“2n analytical platform” to update procedures in ... - ACS Publications

May 6, 2019 - Estimation of the precise time since death was performed on hospital deaths occurred in casualty, by medico-legal and postmortem ...
2 downloads 0 Views 761KB Size
Subscriber access provided by UNIV OF LOUISIANA

Technical Note

“2n analytical platform” to update procedures in thanatochemistry: estimation of Post Mortem Interval in vitreous humor Roberta Risoluti, Silvia Canepari, Paola Frati, Vittorio Fineschi, and Stefano Materazzi Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.9b01443 • Publication Date (Web): 06 May 2019 Downloaded from http://pubs.acs.org on May 7, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

“2n analytical platform” to update procedures in thanatochemistry: estimation of Post Mortem Interval in vitreous humor Roberta Risoluti§*, Silvia Canepari§, Paola Frati†, Vittorio Fineschi† and Stefano Materazzi§

§Department †Department

of Chemistry, “Sapienza” University of Rome, p.le A.Moro 5, 00185 Rome Italy of Anatomical, Histological, Forensic Medicine and Orthopaedic Sciences, ‘‘Sapienza’’, University of Rome,

p.le A.Moro 5, 00185 Rome, Italy

KEYWORDS: thanatochemistry, vitreous humor, PMI estimation, chemometrics, multiparametric approach

ABSTRACT: In this work, a novel multiway approach by spectroscopy and thermogravimetry associated to chemometrics is developed providing a multiparametric characterization of vitreous humor as a function of the time since death. Estimation of the precise time since death was performed on hospital deaths occurred in casualty, by medico-legal and postmortem examination, with no metabolic disorders. Micro and macro elements in vitreous specimen were determined by ICP-OES and were found to be diagnostic in predicting the Post Mortem Interval (PMI). The percentage of bulk and bound water provided by thermal analysis investigation was correlated to spectroscopic analysis and chemometric tools were used to compare results and to develop a model of prediction of PMI. The study reveals a significant role of P, S and Mg in addition to the potassium concentration in determining the death interval and permitting to increase accuracy with respect to conventional procedures and to extend the investigation of PMI to 15 days.

The postmortem estimation of the exact time of death always proves to be a challenge for forensic pathologists and usually requires the interpretation of the cadaveric processes to be determined, as it is not supported by rigorous scientific model 1-3 Among the numerous biological fluid, the promising role of the vitreous humor in postmortem estimation of the time since death was highlighted in literature 4-8, as it demonstrated to be able to preserve its stability over time. A number of statistical approaches or innovative tests for estimation of the time since death have been proposed, suggesting correlations between the Post Mortem Interval (PMI) and the concentration of analytes or metabolites in vitreous humor; among these, the potassium concentration 9-12, the hypoxantines 4, the amino acids 13, 14 and the cysteine determination 15 were found to be useful to predict the PMI in the few hours after death 16. Neverthless, the main concern related to the reported studies, consists of the small range of PMI that can be associated to a particular instrumental response when a single parameter is monitored. In particular, significant differences in the accuracy of the estimation of the time since death may be observed as a function of the technique as well as the formula involved 17, 18. On the contrary, the comparison of complex matrices always requires a multidimensional approach, due to the increasing number of data to be analyzed and interpreted 19- 23, particularly when a biological matrix is involved 24, 25 and the monitoring of its alteration as a function of time is under investigation 7. As a consequence, the importance of a multidisciplinary approach for the characterization of vitreous humor that responds linearly to the postmortem interval, is increasingly emerging 16, 26. In this work, a novel multi-way approach for vitreous humor analysis is provided for an accurate dating of the Post Mortem

