic antioxidant capacity of biological samples, but although BAP is the most recommended procedures [125], the results of the different studies shown that this parameter almost does not change, making the index show no difference [114,115,122], especially when it is used to see the antioxidant potential (BAP/dROMs).
In summary, seems that the Oxidative INDEX can reflex the intensity of the OS better than the dROMs/BAP ratios, but these indexes do not include antioxidant enzymes, important elements in oxidative/antioxidative balance.
3.6. Oxidative Stress Profile. As we explained above, all the ratios and indexes proposal only measure extracellular OS components or oxidative/antioxidative capacities of the biological samples, so missing other elements that also participate in the oxidative balance.
Thus, in another attempt to measure oxidative/antioxidant balance, Cutler et al. in 2005 [126, 127] proposed the so-called "oxidative stress profile," where it considers a lot of markers of oxidative damage, such as MDA, 4-HNE, hydroperoxides, isoprostanes, oxidized nucleic acids, protein carbonyls, antioxidants (endogenous and exogenous), and plus inflammation markers, such as interleukins, measured in both blood and urine samples. In total, these were suggested more than 50 elements in a proposal to assess the oxidative/antioxidative balance [127].
In the interpretation of the results, cut-off values are established for optimum health and intervals outside the reference ones, making a balance between the oxidized components through an oxidation index and an antioxidant index, obtaining an average level of antioxidant protection and other oxidative damages summarized in four categories: (1) high protection-high damage, (2) high protection-low damage, (3) low protection-low damage, and (4) low protectionhigh damage [126,127]. The limitations of this proposal are the high cost due to the large number of parameters to be measured; many of which require a measurement with sophisticated equipment and techniques and also the need to establish the cut-off values to calculate the indexes, mainly of the exogenous antioxidant markers, which are dependent on the diet and lifestyle, so they cannot be generalized in the populations, and finally, the complexity in interpreting the results.  The oxidative/antioxidative balance was exacerbated in patients with potential POPH.  The ROM/BAP ratio was higher in the IMD group.
Patients who receive livingrelated liver transplantation due to IMD are prone to higher ROM/BAP ratio.  computation of different parameters of the oxidative balance. This score includes individual measurements in plasma of free and total MDA (F-MDA, T-MDA) and the GSSG and GSH levels and moreover the urinary isoprostane-PF2α-III level (iPF2α) as the combined oxidative damage. The antioxidant counterpart includes GSH, αand γ-tocopherol, and antioxidant capacity measured by the OXY Adsorbent Test, which the authors called individual antioxidant capacity [128]. For calculation, first the F-MDA, T-MDA, iPF2α, and GSSG levels are log-transformed, then each marker must be standardized according to the formula previously describe, using the raw or log-transformed value as correspond. Later, the average of the standardized oxidants (damage score) and the average of the standardized antioxidants (protection score) must be calculated. Finally, the subtraction of a damage score minus protection score is the OXY score.
Same as for Oxidative INDEX, if the result is near to zero, the levels of all markers are near the average normal values or there is a balance between oxidants and antioxidants, and positive values indicate an imbalance towards the oxidants [128,129]. It has been suggested to compute the OXY score without the urinary iPF2α as a modified OXY score, with similar results [130].
Although the ability of the OXY score to discriminate between different clinical conditions, age, and sex has been reported [127,128], it is necessary to recognize that antioxidant enzymes are not included, being a score limiting [130]. Some drawbacks of the OXY score are the difficulty in calculating it, and the methods of measurement of all the markers are included, because they are complex and expensive; thus, it is not easily applicable to clinical and epidemiological settings.
Another modification to the OXY score is reported, which is a global score for the OS associated with albuminuria to their use in chronic kidney disease patients (Albumin OXY score). This index included protein carbonyl and MDA to calculate the damage score, and superoxide (O 2 −⋅ ) scavenging activity, catalase (CAT), and TAC as the protection index. For the calculating, the authors followed the procedures indicated for the Oxidative INDEX and the OXY score and proposed a new calculation that includes logarithmic transformation, standardization, and final subtraction, integrating both proposals. Being standardized values, the interpretation is equal to the two indexes previously mentioned [131]. The main clinical disadvantages of this index are the complication in the calculation, the cost, and the complexity of some measurements, although the total number of markers to be measured was reduced.
