Effect of Nutritional Status on Coagulation and Inflammation in Patients With Cirrhosis and Variceal Bleeding

Abstract

Background

Malnutrition is a serious concern in patients with liver cirrhosis that is complicated by acute variceal bleeding due to portal hypertension. The Controlling Nutritional Status (CONUT) score is a screening tool used to evaluate altered nutritional status and predict adverse outcomes in patient populations. This study aimed to assess nutritional status using the CONUT score in 151 patients with cirrhosis and esophageal-gastric variceal bleeding (EVB).

Material/Methods

Clinical data from 151 patients with cirrhotic EVB admitted between January 2023 and October 2024 were analyzed. Nutritional status was assessed using the CONUT score. General linear model analysis was used to explore associations between nutritional, coagulation, and inflammatory indices.

Results

Based on CONUT scores, 2.6% (n=4) of patients had normal nutrition, 15.9% (n=24) had mild malnutrition, and 81.5% (n=123) had moderate-to-severe malnutrition. Patients showed significant coagulation impairment, including prolonged prothrombin time (PT) and international normalized ratio (INR), reduced prothrombin activity (PTA), and elevated D-dimer (all P<0.05). Inflammatory markers, including interleukin-6 (IL-6), were also significantly elevated (P<0.05). CONUT scores were positively correlated with PT (rs=0.508, P<0.05), INR (rs=0.515, P<0.05), and IL-6 (rs=0.211, P<0.05) and negatively correlated with PTA (rs=−0.432, P<0.05). General linear model analysis identified PT (OR=1.214), INR (OR=0.172), D-dimer (OR=3.460), and IL-6 (OR=1.439) as independent risk factors for malnutrition (all P<0.05).

Conclusions

Moderate-to-severe malnutrition is highly prevalent in patients with cirrhotic EVB and is independently associated with coagulation dysfunction and systemic inflammation. These findings highlight the need for strengthened nutritional monitoring and individualized interventions to improve patient prognosis.

Introduction

Esophageal-gastric variceal bleeding (EVB) is a life-threatening complication of liver cirrhosis, with an annual incidence of 10% to 15% and a 6-week mortality rate of 20% to 33% after the first bleed [1,2]. A major factor influencing this poor prognosis is malnutrition, which is highly prevalent in cirrhosis, affecting 50% to 90% of patients [3,4]. Malnutrition in this context arises from factors like portal hypertensive enteropathy, impaired hepatic protein synthesis, and dietary restrictions after bleeding [5,6]. This nutritional deficit contributes to clinical deterioration by promoting sarcopenia and immune dysfunction, which can exacerbate portal hypertension and variceal vulnerability [7].

The Controlling Nutritional Status (CONUT) score is an objective screening tool derived from routine blood tests for serum albumin level, total cholesterol level, and total lymphocyte count [8]. It was originally developed and validated for detecting undernutrition in hospitalized patients and has since been applied in various clinical settings, including cirrhosis, to predict clinical outcomes [8,9]. Its advantage lies in its ability to provide a comprehensive assessment of protein reserves and caloric depletion.

In cirrhosis, the interplay between malnutrition, inflammation, and coagulation is critical yet not fully elucidated. Malnutrition can accelerate muscle loss, disrupt intestinal barriers, and exacerbate endotoxemia, leading to Kupffer cell activation and the release of pro-inflammatory cytokines, such as interleukin-6 (IL-6) [10–12]. This inflammatory state, in turn, can inhibit hepatic synthesis of clotting factors and albumin, creating a vicious cycle that worsens coagulation dysfunction and nutritional status [13,14]. While some studies have established a link between hypoalbuminemia and prolonged prothrombin time (PT) [15,16], and others have used the CONUT score to predict outcomes in patients with cirrhosis [17], there is a lack of systematic analysis focusing specifically on the EVB population. Specifically, the relationship between a comprehensive nutritional score like CONUT and key parameters of coagulation, such as international normalized ratio (INR) and D-dimer, and inflammation, such as IL-6, in this high-risk group remains poorly defined, limiting the development of targeted interventions.

Therefore, this study aimed to assess nutritional status using the CONUT score in 151 patients with cirrhosis and bleeding esophagogastric varices, to quantify its correlations with coagulation and inflammatory markers, and to identify independent risk factors for malnutrition among these parameters.