Interval (PMI). A comprehensive evaluation of the micro and macro elements of vitreous humor is performed by Inductively Coupled Plasma - Optical Emission Spectrometry (ICP–OES); the contribution of each element in PMI estimation was derived as a function of time since death and correlated to the chemical composition (percentage of water and the residual metal oxides) obtained by thermoanalytical investigation of vitreous humor. Chemometric tools are used to develop a twoway multiparametric platform to predict the PMI that permits to extend the estimation of the accurate time since death up to 15 days. Results were compared to the outcomes from medico-legal autopsy for a statistical validation of the method. EXPERIMENTAL SECTION Materials Vitreous humor samples (about 2 mL) were collected from male and female through a scleral puncture on the lateral canthus of the eye and stored at -20°C prior to analysis. After drawing the vitreous humor, 2 mL of normal saline was injected in each eye for cosmetic restoration of eyeball. Contaminated samples with tissue fragments were discarded. A detailed description of all the samples under investigation is reported in Table S1. Digestion of vitreous samples prior to ICP-OES with the aim of determining the micro (Cu, Fe, Sr, Tl and Zn) and macro elements (Ca, K, Mg, Na, P and S) was performed using deionized water (Elga Lab Water Purelab Plus), nitric acid (65%, Suprapur, Merck KGaA, Darmstadt, Germany) and hydrogen peroxide (30%, Suprapur, Merck KGaA, Darmstadt, Germany. Yttrium (Y) reference standard solution (Panreac

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Química, Barcelona, Spain) was considered as an internal standard at the concentration of 100 μg/L. Instrumentals Inductively Coupled Plasma - Optical Emission Spectrometry (ICP–OES) from Varian (Vista MPX CCD Simultaneous, Varian, Victoria, Mulgrave, Australia), equipped with an ultrasonic nebulizer (U 5000 AT+ Cetac Technologies, Inc., Omaha, NE, USA) was used to characterize vitreous samples. The method was optimized in order to achieve the most performing conditions of sensitivity and accuracy in vitreous samples. All the operating parameters are reported in Table S2 while the validation results, are summarized in Table S3 and Table S4 respectively. Thermoanalytical characterization of vitreus humor was performed according to the previously optimized method 7. A Perkin Elmer TGA7 Thermobalance (Massachusetts, USA) was used to acquire the thermogravimetric curves. Vitreous specimens without pretreatment (30 µl) were placed into the crucible, and temperature was measured using a thermocouple directly attached to the crucible. The temperature was raised from 20°C to 800°C, with a 10°C/min heating rate, as the best resolution rate, under an atmosphere of air as carrier gas (100 ml/min flow rate). The instrumental response was calibrated using the Curie-point transition of standard metals, as specified by the equipment recommendations and a number of three replicates for each sample were acquired to ensure reproducibility. Derivative Thermogravimetric data (DTG) were also calculated to compare samples and represent the derivative of the function TG (T) with respect T. Analytical strategy A number of 50 vitreous specimens were collected in this study: 37 samples were considered for model development in order to define a characteristic vitreous humor composition. To this aim, different causes of death were evaluated in the range 0-15 days as Post Mortem Interval (PMI). In addition, 13 vitreous samples were used for the cross-internal validation of the model in order to provide the figures of merit for prediction of PMI. Finally, 5 vitreous samples with unknown time of death from medico-legal autopsy were analyzed and processed by using the validated model. The experimental design was built as follows: a preliminary step was performed by the mean of reference standard materials in order to calibrate the spectral response and to evaluate the matrix effect on the sensitivity of the method. To this end, three different pretreatment for digestion of vitreous humor were compared in order to select the most performing outcomes for each element: the acid digestion using HNO3 and H2O2 (3:1) at room temperature or heating sample in a water bath on electric plate; in addition, the improvement in the spectral response was evaluated after microwave-assisted acid digestion using HNO3 and H2O2 (3:1). The same samples were analyzed by thermogravimetry in order to calculate the thermally induced weight losses and to correlate the thermal behaviour to the elemental composition. In a second stage, chemometric tools were applied to perform correlations among two type of instrumental signals and to provide a model of prediction of PMI as a function of a characteristic composition of vitreous humor after the occurring of death. An unsupervised technique based on Principal Component Analysis (PCA) 27 has been used as exploratory method to investigate whether the two-way approach may differentiate samples according to different