3.8. Oxidative Stress Score. All the indexes mentioned until now, except that proposed by Cutler and the Albumin OXY score, include an oxidant part and the non-enzymatic antioxidant capacity and do not take into account the antioxidant enzymes as part of the calculation. Thus, it seems necessary to include at least the main enzymes in the cell antioxidant process: superoxide dismutase (SOD), glutathione peroxidase (GPx), and/or catalase (CAT).
In this sense, Amstad et al. in 1991 [132] showed in transformed cell lines of mice that the balance between the activities of SOD and CAT+GPx is more important to determine the oxidative effect than the absolute activity of a single enzyme. Following this proposal, it has developed a theoretical model of biochemical interaction between these three enzymes showing an additive effect between SOD and GPx, and synergistic effects between GPx and CAT to ensure a global cell protection [133,134]. As we know, the oxygen in aerobic organisms is part of various biochemical reactions, such as the mitochondrial respiratory chain and phagocytosis. One characteristic of these biochemical processes is the transformation of molecular oxygen into O 2 −⋅ , which is dismuted by SOD, forming H 2 O 2 in a first step. Although it is currently recognized that H 2 O 2 is an important molecule in redox regulation and signaling processes, it is also known that it is a molecule harmful for its oxidizing capacity; thus, living organisms have a series of enzymes H 2 O 2 -removing to maintain balance, being glutathione peroxidases as the main second step [8,135]. An imbalance between the first and second steps has as the potential result of increasing H 2 O 2 levels, which is possible because the reaction constant of SOD is higher than GPx, so H 2 O 2 is formed quickly and is not eliminated in its entirety; in addition, this disproportion in the activity between SOD and GPx in favor of the first can produce an increase in lipoperoxidation [136]; therefore, the inclusion of these enzymes as part of an integral OS measurement is important.
Another point to consider in the assessment of OS is to establish which cut-off values of the oxidative/antioxidative imbalance can be considered as harmful, to refer only to the oxidative distress. In this sense, in 2004, we proposed the integral and dynamic measurement of the OS through the evaluation of the antioxidant system efficiency, establishing the categories: (a) efficient antioxidant system (EAS) when the harmful action of the oxidants is effectively counteracted, (b) antioxidant enzymatic deficiency (ANED) if there is an inefficient or insufficient action of antioxidant enzymes, (c) exogenous antioxidants deficiency (EXAD) when there is an insufficient or inefficient action of nonenzymatic antioxidants causing oxidized the molecules, and (d) antioxidant system global deficiency (ASGD) if the enzymatic and nonenzymatic antioxidant components show an imbalance in favor of oxidized molecules [137,138]. Later, we reorganized the interpretation considering the intensity of the OS, proposing four categories: without OS (WOS), low OS (LOS), moderate OS (MOS), and severe OS (SOS). About this, Lushchak in 2014 [139] also expressed the need to classify the OS based on the intensity, suggesting three levels: low, intermediate, and high, where the high can be interpreted as severe OS, very similar to our proposal.
In order to reach our classification, an index was calculated which we call the oxidative stress score (SS) [140]. The SS includes, originally, the measurement of plasma MDA and DNA damage, the activities of the SOD and GPx erythrocyte enzymes, the total antioxidant plasma status (TAS), and the calculation of the SOD/GPx ratio as a marker of oxidation, and the antioxidant gap (GAP). In this sense, GAP represents the antioxidant capacity of other nonmeasured plasma components different to albumin and uric acid, such as antioxidant vitamins, bilirubin, and hormones, can be calculated from TAS, and albumin and uric acid levels multiplied by their value of Trolox equivalent antioxidant capacity (TEAC) using the formula [10]: where TAS, albumin, and uric acid are in μmol/L, TEAC for albumin is 0.69, and TEAC for uric acid is 1.00. Low-molecular-weight antioxidants such as GSH and thiols are not included in the score because both are measured in TAS, and CAT neither is considered due to its role as H 2 O 2 removal is relative in humans, because it is a peroxisomal enzyme [8].
