Real-world Data on Adult Status Epilepticus from Latin America: Clinical Characteristics and Outcomes in a Mexican Tertiary Center

Highlights

• •This study represents the largest real-world cohort of adult status epilepticus (SE) in a Mexican center.
• •Acute symptomatic etiologies were the most common cause of SE.
• •Nonconvulsive SE was associated with poor functional outcomes.
• •Favorable outcomes were significantly associated with prior epilepsy.
• •Findings support risk stratification and tailored management in low-resource settings.

Abstract

Background

Status epilepticus is a life-threatening emergency in neurology practice. Latin America faces a disproportionate epilepsy burden; little data is available on this population type.

Objective

To describe the clinical characteristics, treatment approaches, and functional outcomes of adult patients with SE in a tertiary neurological center in Mexico City.

Methods

We conducted a retrospective cohort study involving 103 adult patients diagnosed with SE from 2020 to 2024. Functional outcomes were measured using the modified Rankin Scale (mRS), with scores 0–2 classified as favorable. Severity was assessed with the STESS. Chi-square, Mann–Whitney U tests, and logistic regression were used to identify variables related to outcomes.

Results

The median patient age was 36 (IQR 18–79), with 50.5% of patients being female. Most had a prior diagnosis of epilepsy (80.6%). Convulsive SE (74.8%) and acute symptomatic etiologies (70.9%) were predominant. Infections were the leading cause among acute symptomatic cases (35%). Refractory SE occurred in 41.7% of cases, and 14.6% progressed to super-refractory SE. Favorable outcomes were significantly associated with prior epilepsy (p = 0.032), and SE subtype (p = 0.003). Multivariate analysis confirmed prior epilepsy as an independent predictor of good outcome (OR = 0.65, 95% CI 0.48–0.87; p = 0.004).

Conclusions

This study offers the most extensive characterization of SE in a Mexican center. Infectious etiologies dominate the clinical landscape, with functional outcome linked to baseline epilepsy history and SE subtype.

1. Introduction

Status epilepticus (SE) is a neurological emergency operationally defined as ≥5 min of continuous seizures or two or more seizures without a complete return to baseline consciousness [1], [2]. Focal SE with impaired consciousness is defined as a single focal seizure lasting longer than 10 min [1]. The proportion of patients with epilepsy (PWE) who develop SE varies with age, etiology, and epilepsy classification. Epidemiological studies estimate that 15–20% of PWE will experience at least one episode of SE during their lifetime, and 12% will present with SE as their initial manifestation [3]. Epilepsy is a primary global health concern, affecting approximately 51.7 million people worldwide [4]. A previous diagnosis of epilepsy, particularly when associated with subtherapeutic anti-seizure medication (ASMs) levels, is the most decisive single risk factor for the development of generalized convulsive SE [5].

The incidence of SE follows a bimodal distribution, with peaks in the first decade of life and after age 60 [6]. The overall incidence is 9 to 41 per 100,000 annually [7]. Despite advances in classification and management, SE remains a significant clinical challenge, as the underlying causes and outcomes may vary based on demographic, and healthcare system factors [8]. In Mexico, only one study provides demographic information on SE [9]. Given this gap, the present study aims to describe the demographic, clinical, and treatment characteristics of adult patients with SE admitted to a tertiary neurological care center in Mexico.

2. Materials and methods

2.1. Patient selection

Following Local Bioethics and Research Committee approval, a retrospective review was conducted to identify patients diagnosed with SE, as defined by the 2015 ILAE position paper on SE, at the National Institute of Neurology and Neurosurgery “Manuel Velasco Suárez” (NINNMVS) in Mexico City between 2020 and 2024. The inclusion criteria were: adult patients (18 years or older), receiving treatment at our institution, and having complete demographic records. Demographic and clinical data were obtained from institutional medical records, including age, sex, past medical history, duration of hospitalization, prior diagnosis of epilepsy, prior ASM use, monthly seizure frequency, and non-neurological comorbidities. Additionally, information on mortality, type of SE (generalized convulsive, non-convulsive, or focal), and treatment for SE was collected. Data extraction was performed by trained research personnel (I.G.O.) using a data collection sheet. A second reviewer independently verified 20% of the entries to ensure consistency and reduce information bias.