amount of each element, while Partial Least SquareDiscriminant Analysis (PLS-DA) 28 was applied for validation and to perform prediction. The first identifies directions in the dataset with higher variability and permits to observe correlations among samples. The regression technique PLSDA permits to build a model of prediction between the acquired measurements (vitreous samples) and class membership information (time since death) on the basis of the directions that are useful to differentiate classes. In addition, a combined plot of scores (investigated samples) and loadings (variables) provided important information about the contribution of each parameter. The combination of PCA and PLS-DA is a so-called supervised technique, meaning that it uses class information to improve sample separation. Chemometric rules were used to compare data by processing a data matrix as follows: 45 samples and 13 variables. The variables included all the investigated metals (Ca, Cu, Fe, K, Mg, Na, P, S, Sr, Tl, and Zn) expressed as (μg/L) and the results from TG analyses expressed as percentage of weight losses (%ww) corresponding to the calculated water amount and residual quantity under thermally induced decomposition. RESULTS AND DISCUSSION Two-Way analytical platform Preliminary, a multiparametric characterization of vitreous was performed for all the collected samples in order to correlate the chemical composition to the time since death. To this aim, eleven elements including potassium (Ca, Cu, Fe, K, Mg, Na, P, S, Sr, Tl and Zn) were considered and the relative abundance in vitreous humor was determined by ICPOES. Among the investigated sample pretreatments, the acid digestion using HNO3 and H2O2 in the ratio 3:1 at room temperature permitted to achieve the most performing results. Principal Component Analysis of the recorded data was used as display method and in particular, a combined plot of scores (results for each sample) and loadings (investigated variables) permitted to identify correlation among data as a function of the PMI. A number of mathematical pretreatments of signals were investigated including Standard-Normal-Variate (SNV) Transformation 29, Multiplicative Scatter Correction (MSC) 30, and normalization 31 as scatter-correction methods, while the Savitzky-Golay (SG) polynomial-derivative filter 32 was considered as spectral-derivation technique. The best separation conditions among samples were achieved when a combined approach based on MSC and second derivative was applied. Results of the chemometric analysis is reported in the biplot in Figure 1, where the samples are represented by symbols and colors are used to differentiate samples according to the PMI, while vectors identify the variables affecting most the distribution of samples. The biplot uses points to represent the scores of the observations on the principal components and vectors to represent the coefficients of the variables on the principal components. Both direction and length of the vectors can be interpreted: the vector position is the direction which has the highest squared multiple correlation with the principal components; the length of the vector is proportional to the squared multiple correlation between the fitted values for the variable and the variable itself. Interestingly, important information may be deducted from Figure 1: first, the chemometric analysis of spectroscopic data

ACS Paragon Plus Environment

Page 2 of 8

24 h

1 day

3 days

5 days

2 days

4 days

6 days

7 days

9 days

11 days

11 days

2

P 1050

1

900

Mg Mg 4 5

1500 750 0 600

600 450 300 150

0 PC 1 (69 %)

1

2

3

4

Figure 1. Chemometric analysis of spectroscopic data: distribution of samples according to the PMI0 (black), PMI1 (orange circle), PMI2 (orange triangle), PMI3 (green circle), PMI4 (green triangle), PMI5 (red circle), PMI6 (red triangle), PMI7 (blue circle), PMI9 (pink circle), PMI11 (pink triangle) and PMI15 (light blue circle).