It was defined the cut-off values of each parameter based on the 90th percentile of healthy young subjects (20-45 y and both sexes), and a score 1 was given to each value above of cut-off (MDA, DNA damage, and SOD/GPx ratio) or under the cut-off (enzymatic and non-enzymatic markers), and 0 if the values are in normal range. After, a sum of all the components was obtained as a SS ranging from 0 to 7, representing the severity of the marker modifications. Then, the SS was categorized as follows [140]: Also, we have excluded DNA damage because their procedure is not easily accessible; in this case, SS is ranging from 0 to 6 [141]. This index can be dichotomous, if SS ≤ 3, we consider as without OS or eustress, and if SS > 3, the subjects have high OS or distress [142].
The SS has been used in several studies since 2005. It has probed in young, mid-aged, and older subjects, both sexes and different study designs (Table 9). This index has shown to be useful to differentiate rural/urban residents [140] and pre/postmenopausal women [141,146]. Also, it has been applied in cognitive impairment [138], chronicdegenerative diseases such as diabetes mellitus and arterial hypertension [142,145], osteoporosis [143], and metabolic syndrome [144]. In addition, it has been used in clinical experimental studies demonstrating therapeutic [148] and of nutritional supplements effects [141,151], as well as the efficiency of controlled physical exercise [147,150].
Other authors have suggested similar proposals to the SS. Initially, Goodman et al. in 2007 [152] developed an OSS with the same idea of us in the form to compute it, but they included prooxidant and antioxidant factors, mainly from diet and lifestyle, to reflect a prooxidant/antioxidant exposure balance. This score is different to our proposal, because they do not perform measurements, rather they use questionnaires to obtain the information and based on them the calculation is made. As the main intention of this index is the exposure balance, the name has been changed to oxidative balance score (OBS). The OBS is a very different index to those included in this review, so we will not discuss it, but it is important to note that it has been used mainly in cases of cancer [141,[153][154][155] and chronic diseases [156][157][158], besides it has been related to OS markers [159,160], inflammation markers [161,162], then too it carried out studies from its relationship with some gene expression and genetic polymorphism [163][164][165][166].
On the other hand, based on the Goodman proposal [152], a study was carried out to determinate the association between adiposity measures and an OSS. The OSS was obtained from the blood measurements of GSH, GPx, vitamin C, MDA, and TAC, and some factors of lifestyle such as smoking, use of anti-inflammatory medication, nutritional supplement, and herbal product, adding a total of 9 components [167]. All the variables were categorized as 0 or 1, as we previously pointed. Although the idea is like to our SS, this index is more complex and included particular parameters such as the consumption anti-inflammatory medication and herbal product that are not necessarily useful in all the researchers.
Another study used six blood oxidative stress markers: MDA, ox-LDL, total antioxidant reactivity (TAR), SOD, GPx, and CAT, and those were scored as 0, 1, and 2 according the mean and standard deviation (SD) of healthy subjects. A score of 0 was assigned if the measurement of MDA, ox-LDL, CAT, or SOD (oxidants) was ranging mean + 1SD or if TAR and GPx (antioxidants) values were mean − 1SD.
When the values were between +1SD and +2SD of the oxidants or −1SD and −2SD of the antioxidants, a value of 1 point was assigned. A value of 2 points was scored if the oxidant values were more than +2SD or the antioxidant values were less than −2SD. In this case, OSS ranging from 0 to 12 and in a dichotomous way, 6 is the cut-off [168]. This OSS seems more accessible to clinical research than the previous, but it needs to be tested in other clinical settings, since it has only been used in phenylketonuria.