Material and Methods

Ethics Statement

This study was reviewed and approved by the Medical Ethics Committee of the Second Affiliated Hospital of Xi’an Jiaotong University, China (approval No. 2024–148). All procedures were conducted in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments. Given that some patients with cirrhosis and EVB may have impaired consciousness or difficulty completing written consent due to acute bleeding, verbal informed consent was obtained from all participants (or their legally authorized representatives for incapacitated patients).

Study Population

Inclusion and Exclusion Criteria

The inclusion criteria were as follows: patients aged 18 to 80 years admitted to the Second Affiliated Hospital of Xi’an Jiaotong University between January 2023 and October 2024, with a confirmed diagnosis of liver cirrhosis (per clinical, imaging, and laboratory criteria) and EVB (diagnosed via upper gastrointestinal endoscopy) [18]. The exclusion criteria were as follows: (1) severe extrahepatic diseases (eg, end-stage renal failure, malignant tumors, severe heart failure); (2) use of coagulation-modifying drugs (eg, warfarin, low-molecular-weight heparin) within 1 month before admission; (3) hematological diseases affecting coagulation (eg, idiopathic thrombocytopenic purpura, hemophilia); (4) acute infection (body temperature >38.5°C, white blood cell count >12×10⁹/L) at admission; and (5) pregnancy or lactation.

Sample Size Justification

The study focused on analyzing associations between nutritional status, coagulation indices, and inflammatory markers. Sample size was determined based on 2 criteria: (1) power analysis using GPower 3.1 software assuming a medium effect size (r=0.3), significance level (α=0.05), and statistical power (1-β=0.8), the minimum required sample size was 85; and (2) recommendation for observational studies, whereby the effective sample size should be 5 to 10 times the number of independent variables [19]. This study included 6 core variables (PT, INR, D-dimer, IL-6, neutrophil-to-lymphocyte ratio [NLR], platelet count), requiring a minimum sample size of 30 to 60 patients. To ensure sufficient statistical power, a total of 151 eligible patients were consecutively enrolled (exceeding the minimum requirements of both criteria).

Randomization, Blinding, and Bias Mitigation

Randomization was irrelevant for this cross-sectional study, as patients were recruited by admission and eligibility. Laboratory technicians were blinded to patients’ clinical details to reduce measurement bias, but clinicians could not be. Selection bias from the single-center design was mitigated by consecutive recruitment and comparing baseline data with published studies. Recall bias was minimal (relying on clinical data), and measurement bias was reduced via standardized protocols and regular analyzer calibration.

Data and Sample Collection

Clinical Data Collection

Demographic (age, sex) and clinical data (etiology of cirrhosis, history of hypertension/diabetes) were extracted from electronic medical records. Laboratory data included coagulation parameters, inflammatory markers, and nutritional indices (serum albumin, total cholesterol, lymphocyte count) measured within 24 hours of admission.

Blood Sample Collection and Processing

All patients fasted for 8 or more hours overnight, and 5 mL of fasting venous blood was collected between 7: 00 and 9: 00 AM by trained nurses using sterile venipuncture. Blood samples for coagulation function tests were collected into vacuum tubes (Shanghai Kehua Bio-Engineering Co, Ltd, China) containing 3.2% sodium citrate. Samples designated for inflammatory and immune indices analysis were collected into vacuum tubes containing EDTA-K2 (Shanghai Kehua Bio-Engineering Co, Ltd, China). After collection, all blood samples were immediately centrifuged at 3000 rpm for 15 minutes at 4°C using a Centrifuge 5427 R (Eppendorf, Germany). Plasma (from citrate tubes) and serum (from EDTA tubes after centrifugation) were separated within 30 minutes of collection and aliquoted into cryogenic vials (Corning, USA). All aliquots were stored at −80°C in a U410 Premium ultra-low temperature freezer (Eppendorf, Germany) within 1 hour of collection for subsequent batch analysis.

