Lobar pneumonia (LP) is a common and severe respiratory disease in children, characterized by acute inflammatory consolidation of one or more lobes of the lung. Owing to the widespread inoculation of antibiotics and the advancement of medical technology, the mortality rate attributable to LP has significantly reduced, but severe cases can still lead to serious complications such as empyema, lung abscess, and acute respiratory distress syndrome, posing a major threat to children’s health [1, 2, 3]. Therefore, early and accurate assessment of the severity of the disease and dynamic monitoring of its progress are of great clinical significance for formulating individualized treatment plans and improving the prognosis of affected children. Globally, LP remains a leading infectious cause of pediatric hospitalization, representing a primary contributor to morbidity and causing increased utilization of healthcare resources [4]. The cornerstone of diagnosis for the disease combines clinical assessment with radiological confirmation, typically by chest X-ray, while its management hinges on the timely administration of appropriate antibiotics. The clinical challenge, however, often lies not in diagnosis but in accurately stratifying disease severity at presentation. This stratification is critical, as it dictates management decisions—from outpatient care to hospital admission, and even to the intensive care unit—directly impacting patient outcomes and antibiotic stewardship.

At present, the clinical evaluation of the severity of LP in children depends on imaging examinations (such as chest X-ray, computed tomography [CT]), inflammatory markers (such as C-reactive protein, procalcitonin, white blood cell count) and clinical manifestations [5, 6, 7]. However, these indicators present some limitations in clinical evaluation of the disease. For example, patients are at risk of radiation exposure during imaging examination and achieving high-frequency dynamic monitoring is challenging [8]. Despite the high sensitivity of the conventional set of inflammatory markers, their inadequate specificity is a major concern, as both infectious and non-infectious factors unrelated to LP could contribute to their altered levels [9]. Therefore, current clinical research on pediatric LP has focused on identifying biochemical indicators that are straightforward to measure and provide specific insight into tissue damage and disease progression.

In this context, the ratio of aspartate aminotransferase (AST) to alanine aminotransferase (ALT)—known as the De Ritis ratio—has received extensive attention in recent years. The ratio was first proposed by Fernando De Ritis in 1957 and was used to assist in the identification of viral hepatitis and alcoholic liver injury. It reflects the pattern of liver cell injury and the degree of mitochondrial dysfunction [10, 11, 12]. Accumulating evidence has proved its broad clinical value beyond the field of liver disease. Multiple studies have shown that AST is widely present in a variety of extrahepatic tissues such as myocardium, skeletal muscle, kidney, red blood cells and lung tissues, and its elevated activity is often closely related to necrosis or severe cell injury [13, 14, 15, 16]. In severe infections such as pneumonia, the release of pathogen toxins, systemic inflammatory response storms, and possible hypoxemia can lead to multiple organ cell damage, especially destruction of cells in the lung tissue and red blood cells, causing significant changes in serum AST levels, thereby affecting the De Ritis ratio [17, 18]. In light of this, we speculate that the De Ritis ratio, a traditional liver function index, has the potential for dynamic monitoring of LP severity in children. The changes may not only indirectly reflect the intensity of systemic inflammatory response caused by pneumonia but may also be related to the degree of damage to the lung tissue itself and other important organs, such as the heart. By dynamically monitoring the trajectory of the ratio, it may be possible to reveal its internal relationship with the changes in pneumonia severity, complications and prognosis, and to provide clinicians with an additional and easy-to-implement evaluation window.

The purpose of this study was to systematically observe the dynamic changes of De Ritis ratio in the course of LP through retrospective analysis, and to explore its correlation with the severity of the disease. It is expected that this study can provide a new laboratory basis for the evaluation system of LP in children and provide a reference for clinical prognosis and intervention timing.

In this study, the clinical data of 120 children with LP were systematically analyzed, and the dynamic changes of De Ritis ratio in the course of the disease and their correlation with the disease severity were discussed. The results showed that the De Ritis ratio of children with severe LP was significantly higher than that of children with ordinary-severity LP at all time points—both before and after treatment, accompanied by a greater alteration ($\mathrm{\Delta }$De Ritis ratio). The ratio also demonstrated excellent diagnostic efficacy for predicting LP severity in pediatric patients (AUC = 0.908), displayed a nonlinear dose–response relationship with the severity of the disease, and yielded a high net benefit during clinical decision-making. These findings strongly suggest that the De Ritis ratio can be used as a new, reliable and clinically valuable biomarker for assessing the severity of LP in children.

