Ciencia y Salud, Vol. 9, No. 1, enero-marzo, 2025 • ISSN (impreso): 0378-7680 • ISSN (en línea): 2613-8751

LEUKO-GLYCEMIC INDEX AS A PREDICTOR OF ACUTE INTRAHOSPITAL COMPLICATIONS IN ADULTS WITH ISCHEMIC STROKE IN THE SALVADOR BIENVENIDO GAUTIER HOSPITAL IN THE DOMINICAN REPUBLIC

Índice leucoglucémico como predictor de complicaciones intrahospitalarias agudas en adultos con enfermedad vasculocerebral isquémica en el Hospital Salvador Bienvenido Gautier de República Dominicana

DOI: https://doi.org/10.22206/cysa.2025.v9i1.3105

Recibido: 5 de abril, 2024 • Aceptado: 14 de abril, 2024

Cómo citar: Francisco Acevedo, A. A., Alcantara de Oleo, M. & Rijo Reyes, M. (2025). Leuko-glycemic Index as a Predictor of Acute Intrahospital Complications in Adults with Ischemic Stroke in the Salvador Bienvenido Gautier Hospital in the Dominican Republic. Ciencia y Salud, 9(1), 19-29. https://doi.org/10.22206/cysa.2025.v9i1.3105

Introduction

Projecting its impact on a global level, ischemic stroke assumes a crucial role in the incidence of morbidity, mortality, and widespread disability¹. In an unexpected revelation from the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2019 Stroke Collaborators, an overwhelming 12.2 million incident cases, 101 million prevalent reports, 143 million disability-adjusted life-years (DALYs), and 6.55 million deaths were attributed to stroke². Indeed, in 2022 nearly 7 million Americans older than 20-year-old reported having had one³. In the Caribbean region, the Dominican Republic is positioned as the third-ranking country in ischemic stroke mortality. The daily incidence is 59 cases per day, extrapolated to 583 weekly reports and 2,333 cases per month, with associated hospital care costs estimated at DOP 700,000.00 (USD 11,967.40)^{4, 5}.

Moreover, hyperglycemia and leukocytosis assessed individually are considered markers of poor prognosis in patients with ischemic stroke^6-8. For this reason, the leuko-glycemic index (LGI) could represent a valuable predictor of complications since it prognosticates in-hospital aggravations and mortality of ischemic etiology combining the inflammatory and metabolic response^9-12.

Previously, Asmat-Vasquez and collaborators reported that 56% of the deceased occurrences due to complications exhibited an LGI exceeding 1,600, with aspiration pneumonia as the primary aggravation in 67% of the cases¹². In addition, Gonzalez and colleagues established the LGI as a robust predictor of complications, identifying dysphagia and pneumonia as common ones¹⁰. However, despite the substantial predictive capacity of the LGI for aggravations, minimal emphasis has been placed on illustrating its efficacy in a Caribbean population. Due to this fact, applying the LGI as an instrument allows for identifying and stratifying high-risk populations for complications.

Methods

Study design

An observational, descriptive, cross-sectional, and retrospective study was conducted at the neurology department of Salvador Bienvenido Gautier Hospital. This community hospital is located in Santo Domingo, the capital of the Dominican Republic.

The primary objective of the investigation was to determine the LGI as a predictor of complications in ischemic stroke in over 18-year-old patients in a hospital in the Dominican Republic. Moreover, the secondary objectives were to establish the relationship of age and sex with the presentation of complications, to specify toxic habits and to list comorbidities of complicated cases, to record leukocyte count and blood glucose, to indicate the level of consciousness according to the Glasgow Coma Scale at the admission of complicated patients, to distinguish the location of the ischemic lesion, to identify the presence of complications and group them, and to classify patients according to their discharge condition.

The study was performed between December 7^th, 2022, and June 5^th, 2023. The Salvador Bienvenido Gautier Hospital Ethics Committee submitted and approved the research protocol. The data was collected by the authors and a contributor (the name was mentioned in the acknowledgment section), and was analyzed and interpreted by the authors.

Participants

The population was constituted of 412 patients who were admitted to the clinical room of the Salvador Bienvenido Gautier Hospital between July and December 2022. The sample consisted of 161 patients.

The inclusion criteria were all patients over 18 years old who were admitted to the clinical ward of the Salvador Bienvenido Gautier Hospital in the period from July to December 2022 with a confirmed diagnosis of ischemic stroke by neuroimaging, as well as with a leukogram and blood glucose reported within the first 72 hours of admission. Nevertheless, the exclusion criteria were pregnant or puerperal patients, those with a documented diagnosis of hemorrhagic stroke, a systemic infection in the 7 days before the stroke, a hematological disease, an autoimmune disease or immunodeficiencies, some type of cancer or in remission, and those who have consumed corticosteroid treatment in the previous 6 months.