In addition, in 2014, two variants of OSS have been published. The first is called OSI, because it analyzes the oxidant/antioxidant balance as the original OSI, but really is an index due to the oxidizing part which includes total lipoperoxides measured as their different forms (conjugated dienes, ketodienes, compounds with conjugated double bonds, conjugated trienes, and MDA), and in the antioxidant part, called the antioxidant defense (AOD), there is an evaluation of α-tocopherol, retinol, and reduced and oxidized glutathione levels plus SOD activity. First is necessary to obtain the result of the division between each value and the mean value of the control group of each parameter. After that, a multiplication of the oxidized components is placed in the numerator and the multiplication of the antioxidant counterpart in the denominator. As the control value is 1, in the interpretation, values above 1 indicate OS [169]. As the calculation is complex, their use seems impractical.
The last proposal includes 12 different blood OS parameters, four oxidized components (protein carbonyls, protein sulfhydryl, MDA, and 8-OHdG), and eight antioxidant components (SOD, GPx, glutathione reductase [GR], CAT, GSH,  The percentage of subjects with OS in the TC group decreases after 6 mo.
The TC exercise has an antioxidative effect in the older adults with MetS.

[151]
Exploratory study of a single group Older adults (71 ± 6 y, uric acid, bilirubin, and vitamin C). In the report, the calculation is not very explicit; only the authors explain that specific Z-scores of each parameter were computed, and after an OSS was calculated without explain how this is performed [170]; therefore, its application is doubtful. One common point in all those proposals is that MDA is used as oxidized marker; this is possible because this parameter is an index of lipoperoxidation involved in both recent and chronic damages, so although much has been questioned, especially its form of measurement, is the marker that best reflects the oxidative state in any health situation or disease. The other point to highlight is the use of the main antioxidant enzymes, SOD and GPx, which importance was discussed at the beginning of this section, together the measurement of the antioxidant capacity, a parameter, in which all the indexes shown here are in agreement, independently of the method by it is evaluated.
Finally, all the proposals that were revised concord in an important aspect; the measurement of the involved oxidative stress markers as an indicator is more sensitive to the imbalance between oxidants and antioxidants than individual parameters.
Indeed, the SS is a proposal with the advantage that it is possible to increase the number of markers (and therefore the total score), both oxidized molecules and antioxidants, being able to include oxidized proteins, DNA damage, and other enzymes or antioxidant molecules; as long as the cutoff values are taken to decide when the marker is increased or decreased, as the case may be.
In this review, we analyzed the main indexes as indicators to measure oxidative stress grades linked to different clinical and health settings, highlighting the advantages and disadvantages of each of them. It should be noted that the modulating mechanisms of OS in humans are complex and interact with each other; hence, the SS can be useful to determine if an individual has a high level of OS and thus proceeds with a possible antioxidant therapy, thus improving your state of health and quality of life.
It is worth mentioning as a limitation of this review that the indexes for measuring oxidative stress analyzed here are not the only ones. We find in the literature review of others, such as the prooxidant-antioxidant balance (PAB) [171] and the oxidation-reduction potential (ORP) [172], which were not included because their fundaments are outside the scope of this work.
Despite the abundant information available so far, more clinical research is needed, especially on prognosis, to assess the usefulness of the different indices, and comparison studies between them, as well as an analysis of the literature of the indicators that evaluate both components, oxidant and antioxidant in a single measurement, as in the PAB, and the redox potential of a sample, such as the ORP.

Conclusion
We made a review of the state of the art of knowledge about the measurement of OS in humans, focused on different ways to integrally evaluate it.
Regardless of the components for the calculation of the different indexes analyzed, all have been able to demonstrate that they can differentiate the OS in pathological or biological stress states; however, not all measure both intracellular and extracellular components involved in the OS; thus, we think they are not equivalent. We cannot say that one index is the best for measuring OS, but we can suggest that there are some proposals that, despite being different, seem to be equivalent in their interpretation since they allow differentiating the types of OS and establishing an intensity or severity, such as Oxy-score and the different proposals of oxidative stress score.
Perhaps an interesting way to make this evaluation is to combine the measurement of various markers, both oxidized and antioxidant molecules (enzymatic and non-enzymatic), besides a simple way to put them together through an index whose calculation is not complicated, such as the oxidative stress index, which has demonstrated to be reliable, practical, and with clinical utility.