As a consequence, a good correlation among samples belonging to the same class (PMI) may be observed and samples resulted well grouped in the plot as a function of the different PMI. In addition, the location of the investigated variables (loadings) in the plot, confirmed the role of K in determining the time of death, resulting in a vector affecting most the PC 1 and suggested that further elements significantly contribute to the differentiation of the samples. In particular, S, Mg and P were found to be responsible for the separation of the samples according to the PC 1, while Na and Ca did not distinguish samples according to the PMI, as the vectors affected most the PC 2. In addition, a negative correlation may be observed between vectors corresponding to Ca and Na. The chemometric analysis of elements in vitreuos humor also revealed a positive correlation of K, S, Mg and P with PC 1 demonstrating an increasing amount of K, S, Mg, and P over time. The trend of these elements with time, obtained by chemometric tools, is also confirmed by the evaluation of the average amount of each element over time, reported in Figure 2. All these results permit to consider these elements as novel predictors of the PMI in post mortem investigations and suggest a valuable contribution of chemometric analysis in considering the multiparametric monitoring of all the elements in a single instrumental process. In a second stage, all the samples included in the study were analyzed by thermogravimetric analysis (TGA) in order to calculate the percentage of water and the residue corresponding to metal oxides, produced under combustive conditions 7. A decreasing amount in the water content may be observed as a function of the PMI, moving from an average value of 97.1 ± 0.5 % (at death, PMI=0 days) to 93.2 ± 0.1 5 (15 days after death). On the contrary, an increasing amount of the residual metal oxides may be calculated moving from the first hours since death (average value of 1.2 ± 0.2 %) to 15 days after death (2.7 ± 0.1 %), as a result of the alteration of the membrane with time. Data from TGA and ICP-OES were compared by chemometric

450

900

300

750

150

600

0 1

450

3

4

6

5

6

Mg7

3 4

2

4 5

3

Mg

5 6 7 9 6 7 9

4

9

9

11

5

SS6

11

15

15

P

Mg

Mg

10 11 15 11 15

7

9

9

11

PMI (days)

11

Mg

Mg

15

150 160.000 16000.000 0 14000.000 140.000

1 2 12000.000 120.000 10000.000 100.000

3

4

5

6

7

S

8000.000 80.000 6000.000 60.000 160.000

S

7

PMI (days)

300

4000.000 40.000 140.000 2000.000 120.000 20.000

120.000 40.000 100.000 20.000

15

S

100.000 0.0000.000 1 2 3 4 5 1 160.000 2 80.000 3 4 5 140.000 60.000

Centinaia

-1

Mg1050

Centinaia

-2

2

300 1050 150 900 0 750 1 2 1 600 2 3

Amount (μg/L)

-3

1

450

Amount (μg/L)

0

-1

P

P

4500 1050 3000 900

750 0

P

tool in order to develop a novel platform permitting to improve samples separation 6according to11 PMI and leading to 6 7 7 9 9 10 11 15 15 P the differentiation of sample belonging to different classes.

Centinaia

provided the discrimination of vitreous samples according to PC 1 (69 % of explained variance).

PC 2 (22 %)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

P

Analytical Chemistry

Amount (μg/L)

Page 3 of 8

7500 7500 6000 6000 4500 4500 3000 3000 12000 1500 1500 10500 0 0 9000 1 2 1 2 3 3 4 4 55 7500 6000 4500 12000 10500 3000 9000 1500 0 7500 1 2 3 6000

S

S 6 6

77 9 10 11 15 9 11 15

PMI (days)

80.000 0.000 Figure 2. Amount of elements recovered in vitreous humor as a S 60.000 2 3 4 5 6 7 9 11 15 function of the time since1 death . 40.000 20.000 0.000 In fact, results from ICP-OES/Chemometrics as well as the 1 2 3 4 5 6 7 9 11 15 previous data by TGA 7 only permitted to differentiate some intervals of PMI (2, 4, 6 days since death). The chemometric evaluation of the combined data from TGA and ICP-OES is reported in Figure 3, where a biplot (combined plot of scores and loadings) was used to correlate samples.