Data collection

The data were obtained through the clinical records reviewed and those were registered using a digital form consisting of 5 sections constituted by 9 multiple-choice questions, and 3 open questions applying the KoboToolbox software, which addressed epidemiological factors such as age, sex, toxic habits, and comorbidities, room analysis and neuroimaging reports, staging, and evolution.

Variables

The dependent variable of the study was the leuko-glycemic index. Whereas, the independent variables were age, sex, toxic habits, comorbidities, leukocyte count, glycemia, level of consciousness, location of the lesion, presence of complications, complications, and discharge condition.

Data processing and analysis

The processing of the information was systematic, using an HP® 15 laptop with Windows 11, the statistical package of Microsoft Excel 2019, and the KoboToolbox software. The tables were processed through Microsoft Word 2019 and graphs were generated using GraphPad Prism version 10.1.2.

Subsequently, we derived the Leuko-Glycemic Index (LGI) for each case within our sample using the following formula¹³.

LGI = (glycemia [mg/dL] · leukocyte count [10⁶/L]) / 1,000

To determine the cut-off value of the LGI of the sample, the Receiver Operating Characteristic (ROC) curve was used, evaluating the Area Under the Curve (AUC) of the sample, which yielded a sensitivity of 66%, a specificity of 69% and an AUC of 0.691 with a standard error of 0.043, and a 95% confidence interval of 0.606 to 0.777 (Figure 1).

From then on, the dichotomy of complicated and uncomplicated cases resulted in a very close sensitivity and specificity, to discern the cut-off point the Youden index was used, following the formula (sensitivity + specificity - 1)¹⁴, extrapolated to:

Youden index = (0.66 + 0.69) – 1 = 0.35

Thence, this allowed us to determine by way of the EPITOOLS software, that the cut-off value of the leuko-glycemic index for the sample was 1,566.

Limitations of the study

The restraints of this investigation were 187 incomplete clinical records that were found by the researchers confining the sample selection. Moreover, and related to the used methodology, non-probability sampling for convenience was used, which could potentially introduce bias and restrict the applicability of the findings. Also, the study period, which was from July to December 2022, might not have captured seasonal variations or long-term trends in ischemic stroke complications.

Results

The study findings indicate that 44.72% (n=72) of cases surpassed the predetermined LGI threshold of 1,566. Notably, among 39.75% (n=64) patients who experienced complications, a substantial 65.62% (n=42) exhibited a heightened LGI, having a linked arithmetic mean of 2,328.39 and an associated standard deviation of 1,530.85. In contrast, in non-complicated patients, 60.24% (n=97) demonstrated a significantly lower LGI, evidenced by an arithmetic mean of 1,475.67 with a corresponding standard deviation of 945.23 (Table 1) (Figure 2).

Furthermore, in the cohort of individuals who experienced complications, a noteworthy 32.81% (n=21) belonged to the age group of 66 to 75 years old, 39.06% (n=25) were female while 60.94% (n=39) were male. Among complicated patients, 39.06% (n=25) presented toxic habits, with smoking, coffee, and tisanes being predominant in 72% (n=18), followed by alcoholism in 40% (n=10). On the other hand, comorbidities were prevalent in 98.43% (n=63), with systemic arterial hypertension having the greatest impact in 88.89% (n=56), pursued by diabetes mellitus in 61.90% (n=39) (Table 2).

From the observation of the sociodemographic variables, we were able to reveal that 65.63% (n=42) of the complicated patients had a high LGI; of whom 54.76% (n=23) were male, 64.29% (n=27) presented toxic habits, 97.62% (n=41) suffered from comorbidities, standing out in 76.19% (n=32) of them diabetes mellitus. Regarding the age group, 4.76% (n=2) were 36-45 years old, 7.14% (n=3) were 46-55 years old, 28.57% (n=12) were 56-65 years old, 28.57% (n=12) were 66-75 years old, 19.05% (n=8) were 76-85 years old, and 11.9% (n=5) were over 85 years old. With this, we demonstrate that elevated LGIs occur in a greater proportion of male patients, diabetics, and consumers of toxic habits, observing a decrease in LGI levels at both ends of these age groups.