2.2. Clinical scales and outcome measures

We retrospectively calculated each patient's Status Epilepticus Severity Score (STESS) to standardize severity [10]. The score was separated into favorable (0–3) and unfavorable (4–6) outcomes. Functional status at discharge was assessed using the modified Rankin Scale (mRS). For analysis, we grouped patients into good (mRS 0–2) and poor functional outcomes (mRS 3–6). Furthermore, data on ICU admission and duration of stay were documented. Patients with incomplete clinical records or without documented discharge outcomes were excluded.

2.3. Status epilepticus and etiologic classification

SE was classified following the four-axis framework proposed by Trinka et al. [1], including semiology, etiology, EEG characteristics, and age. Etiologies were categorized as known (symptomatic) or unknown (cryptogenic). Known causes were further divided into acute, remote (post-stroke or post-encephalitic), and progressive (brain tumors, paraneoplastic process). Acute causes were further divided according to the new classification [11]: Acute-triggering factors in epilepsy (TFE), acute-primary CNS, acute-secondary CNS, and acute-toxic, including drug/alcohol intoxication or withdrawal. Only adult (18–59 years) and elderly (≥60 years) patients were included, as our tertiary care center specializes in adult neurology and does not treat pediatric patients. The classification of SE was established by a certified epileptologist and adult neurologist (E.P.-A) and the ED team based on clinical and EEG findings.

2.4. Treatment protocol

All patients were treated according to the standardized in-house protocol Código Estado Epiléptico (Code Status Epilepticus) protocol from the NINNMVS [12]. The protocol establishes stepwise management beginning with benzodiazepines (midazolam 10 mg IM, lorazepam 0.1 mg/kg IV, or diazepam 0.15–0.2 mg/kg IV), followed—if seizures persist—by second-line intravenous ASMs such as levetiracetam (60 mg/kg, max 4500 mg), valproate (40 mg/kg, max 3000 mg), or phenytoin (15–20 mg/kg) [12]. Refractory SE (RSE) was managed with anesthetic infusions (midazolam, propofol, thiopental, or ketamine) under continuous EEG monitoring for at least 24 h, targeting seizure or burst-suppression. Superrefractory SE (SRSE) is approached with immunotherapy, ketogenic diet, or other advanced measures [12].

2.5. Statistical analysis

SPSS ®version 29 was used for data entry and statistical analysis. Categorical variables were summarized as frequencies and percentages (%), and continuous variables were described using the median and interquartile range (IQR), unless otherwise specified. Before conducting group comparisons, we applied the Shapiro-Wilk test to assess the normality of continuous variables. Afterwards, group comparisons were performed using the Chi-square test for categorical variables and the Mann–Whitney U test for ordinal or non-normally distributed variables. Two-tailed p-values < 0.05 were considered significant in all statistical analyses. Variables with p < 0.05 in bivariate analysis were included in a forward stepwise logistic regression model to identify independent predictors of functional outcome. No Bonferroni correction was applied, as the bivariate analyses in Table 4 were exploratory and aimed to identify potential variables for inclusion in the logistic regression model. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to assess the strength of the associations in Table 4.

Table 4.

Predictors of functional outcome at hospital discharge based on multivariate logistic regression.

		All patients (n = 103)	Good outcome Rankin 0–2 (n = 53)	Poor outcome Rankin 3–6 (n = 50)	OR (IC 95%)	p-value
Previous diagnosis of epilepsy		83 (80.6%)	47 (56.6%)	36 (43.4%)	0.647 (0.481–0.870)	0.004*
Awareness	Preserved	20 (19.4%)	15 (75.0%)	5 (25.0%)	0.769 (0.581–1.017)	0.066
	Impaired	83 (80.6%)	38 (45.8%)	45 (54.2%)	1.301 (0.933–1.722)	0.066
Previous SE		37 (35.9%)	22 (59.5%)	15 (40.5%)	0.908 (0.731–1.127)	0.382
Type of SE	Convulsive	68 (66%)	37 (54.4%)	31 (45.6%)	2.932 (0.568–15.152)	0.199
	NCSE	18 (17.5%)	5 (27.8%)	13 (72.2%)	9.100 (1.389–59.619)	0.021*
	Focal	8 (7.8%)	4 (50%)	4 (50%)	3.500 (0.431–28.447)	0.241

Clinical variables with statistical significance (p < 0.05) in the univariate analysis, along with clinically relevant variables supported by prior literature, were entered into a multivariate logistic regression model using a forward stepwise method. Odds ratios (OR) and 95% confidence intervals (CI) are shown. An OR < 1 indicates a protective association with good functional outcome (Rankin 0–2), whereas an OR > 1 indicates increased odds of poor functional outcome (Rankin 3–6).