ACS Paragon Plus Environment

S

92 90

Analytical Chemistry 88 1 day

3 days

5 days

2 days

4 days

6 days

7 days

15 days

9 days 11 days

2 5

Weight %

4 3 1 2

PC 2 (22 %) PC 2 (8 %) 0

1

H2O 1

2

3

Water release

4

5

6

7

9

11

15

H2O

1

2

3

0

4

5

6

PMI (days)

-1

7

9

11

15

RESIDUO

Metal oxides residue

-1

-2

3 2.7

-3

-4

-3 -3

-2 -2

-1 -1

0 PC01 (87 %) PC 1 (69 %)

1 1

2

3 2

2.4

4 3

2.1

4

Figure 3. Scores plot obtained by PCA of the Two-Way multiparametric approach as a function of the PMI0 (black), PMI1 (orange circle), PMI2 (orange triangle), PMI3 (green circle), PMI4 (green triangle), PMI5 (red circle), PMI6 (red triangle), PMI7 (blue circle), PMI9 (pink circle), PMI11 (pink triangle) and PMI15 (light blue circle).

Weight %

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

24 h 24 h

Page 4 of 8

H2O

86 84 82 80 100 98 96 94 92 90 88 86 84 82 80

1.8 1.5 1.2

0.3 3 2.70 2.4

1

2

2.1

3

4

5

6

PMI (days)

1.8

The interpretation of the resulting biplot provides additional information on samples distribution: a positive correlation to PC1 of the variables K, Mg, P, S and the residue from TGA may be observed while a negative correlation of the water content of vitreous humor to PC1 may be deducted from the location of the corresponding vector in the plot. The main result of the two-way multiparametric analysis, consists of a more performing separation of the samples belonging to the different class of PMI than the single instrumental response. In fact, despite the results from spectroscopic analysis contribute in extending the range of PMI and permit to investigate also samples collected 15 days after death, the Figure 3 demonstrated that a combined multiparametric approach significantly improve the separation of samples according to PMI, especially in the first days since death, leading to a more accurate and sensitive tool for medico-legal investigation. Chemometric investigation of the two-way multiparametric analysis of vitreour humor has shown a good accordance of results between spectroscopic and thermogravimetric data: in fact, as reported in Figure 4, the decrease in the water content and the parallel increase in the metal oxides over time may be associated to the increase in the amount of K, S, Mg and P recovered by ICP-OES, confirming the results are complementary.

RESIDUO

RESIDUO

0.9 0.6

7

9

11

15

1.5

Figure 4. Calculated water content and residue of metal oxides 1.2 RESIDUO from 0.9 TGA as a function of the time since death. 0.6 0.3 The results obtained by chemometric evaluation of data from 0 1 techniques different highlights the great potential of the two2 3 4 5 6 9 11 way approach in providing a7 more accurate prediction of the 15 time since death, as samples were found to be better separated according to PMI, thus permitting to estimate the correct day of death.

Validation On the basis of the promising results, a model of prediction for the PMI estimation in vitreous humor was developed and validated by Partial Least Square-Discriminant Analysis (PLS-DA). For model calibration, the 75 % of the dataset (37 vitreous specimens) was used, while the 25 % of samples (the remaining 13 samples) was considered for validation. Figures of merit of the model were calculated in order to evaluate the performances of the model and the prediction ability in estimating the PMI: in particular, the Root-Mean-Squared Error of Calibration (RMSEC), RootMean-Squared Error of CrossValidation (RMSECV), and Root-Mean-Squared Error of Prediction (RMSEP) 30, 32 . The results of the model are reported in Table 1, as the percentages of the objects that were correctly classified (correctclassification rates). Table Prediction of analytical the Two-Way multiparametric Table 1 1 The prediction abilityability of the Two-way platform by PLS-DA approach by PLS-DA PMI