Regarding leukocyte count, complicated patients exhibited an arithmetic mean of 10.80 × 10⁶/L, while glycemia resulted in a mean of 210.25 mg/dL. In contrast, non-complicated patients had an arithmetic mean leukocyte count of 8.32 × 10⁶/L, and a mean of 176.1 mg/dL in their glycemia (Table 3).

In assessing the level of consciousness upon admission using the Glasgow Coma Scale (GCS) among patients who developed complications, 51.56% (n=33) showed a mild alteration, 28.12% (n=18) a moderate alteration, and 20.31% (n=13) a severe alteration. Regarding lesion location, a total of 23.60% (n=38) exposed multifocal lesions. Notably, the frontal lobe was identified as the most affected for 11.18% (n=18) of cases (Table 4).

Moreover, in the 42 complicated patients that elevated the LGI, 59.52% (n=25) of them presented a mild level, 21.43% (n=9) a moderate level, and 19.05% (n=8) a severe level; accompanied by a multifocal location and a greater involvement in the frontal lobe territory. However, of the 22 patients who succumbed to complications, 45.45% (n=10) of them were admitted with a severe level of consciousness according to the GCS, 36.36% (n=8) with a moderate level and 18.18% (n=4) with a mild level. Therefore, we denote that the location of the lesion influences the level of consciousness according to the GCS if it is combined with comorbidities, toxic habits, and middle and later adulthood, leading to the appearance of fatal complications for patients.

Likewise, 39.75% (n=64) manifested complications while 60.25% (n=97) did not. Among the reported aggravations, 39.06% (n=25) developed aspiration pneumonia, followed by structural epilepsy in 23.43% (n=15) and urinary tract infection in 18.75% (n=12). Additionally, 62.50% (n=40) of patients presented other diverse complications such as cerebral edema, electrolyte imbalance, acid-base disorders, acute kidney disease, and others (Table 5) (Figure 3). Furthermore, in terms of the patients’ outcomes, 86.33% (n=139) were discharged alive, while 13.66% (n=22) succumbed due to complications (Table 5).

Discussion

The examination of the LGI in our study has revealed compelling insights into its potential as a predictive biomarker for complications in the studied population. A determined LGI cut-off point of 1,566, surpassed by a substantial portion of complicated cases, highlights its discriminative capability. Similarly, Gonzalez and colleagues found that complications increased with higher LGI values, especially beyond the 1,021 cut-off point¹⁰.

Moreover, our examination of demographic factors showcased a notable portion of complicated cases’ age distribution in the 66 to 75-year-old. Additionally, gender composition revealed a higher occurrence of aggravations in males compared to females. These align with the findings of Hernandez-Sanchez and collaborators, who reported a mean age of 71.4 years old, predominantly men; and with Goldstein, who estimated a 9% 30-day stroke mortality rate due to complications for 65-74 years old^{15, 16}.

The inspection of toxic habits and comorbidities among complicated patients further enhances our understanding of the multifactorial nature of complications. Smoking emerged as a predominant toxic habit while co-occurring diseases such as systemic arterial hypertension, diabetes mellitus, and previous stroke had a substantial impact. This correlates with the study conducted by Asmat-Vasquez, which noted that the most prevalent associated conditions were systemic arterial hypertension and diabetes mellitus¹².

The investigation into leukocyte count and glycemia provided significant differences in these parameters reinforcing their conjugated potential as objective indicators of condition severity. Nevertheless, as reported by Gonzalez et al., the mean leukocyte count and glycemia were 9.36 × 10⁶/L and 123.50 mg/dL, respectively¹⁰.

Likewise, complicated patients showed mild alteration in the level of consciousness and multifocal lesions with a predominance in the frontal lobe, contributing to the complexity of the clinical presentation. However, Asmat-Vasquez and collaborators, reported 61% presenting a moderate level of consciousness upon admission¹². This observed phenomenon may be attributed to the time gap between the identification of the condition by family members and the subsequent admission to the hospital¹⁶. Correspondingly, Ropper and associates, underline the frontal lobe as most affected in convoluted scenarios¹⁷.

Finally, outcomes analysis revealed that most patients were discharged alive, suggesting a favorable prognosis for the majority. However, a portion of patients experienced complications, particularly aspiration pneumonia. In contrast, Zarranz and colleagues reported complication rates ranging from 60% to 95%¹⁸, this could be associated with the growing aging trend of the population¹⁹. According to Chamorro, early post-stroke infections often result from induced immunosuppression due to brain parenchyma damage²⁰, and Asmat-Vasquez et al. identified pneumonia as the primary complication in about 67% of cases, and 93.59% of cases being discharged alive¹².