3. Results

3.1. Patient sample selection, past medical history, and clinical characteristics

Out of 120 patients evaluated, 103 met the inclusion criteria and were included in the final analysis. Of these, 52 (50.5%) were female. The median age at admission was 36 years (IQR: 18–79). Comorbidities were frequent (64.1%), most commonly hypertension and type 2 diabetes (Table 1). A prior diagnosis of epilepsy was present in 80.6% (n = 83) of patients; among them, the most frequent etiology was structural epilepsy (39.8%, n = 41). Their clinical characteristics are displayed in Supplementary Table 1s.

Table 1.

Status epilepticus cohort demographics.

	n (%) or median [IQR]
Sex, female	52 (50.5%)
Age	36 [18–79]
Right-handed dominance	91 (88.3%)
Etiology of SE
Acute symptomatic*	73 (70.9%)
•Acute-secondary CNS	41 (56.1%)
•Acute-TFE	16 (21.9%)
•Acute-primary CNS	12 (16.4%)
•Acute-toxic	4 (5.4%)
Remote symptomatic	8 (7.8%)
Progressive symptomatic	7 (6.8%)
Unknown (cryptogenic)	15 (14.6%)
Type of SE
Generalized convulsive	77 (74.8%)
Non-convulsive	18 (17.5%)
Focal	8 (7.8%)
Evolution of SE
Refractory SE	43 (41.7%)
Super Refractory SE	15 (14.6%)
History of previous SE	37 (35.9%)
Past medical history
Non-neurologic comorbidities	66 (64.1%)
Arterial hypertension	17 (16.5%)
Type 2 diabetes mellitus	11 (10.7%)
Depression	8 (7.8%)
FND	9 (8.7%)

SE: status epilepticus; FND: functional neurological disorder; TFE: triggering factors in epilepsy.

*Acute-secondary CNS: systemic infection or metabolic disturbances; Acute-TFE: withdrawal, low levels, or inappropriate prescription of ASMs in patients with pre-existing epilepsy; Acute-primary CNS: cerebrovascular diseases, active CNS infections, or head trauma; Acute-toxic: drug or alcohol intoxication and withdrawal.

3.2. Status epilepticus characteristics and in-hospital management

Generalized convulsive SE was most predominant (74.8%), followed by non-convulsive SE (17.5%) and focal SE (7.8%). The most common etiology was acute symptomatic (n = 73, 70.9%). Further information is clarified in Table 1. The median length of hospital stay was 3 days (IQR 0–133). Nearly all patients received a first-line benzodiazepine, most commonly midazolam in 99 (96.1%), followed by lorazepam in 3 (2.9%) and orodispersible clonazepam in 1 (1%). Regarding second-line ASMs, levetiracetam was the most frequently administered in 38 patients (36.9%), followed by valproic acid and phenytoin. (Table 2) The management was standardized according to our in-house protocol, as mentioned in the methods section. However, in some cases, it was modified based on medication availability. Neuroimaging was performed upon ED admission for 56 (54.3%) patients. Among them, 35 (34.0%) patients underwent a non-contrast CT scan, 11 (10.7%) underwent a contrast-enhanced CT scan, and 10 (9.7%) underwent a brain MRI.

Table 2.

In-hospital management and treatment strategies in patients with status epilepticus.

Treatment strategies
First-line benzodiazepine
Midazolam	99 (96.1%)
Lorazepam	3 (2.9%)
Orodispersible Clonazepam	1 (1.0%)
Second-line treatment
Levetiracetam	38 (36.9%)
Valproic acid	30 (29.1%)
Phenytoin	17 (16.5%)
Neurological ICU
Neuro ICU Admission	27 (26.2%)
Mean neuro ICU stay duration (days)	5.3 ± 12.0

Percentages are based on the total number of patients (n = 103). Abbreviations: RSE, Refractory Status Epilepticus; ICU, Intensive Care Unit.