Traditionally, the De Ritis ratio is mainly used for the differential diagnosis of hepatobiliary diseases. However, building on existing literature, the present study provides additional evidence that the clinical significance of the De Ritis ratio extends beyond liver disease. AST is not only present in the cytoplasm of liver cells, but also in mitochondria [21]. It is also widely distributed in various extrahepatic tissues such as myocardium, skeletal muscle, kidney, red blood cells, and lung tissues. ALT is mainly localized in the cytoplasm of hepatocytes [22]. In the pathological process of LP, especially severe LP, the massive release of bacterial toxins, strong systemic inflammatory response, and possible hypoxemia can lead to direct cell damage and necrosis in multiple organs [23, 24]. We speculate that extensive inflammation and destruction of parenchymal cells in lung tissue, hemolysis (leading to destruction of red blood cells and release of AST), and potential myocardial injury (such as toxic myocarditis) may be accountable for the significant elevation of serum AST. In contrast, ALT levels did not exhibit significant changes in the contexts of pulmonary inflammation. Therefore, the increase in the AST/ALT ratio may essentially reflect the extensive tissue and cell damage caused by LP, especially the severe damage or necrosis of mitochondria-rich cells. In this study, the higher levels of NEU, NLR, PCT, CRP and other inflammatory indicators in the severe pneumonia group indicate the presence of a more severe inflammatory response, which may underlie the observed increase in the De Ritis ratio. The process of its dynamic decline may be closely related to the control of inflammation and the gradual repair of tissue damage.

The core finding of this study is the strong correlation between the De Ritis ratio and LP severity. The severe pneumonia group not only had a significantly higher baseline De Ritis ratio before treatment, but its ratio also persistently exceeded that of the common pneumonia group at every evaluated time point during the treatment process. More importantly, the magnitude of its decline ($\mathrm{\Delta }$De Ritis ratio) was found to be more pronounced in the severe pneumonia group, indicating augmented tissue damage in critically ill children, and effective treatment promotes a significant decline in this indicator; these findings suggest that the De Ritis ratio serves a dual function in both LP evaluation and treatment efficacy monitoring. ROC curve analysis showed that the AUC of the De Ritis ratio in the diagnosis of severe LP before treatment was as high as 0.908, coupled with a high specificity (94.76%), indicating a high clinical utility value of the index marked by its high accuracy in severe disease identification and low false positive rate.

Hierarchical multiple linear regression further revealed that gender, age, BMI, and the range of lung lesions were independent influencing factors of $\mathrm{\Delta }$De Ritis ratio. Tissue damage is more significant in older children due to a more intense immune response; bilateral lobar lesions generally indicates a heavier infection load and a wider range of inflammation, which is consistent with a higher De Ritis ratio. These findings indicate that when using this ratio for evaluation, it is necessary to comprehensively interpret the specific demographic characteristics and disease features of the pediatric patients. Furthermore, the nonlinear dose–response relationship between the De Ritis ratio and the risk of severe LP detected in the restricted cubic spline analysis indicates that when the ratio exceeds a certain critical value or threshold, the risk of severe pneumonia will no longer increase at a constant rate; instead, it will rise sharply and disproportionately to the De Ritis ratio.

It is noteworthy that our analysis yielded two complementary De Ritis ratio values for clinical assessment, both derived from the ROC analysis framework. The primary analysis provided an optimal diagnostic cutoff of 1.52, which offers the best balance between sensitivity (83.42%) and specificity (94.76%) for discriminating between common and severe pneumonia at presentation. Furthermore, internal validation via bootstrapping refined our understanding of this threshold. The bootstrapping procedure, which assesses the stability of the ROC results, generated a 95% confidence interval for the optimal cutoff ranging from 1.28 to 1.41, with a point estimate of 1.34. The lower boundary and point estimate are clinically informative. A ratio above 1.52 offers strong diagnostic confidence for severe pneumonia, while values between 1.34 and 1.52 indicate a high-risk or early warning zone. Patients with De Ritis ratios within the 1.34–1.52 interval may not meet the definitive diagnostic threshold but are likely at significantly increased risk for severe disease, warranting closer clinical monitoring and a lower threshold for escalating care.