Laboratory Detection Methods

Coagulation Function Detection

Coagulation parameters, including PT, activated partial thromboplastin time (APTT), INR, prothrombin activity (PTA), and D-dimer, were assayed using a Sysmex CS-5100 automated coagulation analyzer (Sysmex, Japan). With the exception of the D-dimer reagent from SEKISUI Medical Co, Ltd, Japan, all supporting reagents and parameters were supplied by Siemens Healthineers, Germany. The International Sensitivity Index (ISI) is approximately 1. The INR was calculated by raising the ratio of the patient’s PT to the normal control PT to the power of the ISI: INR=(patient PT/normal control PT)^ISI. The analyzer was calibrated daily with commercial quality control materials, and both intra- and inter-assay coefficients of variation were maintained below 5% for all parameters.

Inflammatory and Immune Biomarker Analysis

According to the manufacturer’s instructions, serum IL-6 levels were quantitatively determined using a chemiluminescent immunoassay analyzer (IMMULITE 2000 XPi, Siemens) and matching kits (Siemens, Germany). Serum procalcitonin (PCT) levels were quantified by chemiluminescent immunoassay on a Wan200+ automated analyzer (UMIC, China), using dedicated reagent kits provided by Xiamen Wantai Kairui Biotechnology Co, Ltd, China. Neutrophil and lymphocyte counts were performed on a XN-9000 automated hematology analyzer (Sysmex, Japan) using the manufacturer’s reagents. NLR was calculated as the ratio of absolute neutrophil count to absolute lymphocyte count. The systemic immune-inflammation index (SII) was calculated using the following formula [20]: SII=(absolute neutrophil count×absolute platelet count)/absolute lymphocyte count, where platelet count was also measured via the XN-9000 automated hematology analyzer (Sysmex, Japan) with an impedance method.

Nutritional Status Evaluation

Nutritional risk stratification in this study was based on the patients’ CONUT scores [8]. The specific criteria are shown in Table 1.

Table 1.

Criteria for Controlling Nutritional Status (CONUT) nutritional risk stratification according to scores: normal: 0–1; mild malnutrition: 2–4; moderate malnutrition: 5–8; severe malnutrition: 9–12.

Indicator (unit)	Scoring criteria
Serum albumin (g/L)	≥35 (0); 30.0–34.9 (2); 25.0–29.9 (4); <25.0 (6)
Total cholesterol (mg/dL)	≥180 (0); 140–179 (1); 100–139 (2); <100 (3)
Lymphocyte count (×109/L)	≥1.6 (0); 1.2–1.5 (1); 0.8–1.1 (2); <0.8 (3)

Statistical Analysis

Statistical analysis was performed using SPSS 20.0 software (IBM Corp, Armonk, NY, USA). Variables were first tested for normality using the Shapiro-Wilk test. Normally distributed continuous variables (eg, age) are expressed as mean±standard deviation. Non-normally distributed continuous variables (eg, PT, IL-6) are expressed as median (25^th–75^th percentiles); Categorical variables (eg, sex, nutritional status) are expressed as number of cases and percentage (n,%). Spearman correlation analysis was used to measure the relationships between CONUT scores and coagulation and inflammatory indices, with correlation coefficients (rs) and 95% CIs reported. General linear model analysis was used to identify independent risk factors for malnutrition, with odds ratios and 95% CIs calculated. Variables with P<0.10 in univariate analysis were included in the multivariate general linear model. A 2-tailed P value <0.05 was considered statistically significant.

Results

Nutritional Status Distribution Assessed by CONUT Score

Among the 151 enrolled patients with cirrhosis and acute EVB, nutritional status was categorized using the CONUT score. The distribution showed a high prevalence of malnutrition: 4 cases (2.6%) were classified as normal, 24 (15.9%) with mild malnutrition, 72 (47.7%) with moderate malnutrition, and 51(33.8%) with severe malnutrition (Table 2). This distribution underscores the high prevalence of nutritional impairment in this cohort.

Table 2.

Nutritional status stratification by Controlling Nutritional Status (CONUT) score in patients with cirrhosis and acute esophageal variceal bleeding.

Nutritional status	Cases	Percentage (%)
Normal	4	2.6%
Mild malnutrition	24	15.9%
Moderate malnutrition	72	47.7%
Severe malnutrition	51	33.8%

Baseline Coagulation and Inflammatory Profiles

Baseline assessments were conducted to characterize the coagulation and inflammatory status of patients with cirrhosis and acute EVB, which was a key focus of the study. As presented in Table 3, the cohort exhibited significant abnormalities in coagulation indices and inflammatory markers.

Table 3.

Baseline coagulation and inflammatory marker levels in the study population.