A total of 43 (41.7%) patients were classified as RSE, and 15 (14.6%) progressed to SRSE. Admission to the ICU was required in 27 (26.2%) patients, with a mean stay of 5.3 ± 12.0 days. Among patients who developed RSE, only 5 (11.6%) achieved a good functional outcome at discharge, while 11 (25.6%) had a poor outcome. However, this difference was not statistically significant compared to patients without RSE (p = 1.000). Similarly, among patients who progressed to SRSE, 3 (20.0%) had good outcomes, and 2 (13.3%) had poor outcomes, with no significant difference compared with non-SRSE cases (p = 0.367). Death at discharge (mRS = 6) occurred in 9 patients (8.7%). No statistically significant differences were observed across key clinical variables (e.g., age, sex, previous SE).

3.3. Association between clinical variables and outcomes: Rankin and STESS scales

A previous diagnosis of epilepsy was significantly more frequent among patients with favorable outcomes (56.6%, n = 47) compared to those with poor outcomes (43.4%) (p = 0.032). Furthermore, preserved awareness was associated with better outcomes (p = 0.019). The type of SE was also significantly associated with prognosis (p = 0.003). (Table 3).

Table 3.

Bivariate comparison of clinical characteristics by functional outcome at hospital discharge (Rankin 0–2 vs. 3–6).

	Good outcome Rankin 0–2 (n = 53)	Poor outcome Rankin 3–6 (n = 50)	p-value
Sex, female	27 (51.9%)	25 (48.1%)	0.924
Age	33 [25–50]	51.5 [25–60]	0.076
Previous SE	22 (41.5%)	16 (32%)	0.224
Refractory SE	23 (43.4%)	20 (40%)	0.727
Super-Refractory SE	6 (11.3%)	9 (18%)	0.337
Type of SE	Convulsive: 44 (83%)	Convulsive: 33 (66%)	0.003*
	NCSE/Focal: 9 (16.9%)	NCSE/Focal: 17 (34%)
Etiology of SE	Acute: 34 (64.2%)	Acute: 39 (78%)	0.463
	Remote: 5 (9.4%)	Remote: 3 (6%)
	Progressive: 4 (7.5%)	Progressive: 3 (6%)
	Cryptogenic: 10 (18.9%)	Cryptogenic: 5 (10%)
Previous diagnosis of epilepsy	47 (56.6%)	36 (43.4%)	0.032*
Previous epilepsy surgery	8 (15.1%)	4 (8%)	0.262
Preserved awareness	15 (28.3%)	5 (10%)	0.019*
Non-neurologic comorbidities	32 (60.4%)	34 (68%)	0.420
Hospitalization, days	4 [1–16]	4 [1–28]	0.241
ICU stay, days	0 [0–1.5]	0 [0–6.75]	0.686

Statistically significant values are boldly indicated with an asterisk (p < 0.05). Continuous variables are reported as medians with interquartile ranges. Categorical variables were analyzed using the Chi-square test and are presented as absolute frequencies and percentages. Due to non-normal distribution, continuous variables were analyzed using the Mann–Whitney U test.

In the multivariate analysis (Table 4), a previous diagnosis of epilepsy remained associated with a good functional outcome (OR = 0.65, 95% CI 0.48–0.87; p = 0.004). Preserved awareness showed a non-significant trend toward better outcome (OR = 0.77, 95% CI 0.58–1.02; p = 0.066). History of previous SE was not significantly associated with outcome (p = 0.382). Regarding SE subtypes, non-convulsive SE was independently associated with poor functional outcome (OR = 9.10, 95% CI 1.39–59.62; p = 0.021), whereas convulsive and focal SE were not significant predictors. These findings highlight the prognostic relevance of seizure semiology, with non-convulsive SE being a strong predictor of worse outcomes in our cohort.

The STESS score was calculated for all patients to assess SE severity and its association with outcomes. The median STESS score was 2 (IQR 1–3). Scores of 0 and 1 were recorded in 14.6% and 20.4% of patients, respectively, indicating a lower severity profile. Meanwhile, scores of 4 or higher were seen in 11 patients (10.7%). These findings suggest that most patients in our cohort presented with moderate STESS scores, though a subset exhibited high-severity features requiring closer prognostic evaluation. Furthermore, no significant differences were found between outcome groups regarding sex, age, age at epilepsy onset, or epilepsy duration. There was no association between ICU admission and patient outcomes.