Correct Classification rate (%) Calibration

Cross-Validation

Prediction

0

86.6

86.6

100.0

1

83.3

88.8

94.4

2

80.0

93.3

93.3

3

91.6

91.6

100.0

4

93.3

93.3

100.0

5

83.3

91.6

91.6

6

66.6

93.3

93.3

7

66.6

83.3

83.3

9

77.7

77.7

88.8

11

66.6

83.3

83.3

15

83.3

91.6

91.6

ACS Paragon Plus Environment

Page 5 of 8

The selection of five latent variables permitted to obtain the lowest RMSECV (equal to 0.05), representing 79 % of the X explained variance and 21 % of the variance in Y. In addition, five unknown samples were analyzed by the two-way multiparametric platform and all the samples were correctly predicted by the model (Figure 5) and confirmed by further forensic autopsies.

The Supporting Information is available free of charge on the ACS Publications website. Authors reported instrumental condition of ICP-OES analysis and results for quantification as well as a detailed list of the collected vitreous humor sample.

AUTHOR INFORMATION Corresponding Author * Roberta Risoluti, Department of Chemistry, Sapienza University of Rome, Piazzale Aldo Moro 5, 00185 Rome, Italy; Tel +390649913616 fax: +390649387137 e-mail address: [email protected]

5

Author Contributions

4

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

3 2

PC 2 (21 %)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Notes

1

The authors declare no competing financial interest.

0

ACKNOWLEDGMENT

-1 -2 -3

-4

-3

-2

-1

0 PC 1 (79 %)

1

2

3

4

5

Figure 5. Scores plot obtained by PLS-DA model of the Two-Way multiparametric approach as a function of the PMI and prediction of unknown samples

CONCLUSIONS A Two-Way multiparametric platform was developed and validated for the thanatochronological characterization of vitreous humor. Inductively Coupled Plasma - Optical Emission Spectrometry (ICP–OES) and thermogravimetric analysis (TGA) permitted a multiresidual determination of the elements, the water content and the residual metal oxides in vitreous humor, and the identification of characteristic pattern of distribution of each parameter according to the time since death. In particular, the study of biplot performed by chemometrics suggested new insights into the role and the significance of novel diagnostic parameters. In fact, in addition to potassium, the study revealed for the first time, that P, S, Mg and the water content may be considered as good predictors of the PMI, that permit to extend the range of estimation of the time since death until 15 days. These features were found to be promising for PMI estimation, providing for an update of the procedures in medico-legal investigations. In addition, the most performing results were achieved for samples separation according to PMI only when a multi-way approach was involved, permitting to differentiate samples day by day. In conclusion, a model of prediction of PMI in vitreous humor was developed and the results demonstrated a significant improvement in accuracy of the PMI estimation with respect to conventional analytical techniques actually used, as all the processed samples were correctly predicted.

ASSOCIATED CONTENT Supporting Information

Authors appreciate the financial support from the PRIN 2015 (Prot. n. 201545245K ) that permit the development to the research project.

REFERENCES (1) Henssge, C.; Madea, B. Forensic Sci Int. 2007, 165, 182–4. (2) Madea, B. Methods for determining the time since death. For Sci Med Pathol. 2016, 12, 451–85. (3) Madea, B. Estimation of the Time Since Death. 3rd ed. New York, CRC Press 2016. (4) Madea, B.; Kaferstein, H.; Hermann, N.; Sticht, G. Forensic Sci Int. 1994, 65, 9–31. (5) Coe, J.I. Am J Pathol. 1969, 51, 741–50. (6) Coe, J.I. Forensic Sci Int. 1989, 42, 201–13. (7) Materazzi, S.; Gullifa, G.; Fabiano, M.A.; Frati, P.; Santurro, A.; Scopetti, S.; Fineschi, V.; Risoluti, R. Journal of Thermal Analysis and Calorimetry 2017, 130 (1), 549-557 (8) Coe, J.I.; Apple, F.S. J Forensic Sci. 1985, 0, 828–35. (9) Rognum, T.O.; Holmen, S.; Musse, M.A.; Dahlberg, P.S.; Stray-Pedersen, A.; Saugstad, O.D.; Opdal, S.H. Forensic Science International 2016, 262, 160–165 (10) Tumram, K.; Ambade, N.V.; Dongre. A.P. Alexandria Journal of Medicine 2014, 50, 365–368 (11) Di Donna, L.; Napoli, A.; Sindona, G.; Athanassopoulos, C. J. Am. Soc. Mass Spectrom. 2004, 15, 1080-1086 (12) Swain, R.; Kumar, A.; Sahoo, J.; Lakshmy, R.; Gupta, S.K.; Bhardwaj, D.N.; Pandey, R.M. Journal of Forensic and Legal Medicine 2015, 36, 144-148 (13) Patrick, W.J.; Logan, R.W. Arch Dis Child. 1988, 63, 660–2. (14) Croxton, R.S.; Baron, M.G.; Butler, D.; Kent, T.; Sears, V.G. Forensic Sci Int. 2010, 199, 93–102. (15) Ansari, N.; Lodha, A.; Menon, S.K. Biosensors and Bioelectronics 2016, 86, 115–121 (16) Chandrakanth, H.V.; Kanchan, T.; Balaraj, B.M.; Virupaksha, H.S., Chandrashekar, T.N. Journal of Forensic and Legal Medicine 2013, 20, 211-216 (17) Ortmann, J.; Markwerth, P.; Madea, B. Forensic Science International 2016, 269, 1–7 (18) Madea, B.; Rodig, A. Forensic Science International 2006, 164, 87–92

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(19) Ferreira, S.L.C.; Lemos, V.A.; de Carvalho, V. S.; da Silva, E.G.P.; Queiroz, A.F.S.; Felix, C.S.A. da Silva, D.L.F; Dourado, G.B.; Oliveira, R.V. Microchemical Journal, 2018, 140, 176–182 (20) Risoluti, R.; Gregori, A.; Schiavone, S.; Materazzi, S. Anal Chem, 2018, 90-7, 4288-4292 (21) Materazzi, S.; Risoluti, R.; Pinci, S.; Romolo, F. S. Talanta 2017, 174, 673−678. (22) Materazzi, S.; Peluso, G.; Ripani, L.; Risoluti, R. Microchem. J. 2017, 134, 277−283. (23) Rocío, L.; Pérez, G.; Escandar, M. Sustainable Chemistry and Pharmacy, 2016, 4, 1–12 (24) Risoluti, R.; Materazzi, S.; Sorrentino, F.; Maffei, L.; Caprari, P. Talanta, 2016, 159, 425−432. (25) Risoluti, R.; Materazzi, S.; Sorrentino, F.; Bozzi, C.; Caprari, C. Talanta, 2018, 183, 216–222 (26) Madea, B. Forensic Sci Int., 2005, 151, 139–49. (27) Massart, D.L.; Vandeginst, B.G.M.; Buydens, L.M.C.; De Jong, S.; Lewi, P.L.; Smeyers-Verbeke, J. Handbook of Chemometrics and Qualimetrics. Part B, 20B, Elsevier, Amsterdam, 1998, 88–103. (28) Barker, M.; Rayens, M.W. J. Chemomet. 2003, 17, 166–173. (29) Geladi, P.; MacDougall, D.; Martens, H. Appl. Spectrosc. 1985, 39, 491−500. (30) Miller, J. N.; Miller, J. C. Statistics and Chemometrics for Analytical Chemistry; Prentice Hall: Harlow, England, 2000. (31) Savitzky, A.; Golay, M. J. E. Anal. Chem. 1964, 36, 1627−1639. (32) Mark, H.; Workman, J. Chemometrics in Spectroscopy; Academic Press: Amsterdam, 2007.

ACS Paragon Plus Environment

Page 6 of 8

Page 7 of 8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

TOC of the manuscript

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Paragon Plus Environment

Page 8 of 8