Conclusions

This research aimed to prove that LGI can effectively predict complications in ischemic stroke. Despite facing difficulties in gathering information from 187 incomplete clinical records, acknowledging the predictive potential of LGI enables healthcare providers to conduct timely risk assessments and interventions, empowering them to proactively manage stroke aggravations and improve patient outcomes.

Based on these outcomes, practitioners should emphasize the necessity of smoking cessation, adopting a Mediterranean diet, promoting physical exercise, and ensuring stringent medical control of comorbidities. Identifying predominant complications, like aspiration pneumonia, underscores the need for early identification, close monitoring, and preventive measures, including an upright posture during meals and sleep, a full liquid diet, speech therapy, avoiding solid oral medication, and caregiver education. To better understand the implications of these results, future studies could delve into widening the sample size, orchestrating multicenter investigations concentrated on high-risk populations, and integrating the National Institute of Health Stroke Scale (NIHSS) as a compass to navigate the intricate landscape of stroke impact, facilitating strategic post-acute care.

Acknowledgments

We are infinitely grateful to the director and the Department of Neurology of the Salvador Bienvenido Gautier Hospital. We want to thank Dr. Anderson Tavarez Gonzalez for his valuable contribution as a data collector in this research.

Authors’ contribution

The authors confirm their contribution to the paper as follows: conception and design of the study: Alarice Francisco, Mabel Alcantara y Misael Rijo; Data Collection: Alarice Francisco, Mabel Alcantara y Misael Rijo, analysis and interpretation of results: Alarice Francisco, Mabel Alcantara y Misael Rijo; preliminary manuscript preparation: Alarice Francisco. All authors reviewed the results and approved the final version of the manuscript.

Ethical considerations

The ethics committee of the study hospital approved the investigation. In the same way, the present project was governed under the principles of privacy and confidentiality promulgated in the Declaration of Helsinki of 1964 and its revision of 1975. Likewise, it was based on the bioethical principles of non-maleficence, beneficence, and justice. No conflicts of interest were declared by the authors.

References

1. Choreno-Parra JA, Carnalla-Cortes M, Guadarrama-Ortiz P. Ischemic cerebrovascular disease: extensive review of the literature for the first contact physician. Med Int Mex. 2019;35(1):61-79. Available from: https://doi.org/10.24245/mim.v35i1.2212.

2. GBD 2019 Stroke Collaborators. Global, regional, and national burden of stroke and its risk factors, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Neurol. 2021;20(10):795–820. Available from: https://doi.org/10.1016/S1474-4422(21)00252-0

3. Smith WS, Claiborne Johnston S, Hemphill JC. Introduction to Cerebrovascular Diseases. In: Loscalzo J, Kasper DL, Dongo DL, Fauci AS, Hauser SL, Jameson JL, editors. Harrison’s Principles of Internal Medicine. 21st ed. Boston: McGraw Hill; 2022. p. 12,090-12,123.

4. Reyes N. They register 59 strokes every day in the country [Internet]. Listin Diario; 2023 May [cited 2024 Jan 12]. Available from: https://listindiario.com/la-republica/sector-salud/20230515/registran-59-accidentes-cerebrovasculares-dia-pais_753463.html

5. Central Bank of the Dominican Republic. Dollar exchange rate history [Internet]. Central Bank of the Dominican Republic; 2024 Jan [cited 2024 Jan 12]. Available from: https://www.bancentral.gov.do/SectorExterno/HistoricoTasas

6. Cuesta-Mero PL, Garcia-Romo LA, Villacres-Lopez AD. Leukoglucemic index as a prognostic marker for complications in patients with acute myocardial infarction. Rev Colomb Cardiol. 2021;28(2):119-127. Available from: https://doi.org/10.24875/rccar.m21000023

7. Reyes-Villarreal JC, Morales-Gutierrez JE, Lopez-Zamora B, Reyes-Navarro GV, Ordonez-Gonzalez I, Triana-Gonzalez S, et al. Leuko-glycemic index in patients with ST elevation myocardial infarction. Rev Med Inst Mex Seguro Soc. 2022;60(2):142-8.

8. Rodriguez Jimenez A, Fardales Rodriguez R, Toledo Rodriguez E, Quintana Canizarez G. Leuko-glycemic Index as a Prognostic Factor after Acute Myocardial Infarction with ST-segment Elevation. Finlay. 2019;9(2):1-10.

9. Ramos-Hernandez WM, Soto LF, Del Rosario-Trinidad M, Farfan-Morales CN, De Jesus-Gonzalez LA, Martinez-Mier G, et al. Leukocyte glucose index as a novel biomarker for COVID-19 severity. New York: Sci Rep; 2022. Available from: https://doi.org/10.1038/s41598-022-18786-5

10. Gonzalez FM, Cutiller SA, Paulin F, Paz Rivero M, Ragusa MA, Marovelli L, et al. Leuko-glycemic index as a preacher of complications in patients with ischemical stroke. Neurologia Argentina. 2020 Jun;12(3):159-164. Available from: https://doi.org/10.1016/j.neuarg.2020.04.001

11. Garcia Alvarez PJ, Garcia Albero AP, Santana Alvarez CJ. Leukoglycaemic index as a predictor in short term of mortality in stroke. Arch. med. Camagüey [Internet]. 2018 Apr [cited 2024 Jan 14]; 22(2):163-170. Available from: http://scielo.sld.cu/scielo.php?script=sci_arttext&pid=S1025-02552018000200006&lng=es

12. Asmat-Vasquez JCM, Diaz-Paz K. High Leuko-Glycemic Index As An In-Hospital Mortality Predictor In Patients With Acute Ischemical Ictus. Rev Ecuat Neurol. 2021 Sept;30(2):18-25. Available from: https://doi.org/10.46997/revecuatneurol30200018

13. Padilla-Cueto D, Hernandez-Negrin H, Ramirez-Gomez JI, Perez-Valdivia A, Cardenas-Sanchez AL, Alfonso-Izquierdo A. The leucoglycaemic index is a predictor of all-cause mortality per year in Cuban patients with ST-segment elevation acute myocardial infarction. CorSalud [Internet]. 2019 Jan-Mar [cited 2024 Jan 15];11(1):21-29. Available from: http://scielo.sld.cu/scielo.php?script=sci_arttext&pid=S2078-71702019000100021&lng=es

14. Sahngun Nahm F. Receiver operating characteristic curve: overview and practical use for clinicians. Korean J Anesthesiol. 2022 Jan 18;75(1):25-36. Available from: https://doi.org/10.4097/kja.21209

15. Hernandez-Sanchez MM, Lozano-Nuevo JJ, Suarez-Cuenca JA, Mendoza-Portillo E, Vergara-Orduna FI, Palacios-Perez E, et al. Leuco-glycemic index associated to complications in atherothrombotic cerebral ischemia. Med Int Mex. 2019 Feb;35(1):39-44. Available from: https://doi.org/10.24245/mim.v35i1.2180

16. Goldstein LB. Ischemic cerebrovascular disease. In: Goldman L, Schafr AI, editors. GOLDMAN-CECIL MEDICINE. 26th ed. Philadelphia: Elsevier; 2020. p. 2,396-2,407.

17. Ropper AH, Samuels MA, Klein JP, Prasad S. Stroke and Cerebrovascular Diseases. In: Ropper AH, Samuels MA, Klein JP, Prasad S, editors. Adams and Victor’s Principles of Neurology. 11th ed. New York: McGraw-Hill Education; 2019. p. 798-905.

18. Castillo J, Luna A, Rodriguez-Yanez M, Ugarriza I, Zarranz JJ. Cerebral vascular diseases. In: Zarranz JJ, editor. Neurology. 6th ed. Barcelona: Elsevier; 2018. p. 301-357.

19. Soto A, Guillen-Grima F, Morales G, Munoz S, Aguinaga-Ontoso I, Fuentes-Aspe R. Prevalence and incidence of ictus in Europe: systematic review and meta-analysis. An Sist Sanit Navar. 2022 Apr;45(1): e0979. Available from: https://doi.org/10.23938/ASSN.0979

20. Chamorro Sanchez A. Cerebral vascular accidents. In: Rozman C, Cardellach F, editors. Farreraz Rozman’s Internal Medicine. 19^th ed. Barcelona: Elsevier; 2020. p. 1375-1388.

_______________________________

¹ MD. Department of Neurology, Salvador Bienvenido Gautier Hospital, Santo Domingo, Dominican Republic. ORCID: https://orcid.org/0000-0003-0153-0903. Email: alariceafrancisco@gmail.com

² MD. Department of Neurology, Salvador Bienvenido Gautier Hospital, Santo Domingo, Dominican Republic. ORCID: https://orcid.org/0009-0005-8021-4467. Email: 100389470ma@gmail.com

³ MD. Department of Neurology, Salvador Bienvenido Gautier Hospital, Santo Domingo, Dominican Republic. ORCID: https://orcid.org/0009-0004-2532-4658. Email: misaelrijoreyes@gmail.com