4. Discussion

This study presents a comprehensive clinical characterization of adult patients with SE managed at a tertiary care center in Mexico, addressing a notable gap in the regional data. Our findings indicate that acute symptomatic etiologies are the leading causes of SEs in our setting, with convulsive SE being the predominant clinical subtype. Importantly, we found that a history of epilepsy was significantly associated with favorable functional outcomes at hospital discharge.

The predominance of acute symptomatic etiologies in our cohort aligns with previous reports from both high and low-income countries [13]. However, unlike studies from high-income regions, infections accounted for a substantial proportion of acute symptomatic SE in our population, reflecting persistent public health challenges in infectious disease control.

In Latin America, a retrospective case series by San-Juan et al. at our center focused on non-convulsive SE, where idiopathic etiology (56%) and uncontrolled epilepsy (48%) were the leading causes [13]. Although limited to NCSE, their findings complement our data and highlight the heterogeneity of SE presentations in Mexico. Notably, the mortality rate reported in San-Juan et al.'s NCSE-focused cohort was 7.3% (3 out of 41 patients), which is comparable to the 8.7% mortality observed in our SE population (9 out of 103 patients). Although prior studies from low-middle countries found increased mortality among older adults [14], our analysis did not demonstrate a significant difference by age group. In our cohort, 19 of 103 patients (18.4%) were aged ≥60 years. Among them, four individuals (21.1%) had a poor functional outcome at discharge. However, this proportion did not differ significantly from that observed in younger patients (p = 0.947), suggesting that age was not independently associated with worse outcomes in our sample. The limited sample size and potential selection bias in our study population may explain this discrepancy.

RSE occurred in 41.7% of patients (n = 43), a rate comparable to global estimates [15]. Jan Novi et al. found that RSE occurs in between 31% and 43% of patients presenting an SE episode, and almost all require coma induction for treatment [15]. Recent multicenter efforts, such as the MORSE CODE study by Chiu et al., have aimed to characterize RSE [16]. While this initiative offered valuable insights into international variations in SE management, Mexico was omitted. Despite an estimated 1.5 to 2 million people in Mexico living with epilepsy [17], published research on SE within the Mexican population remains limited. Reséndiz-Aparicio et al. conducted a multicenter epilepsy registry involving 89 hospitals across Mexico from 2021 to 2022. They analyzed 6,653 patients, of whom 12.4% had a history of SE associated with immune and infectious etiologies [18]. These findings highlight that individuals with epilepsy remain at risk of developing SE, particularly in the context of poor adherence, structural lesions, or infections. A single study by Hernandez-Dominguez et al. addressed SE in Mexican adults, involving 19 patients. The mean age was 39, and 53% of the participants were male [9]. The most frequent causes of SE in their study were structural brain abnormalities, non-specific inflammation, infections, and cryptogenic SE. Of the 17 survivors, 10 had a modified Rankin Scale of 4–5 points. Our study expands this evidence by characterizing a larger cohort of adult SE patients at a tertiary center.

A study by Murthy et al. in India offers insights into managing convulsive SE in developing countries [19]. They found acute symptomatic causes in over half of SE cases, mainly infections, and a mortality rate of 10.5%, comparable to our 8.7%. In their study, 49% of patients with acute symptomatic etiology achieved a good outcome [19]. This is similar to our finding that 51.5% of patients reached a favorable outcome (mRS 0–2), despite the shorter follow-up period. Notably, both studies underscore the challenges posed by acute symptomatic SE in resource-limited settings. In Colombia, a study of 395 SE patients identified ASM switching, mainly due to administrative barriers, as the primary trigger [20]. Similarly, in our cohort, lack of adherence or interruption of chronic ASM therapy was a major acute symptomatic trigger for SE, reflecting comparable challenges in continuity of care.

4.1. Lines of treatment and ICU admission

The most effective treatment for early SE is intravenous benzodiazepines, which can promptly control SE in about two-thirds of all patients. The most commonly used first-line treatments are lorazepam, midazolam, diazepam, and clonazepam [21]. Phenytoin/fosphenytoin, valproic acid, levetiracetam, phenobarbital, and lacosamide are commonly prescribed ASMs for treating established SE [22]. In our cohort, midazolam was the most frequently administered first-line benzodiazepine (96.1%), which is similar to reports from European and North American settings [23]. Following benzodiazepine administration, levetiracetam and valproic acid were the most frequently used second-line agents in our cohort. In contrast, lacosamide is among the most commonly used second-line agents in international cohorts [23]. This difference likely reflects variations in drug availability, institutional protocols, and cost considerations in middle-income settings. ICU admission was required for 26.2% of our patients, with a mean length of stay of 5.3 days. This aligns with previous studies in low- and middle-income countries, which report high ICU demand in SE management [19], [24], [25]. However, unlike Varelas et al. [25], we found no significant relationship between ICU admission and functional outcomes, suggesting that ICU admission in our setting may not adequately predict prognosis in SE.