Index	N	Median	P25	P75
Coagulation indices
PT	151	14.45	12.80	16.70
INR	151	1.32	1.17	1.51
PTA	151	57.20	47.63	71.05
APTT	151	37.60	31.30	45.60
D-dimer	151	1530.00	740.00	4955.00
Inflammatory-related indices
PCT	151	0.19	0.09	0.51
IL-6	151	23.06	9.60	49.16
SII	151	386.69	187.34	914.24
NLR	151	4.17	2.66	8.86
Neutrophil count	151	3.32	1.88	5.62
PLT	151	84.00	59.25	141.50
Lymphocyte count	151	0.73	0.52	1.06

PT – prothrombin time; INR – international normalized ratio; PTA – prothrombin activity; APTT – activated partial thromboplastin time; PCT – procalcitonin; IL-6 – interleukin-6; SII – absolute neutrophil count×absolute platelet count/absolute lymphocyte count; NLR – neutrophil-to-lymphocyte ratio; PLT – platelet count; P25 – 25^th percentile; P75 – 25^th percentile.

Coagulation Indices

The cohort exhibited prolonged PT (median 14.45 seconds) and INR (median 1.32), reduced PTA (median 57.2%), elevated D-dimer (median 1530 ng/mL), prolonged APTT (median 37.6 seconds), and decreased platelet count (median 84×10⁹/L).

Inflammatory Markers

Levels of IL-6 (median 23.06 pg/mL), NLR (median 4.17), SII (median 386.69), and PCT (median 0.19 ng/mL) were elevated (Table 3). These findings confirm the presence of significant coagulation dysfunction and systemic inflammation in patients with cirrhosis and EVB, providing a foundational basis for subsequent analyses of the associations between nutritional status and these 2 pathways.

Correlations Between CONUT Score and Coagulation and Inflammatory Indices

Bivariate correlation analyses were performed to explore the relationships between the CONUT score and key parameters.

Correlations Between CONUT score and Coagulation Indices

The CONUT score showed strong positive correlations with PT (rs=0.508), INR (rs=0.515), APTT (rs=0.388), and D-dimer (r s=0.370) (all P<0.01). It was negatively correlated with PTA (rs=−0.432) and platelet count (rs=−0.286) (both P<0.01).

Correlations Between CONUT Score and Inflammatory Indices

Positive correlations were observed between the CONUT score and PCT (rs=0.174, P<0.05), IL-6 (rs=0.211, P<0.05), and NLR (rs=0.275, P<0.01). A negative correlation was found with lymphocyte count (rs=−0.405, P<0.01). No association was found with neutrophil count (P>0.05) (Table 4). These results indicate that a higher CONUT score (worse nutrition) was associated with more severe coagulation impairment and inflammatory activation. A pairwise correlation analysis, visualized in a heatmap, further confirmed that the CONUT score showed strong positive correlations with coagulation and inflammatory indices, with deeper color intensity indicating a stronger correlation (Figure 1).

Table 4.

Correlation analysis between Controlling Nutritional Status (CONUT) score and coagulation and inflammatory indices. CONUT scores were positively correlated with coagulation indicators (PT, INR, APTT, and D-dimer) and inflammatory indicators (PCT, IL-6, and NLR).

Screening tool	Variable	Spearman	P
		Rs (95% CI)
	PT	.508 (0.365, 0.626)	0.000
	INR	.515 (0.377, 0.631)	0.000
	PTA	−.432 (−0.567, −0.269)	0.000
	APTT	.388 (0.234, 0.512)	0.000
CONUT	D-dimer	.370 (0.205, 0.520)	0.000
	PCT	.174 (0.001, 0.340)6	0.048
	IL-6	.211 (0.040, 0.379)6	0.020
	NLR	.275 (0.122, 0.420)	0.001
	Neutrophil count	.031 (−0.121, 0.175)	0.706
	PLT	−.286 (−0.426, −0.129)	0.000
	Lymphocyte count	−.405 (−0.526, −0.226)	0.000

PT – prothrombin time; INR – international normalized ratio; PTA – prothrombin activity; APTT – activated partial thromboplastin time; PCT – procalcitonin; IL-6 – interleukin-6; NLR – neutrophil-to-lymphocyte ratio; PLT – platelet.

Figure 1.

Heatmap of correlations between Controlling Nutritional Status (CONUT) score and laboratory indices.

The heatmap displays Spearman’s correlation coefficients (rs) between the CONUT score and various coagulation and inflammatory parameters. A deeper red color indicates a stronger positive correlation, while a deeper blue indicates a stronger negative correlation. PT – prothrombin time; INR – international normalized ratio; PTA – prothrombin activity; APTT – activated partial thromboplastin time; PCT – procalcitonin; IL-6 – interleukin-6; NLR – neutrophil-to-lymphocyte ratio; neutrophil – neutrophil count; PLT – platelet; lymphocyte – lymphocyte count.

Influence of Nutritional Status on Coagulation and Inflammation

Logistic multivariate regression analysis was performed to further clarify how the severity of malnutrition affected coagulation and inflammatory outcomes, refining the correlation results observed in the previous section. All analyses were adjusted for potential confounding factors, including age, sex, body mass index, history of diabetes, and history of hypertension (Table 5).

Table 5.

Comparison of coagulation and inflammatory markers by nutritional status severity. The levels of coagulation indicators (INR and D-dimer) and inflammatory indicators (IL-6 and NLR) in the severely malnutrition group were significantly elevated with statistical differences. Data in the table are estimated means (95% CI), adjusted for age, gender, BMI, diabetes history, and hypertension history using a general linear model.

Index	Normal or mild malnutrition (n=28)	Moderate malnutrition (n=72)	Severe malnutrition (n=51)	P
PT	1.109 (1.062, 1.155)	1.168 (1.139, 1.196)	1.214 (1.180, 1.248)	0.002
INR	0.064 (0.029, 0.098)	0.136 (0.115, 0.156)	0.172 (0.147, 0.197)	<0.001
PTA	1.787 (1.680, 1.894)	1.710 (1.645, 1.775)	1.697 (1.619, 1.776)	0.384
APTT	1.520 (1.445, 1.596)	1.574 (1.529, 1.620)	1.597 (1.541, 1.653)	0.278
D-dimer	3.058 (2.852, 3.264)	3.207 (3.082, 3.331)	3.460 (3.306, 3.613)	0.005
PCT	−0.839 (−1.104, −0.574)	−0.651 (−0.791, −0.512)	−0.631 (−0.795, −0.466)	0.395
IL-6	1.045 (0.774, 1.316)	1.364 (1.208, 1.520)	1.439 (1.267, 1.612)	0.045
NLR	0.503 (0.373, 0.633)	0.669 (0.591, 0.793)	0.697 (0.602, 0.793)	0.045
Neutrophil count	0.524 (0.397, 0.650)	0.509 (0.433, 0.586)	0.493 (0.400, 0.586)	0.927
PLT	2.061 (1.951, 2.170)	1.965 (1.889, 2.031)	1.827 (1.746, 1.907)	0.002
Lymphocyte count	0.020 (−0.070, 0.111)	−0.160 (−0.215, −0.105)	−0.204 (−0.271, −0.137)	<0.001

PT – prothrombin time; INR – international normalized ratio; PTA – prothrombin activity; APTT – activated partial thromboplastin time; PCT – procalcitonin; IL-6 – interleukin-6; NLR – neutrophil-to-lymphocyte ratio; PLT – platelet.

Influence of Nutritional Status on Coagulation

The severe malnutrition group had significantly higher INR (0.172 vs 0.064), higher D-dimer levels (3.460 vs 3.058; P=0.005), and lower platelet counts (1.827 vs 2.061) (Table 5).

Influence of Nutritional Status on Inflammation

The severe malnutrition group exhibited significantly higher IL-6 (1.439 vs 1.045) and NLR (0.697 vs 0.503; P=0.045) (Table 5). No significant differences were observed for PTA (P=0.384), APTT (P=0.278), PCT (P=0.395), or neutrophil count (P=0.927). These findings directly link the severity of malnutrition, as assessed by the CONUT score, to progressively worse coagulation dysfunction and inflammatory activation.

Discussion

Our study demonstrates a high prevalence of malnutrition, as assessed by the CONUT score, among patients with cirrhosis and acute EVB. We further identified significant correlations between the severity of malnutrition and a spectrum of abnormalities in coagulation parameters (prolonged PT, INR, APTT; elevated D-dimer; decreased PTA and platelet counts) and inflammatory markers (elevated IL-6, NLR, PCT, SII). Multivariate analysis confirmed several of these parameters as independent risk factors for malnutrition, underscoring a tight interplay between nutritional status, coagulation competence, and systemic inflammation in this patient population.

These findings are consistent with and expand upon the existing literature. Malnutrition plays a critical role in cirrhosis, acting as both a complication and as a contributor to poor outcomes [6,21]. Our observed malnutrition prevalence of 97.4%, with 81.5% being moderate-to-severe, is notably higher than that of general reports in cirrhosis [22], highlighting the acute metabolic catabolism and stress superimposed by variceal hemorrhage. The coherence of our coagulation and inflammatory profiles with those reported by Li Qian et al [23] in a broader cirrhotic cohort validates our methodology. However, by focusing specifically on the acute EVB setting, our data suggest that the bleeding event serves as a potent amplifier of these perturbations, likely through stress-induced inflammation and a further acute compromise of the liver’s synthetic function.

The robust associations revealed by the Spearman analysis provide a mechanistic link between the CONUT score and hemostatic dysfunction. The positive correlations with PT, INR, APTT, and D-dimer, alongside negative correlations with PTA and platelets, are consistent with known pathophysiology. This includes impaired synthesis of vitamin K–dependent clotting factors [24,25], hepatic dysfunction exacerbated by hypoalbuminemia, hyperfibrinolysis due to reduced antifibrinolysin [26,27], and thrombocytopenia from portal hypertension–induced hypersplenism [28,29]. Our study strengthens this evidence by demonstrating that these coagulation defects are quantitatively correlated with the severity of a comprehensive nutritional deficit.

Similarly, the correlations between the CONUT score and inflammatory markers (IL-6, NLR) and lymphocytes reinforce the concept of malnutrition as a pro-inflammatory state. This likely stems from the synergy of intestinal barrier disruption, endotoxin translocation [11,12], and a diminished anti-inflammatory capacity associated with hypoalbuminemia [30,31]. The CONUT score, by integrating albumin levels and lymphocyte counts, inherently captures key facets of this inflammatory milieu, offering a more holistic clinical tool than isolated markers.

The clinical utility of our findings is underscored by the multivariate analysis, which identified PT, INR, D-dimer, IL-6, NLR, platelets, and lymphocytes as independent risk factors for malnutrition. In practice, these readily available laboratory parameters could serve as valuable proxies for nutritional status, alerting clinicians to the need for intensified nutritional support and anti-inflammatory strategies even when a formal CONUT score is not calculated. This is particularly relevant for mitigating coagulation risks and interrupting the vicious cycle linking malnutrition, inflammation, and bleeding.

However, the interpretation of our findings must be considered in the context of several limitations. First, the single-center, cross-sectional design of our study inherently limits the ability to establish causality and may introduce selection bias. While we identified significant associations, the temporal relationships between nutritional status, coagulation, and inflammation remain to be elucidated through longitudinal studies. Second, although we controlled for major confounding variables, residual confounding from unmeasured variables, such as subclinical infection or acute kidney injury, cannot be entirely ruled out. Third, regarding methodological limitations, the use of a single nutritional assessment tool (CONUT score), despite its validated objectivity, might not capture all dimensions of malnutrition. Future studies incorporating a combination of tools (eg, anthropometrics, functional assessments) would provide a more comprehensive picture. Finally, the specific assays and analyzer platforms used, while standardized and controlled, could yield slightly different absolute values, compared with other laboratory systems, which is a general consideration for the reproducibility of the exact numerical correlations we report. Future multi-center studies with larger, prospective cohorts and integrated mechanistic analyses (eg, of gut microbiome or cytokine dynamics) are needed to validate and extend our conclusions.

Conclusions

In conclusion, this study provides compelling evidence that malnutrition, as quantified by the CONUT score, is highly prevalent and is intricately linked with coagulopathy and systemic inflammation in patients with cirrhosis and acute EVB. The CONUT score emerges as a practical and integrative clinical tool that can effectively stratify patient risk and guide timely, multifaceted interventions aimed at improving nutritional status, correcting coagulation defects, and modulating inflammation. Addressing the limitations of this research in future work will be crucial to further refine these clinical strategies and validate their effect on long-term outcomes.