Application of the De Ritis ratio in LP severity stratification has significant clinical advantages: First, it is economical and efficient. As both the AST and ALT are conventional biochemical indicators, their detection is relatively inexpensive and rapid owing to the existing assays. Second, the De Ritis ratio is conducive to dynamic monitoring, without causing radiation exposure, and can be repeatedly detected to evaluate the disease progression and monitor treatment efficacy. Third, the De Ritis ratio has higher specificity than the conventional indicator. Compared with the traditional inflammatory markers such as CRP and PCT, it provides a direct assessment of the degree of tissue damage or the severity of pneumonia. Based on our clinical decision curve analysis, the De Ritis ratio can inform clinical decision-making, offering higher net benefit to patients within a wide range of threshold probabilities. This shows that incorporating this indicator into the clinical decision-making system (such as whether to upgrade treatment or whether to transfer patients to pediatric intensive care unit [PICU]) has practical clinical application value, which can facilitate early and accurate identification of high-risk children, as well as optimization of treatment strategies, thereby improving patient prognosis.

Several limitations of this study should be highlighted. Firstly, this is a single-center retrospective study, which may be subject to selection bias. Secondly, the sample size adopted was relatively limited, necessitating more multicenter studies with larger samples to validate our findings in the future. Thirdly, despite the implementation of a comprehensive set of microbiological tests to exclude viral pneumonia (using PCR for common respiratory viruses), the possibility of undetected or emerging viral pathogens cannot be entirely ruled out, which might represent a potential source of misclassification bias. Fourth, the lack of a prospective a priori sample size calculation weakens the rigor of the study design to some extent. Although this study demonstrates a correlation, the specific mechanisms underlying the increased De Ritis ratio—such as the relative contributions from lung tissue, myocardium, or red blood cells—require further in-depth investigation through basic research using animal models and targeted marker analyses. Finally, this study did not analyze pathogen-specific $\mathrm{\Delta }$De Ritis ratio caused by a range of pathogens, such as Mycoplasma pneumoniae and Streptococcus pneumoniae, which represents a potential direction for future research.

In summary, the De Ritis ratio was significantly increased in children with LP, especially in cases of severe pneumonia, and its dynamic changes were closely associated with the disease severity. The ratio is a readily obtainable, low-cost biomarker with high diagnostic efficiency and substantial clinical utility. When combined with existing clinical indicators, it has the potential to contribute to a comprehensive system for early assessment and dynamic monitoring of LP severity in children, provide robust support for accurate clinical decision-making, and ultimately improve patient outcomes.

• The De Ritis ratio is significantly elevated in children with severe lobar pneumonia and demonstrates distinct dynamic changes throughout the disease course, with a more pronounced decrease observed in severe cases.

• This ratio exhibits high diagnostic accuracy for severe lobar pneumonia (AUC = 0.908), with an optimal cutoff value of 1.52.

• A nonlinear dose–response relationship exists between the De Ritis ratio and severe lobar pneumonia.

• The $\mathrm{\Delta }$De Ritis ratio is influenced by patient factors such as baseline De Ritis ratio, age, body mass index, and the presence of bilateral lung lesions.

• As an economical and readily available biomarker, the De Ritis ratio shows substantial clinical utility for severity assessment, dynamic monitoring, and clinical decision-making in the contexts of pediatric lobar pneumonia.

All experimental data included in this study can be obtained by contacting the corresponding author if needed.

RL designed the research study, analyzed the data and drafted the manuscript. SSC performed the research. QQY collected the data. All authors contributed to the important editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was conducted following the principles of the Declaration of Helsinki and was approved by the Ethics Committee of the Affiliated Hospital of Xuzhou Medical University (Approval No.: XYFY2025-KL475-01). Written informed consent was obtained from the parents or legal guardians of all participating children.

Not applicable.

This research received no external funding.

The authors declare no conflict of interest.