4.2. Recurrence

In our cohort, 35.9% of patients had a history of SE, slightly higher than the reported recurrence rate in the literature (13–32%) [26], [27]. However, we did not observe a significant association between previous SE episodes and worse functional outcomes. This contrasts with the German study by Bauer et al., where functional outcomes worsened with each recurrence [27]. Our findings may reflect a different patient profile and warrant further investigation with longer follow-up. Although SE treatment and outcome predictors have been well studied, data on recurrent episodes remain limited, despite evidence that patients experience recurrences [28].

4.3. Strengths and limitations

Our study benefits from standardized classification systems, the latest ILAE criteria for SE, the Status Epilepticus Severity Score (STESS), and the modified Rankin Scale (mRS), to ensure consistent and reliable assessment of patient outcomes. These tools strengthened the analysis of our large cohort from a leading neurological center in Mexico. Unlike most published data from high-income countries, this study contributes novel insights from a Latin American population, where healthcare access may differ. However, as our study assesses functional outcomes only at discharge, it may not reflect long-term outcomes. This limitation can be addressed through further follow-up studies. Also, as a single-center study, the findings may not be generalizable to other populations. Also, the relatively low number of focal and non-convulsive SE cases may reflect selection bias inherent in the study setting, particularly regarding the availability of continuous EEG monitoring.

4.4. Implications and future directions

Our findings could help the implementation of treatment guidelines adapted to Latin American healthcare settings, where acute symptomatic etiologies predominate and access to healthcare may be limited.

5. Conclusions

In this cohort, acute symptomatic etiologies were the most prevalent causes of SE, with generalized convulsive SE being the predominant clinical presentation. Despite the substantial disease burden in the region, local data on severe epilepsy presentations, such as SE, are scarce. This study importantly offers valuable insights from a Latin American country, contributing to a more inclusive understanding of SE across diverse healthcare contexts.

Ethical statement

This study was approved by the Local Research and Ethics Committee of the National Institute of Neurology and Neurosurgery Manuel Velasco Suarez (NINNMVS), Mexico City. The study was conducted in accordance with the Declaration of Helsinki. Given the retrospective design, informed consent was waived. All patient data were anonymized to ensure confidentiality. This study adhered to the international STROBE guidelines for reporting observational studies.

CRediT authorship contribution statement

Irene Gómez-Oropeza: Writing – review & editing, Writing – original draft, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. María F. Castelo-Pablos: Writing – original draft, Software, Investigation. Iracema Santizo-Nanduca: Investigation, Data curation, Conceptualization. Irving Fuentes-Calvo: Validation, Software, Investigation, Formal analysis, Data curation. Daniela Deustúa-Hernández: Writing – original draft, Investigation. Oscar Esquivel-Zapata: Writing – original draft, Methodology, Formal analysis, Conceptualization. Salvador Martínez-Medina: Writing – original draft, Conceptualization. Mariana Peschard-Franco: Investigation, Conceptualization. Pilar Robles-Lomelin: . Iris E. Martínez-Juárez: Visualization, Validation, Supervision, Conceptualization. Ashuin Kammar- García: Supervision, Methodology, Investigation. Claudia Cervera-Martínez: Investigation. Valeria I. Bravo-Osorno: Conceptualization. Gabriel Manrique-Gutierrez: Investigation. Erika Palacios Rosas: Supervision, Methodology, Conceptualization. Juan C López-Hernández: Visualization, Validation, Supervision. Anwar Santos-García: Visualization, Supervision, Investigation. Raúl Medina-Rioja: Validation, Supervision, Conceptualization. Sonia I Mejia-Peréz: Writing – review & editing, Supervision, Methodology, Conceptualization. Elma Paredes-Aragón: Writing – review & editing, Visualization, Validation, Supervision, Methodology, Investigation, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: