Introduction

LOLA has been used to treat hyperammonemia caused by various acute and chronic liver diseases, such as cirrhosis, fatty liver, and hepatitis, and is particularly suitable for relieving central nervous system symptoms caused by liver diseases and for the rescue of hepatic coma. Additionally, LOLA has shown potential applications in protecting against the side effects of chemotherapy in cancer patients and in promoting liver function recovery after surgery1. It can directly participate in the metabolic processes of hepatocytes. In the body, LOLA breaks down into aspartate and ornithine, both of which play key roles in ammonia detoxification, energy metabolism, and the repair and regeneration of hepatocytes. Ornithine, as a key substance in the urea cycle, promotes the conversion of ammonia to urea, thereby reducing blood ammonia levels; aspartate, on the other hand, participates in the tricarboxylic acid cycle, providing energy for liver cells2. Additionally, aspartate helps maintain metabolic balance within liver cells, promoting their normal physiological functions and positively impacting liver function improvement. LOLA can also enhance the activity of glutathione-S-transferase and other detoxifying enzymes in the liver. These enzymes can transform harmful substances in the body, making them easier to excrete and promoting the metabolism and excretion of bilirubin, thereby improving the liver’s detoxification function and reducing its detoxification burden3. LOLA has antioxidant properties, which can reduce the production of free radicals and protect the integrity of hepatocytes. In summary, LOLA helps the liver better process metabolic waste, promotes the repair and regeneration of damaged hepatocytes, thereby alleviating the symptoms of hepatobiliary diseases and improving the condition4,5. Because clinical trials have relatively limited sample sizes and shorter trial periods, it is difficult to detect rare or adverse events that only occur after long-term use of the drug.

The safety of a drug is a dynamic issue that requires continuous attention. Traditional clinical trials, although they have strictly assessed the safety and efficacy of drugs before they are marketed, often cannot cover all possible adverse reactions due to the limitations of trial design. The clinical application of LOLA is primarily focused on treating hyperammonemia induced by acute and chronic liver diseases. It can effectively reduce blood ammonia levels, thereby alleviating the liver burden and promoting the recovery of patients’ neurological functions. The adverse events (AEs) of LOLA include central nervous system injury, hypochromasia, splenic injury, hyperchloraemia, anisocytosis, etc. Among these, hepatitis fulminant and hepatic failure are fatal complications.

This study aims to systematically and comprehensively explore the toxicity and basic characteristics of LOLA in central nervous system injury diseases. Given the widespread use of LOLA in the treatment of hepatobiliary disorders system diseases and its potential toxicity risks, this study hopes to provide a scientific basis for clinical decision-making, optimize treatment plans, reduce the occurrence of adverse events, and ultimately improve patients’ therapeutic outcomes and quality of life.

Methods

Study design and data sources

All our data comes from the FAERS Database (https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html). This system consists of adverse event and medication error reports voluntarily submitted to the FDA by healthcare professionals and consumers. The database comprises seven subsets: patient information, drugs, indications, outcomes, adverse reactions, medication timing, and reporting country. These subsets are linked and analyzed through the personal safety report code field. The data processing and analysis software used in this study is R software (version 4.4.0)6. Data preparation: After extracting the core variables (including patient ID, case ID, drug name, reaction term, report date, and age) from the FAERS dataset, we first converted all character variables to a uniform format (e.g., converting drug names and reaction terms to lowercase) to avoid false duplicates caused by case sensitivity. Identification of duplicates: We defined duplicates as records that were identical in all key variables mentioned above. Validation: After removing duplicates, we verified the effectiveness of the process by checking the total number of records before and after cleaning, and randomly sampling 100 records to confirm that no identical entries remained in the cleaned dataset.

Data cleaning

Since the FAERS database can be reported by different institutions and the public, there are multiple versions of duplicate cases and non-standardized data. Therefore, before analysis, it is necessary to remove duplicates using the primaryid and caseid in the seven files. The primaryid is the unique number for identifying FAERS reports and the main linking field in the data files, while the caseid is the number for identifying cases. Each caseid corresponds to one Individual Case Safety Report (ICSR)7. When an ICSR is submitted multiple times subsequently, its caseid remains unchanged, but the primaryid will generate a new, larger number. Generally, the more updated report is retained, that is, for data with the same caseid in the DEMO table, the report with the largest primaryid is kept. The deduplication method applies only to follow-up reports from the same source (e.g., updated ICSRs from the same pharmaceutical company or healthcare provider). This deduplication is only applicable to follow-up reports and does not account for multiple reports submitted by different individuals.

Data extraction and standardization

Data from 2004Q1 to 2024Q3 with ‘LOLA’ as the search term were extracted from the FAERS Database. Duplicate reports from the same patient were removed. The collected data mainly included the following characteristics of the drug-related reports: patient age, patient gender, reporter type, adverse event outcome (death, life-threatening, hospitalization, disability, and/or others), drug and its indication. The data were standardized using the Preferred Terms (PT) in version 26.1 of the Medical Dictionary for Regulatory Activities (MedDRA), and adverse events were classified into different systems using the System Organ Class (SOC).

Data statistical analysis

To reduce the probability of false-positive signals, this study employed four methods, namely the Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), and Empirical Bayes Geometric Mean (EBGM), to evaluate the correlation between drugs and adverse reaction events. The larger the signal value of an adverse event, the stronger the adverse event signal, indicating a stronger correlation between LOLA and the target adverse event. The calculation formulas and evaluation criteria of the four methods are listed in Table 1. All data cleaning, statistical analysis, and data visualization were performed using R software (version 4.4.0).

Table 1 Four major methods used for signal detection.
Full size table

Results

Descriptive analysis

A total of 413 reported cases of LOLA were obtained. Among them, 200 cases were male, 196 were female, and 17 cases were of unknown gender, as shown in Fig. 1. The majority of patients were aged 41–65 years old and ≥ 65, with 199 cases. The main reporting countries were Poland, Germany, and Korea, as shown in Fig. 2. The majority of reporters were Physicians, accounting for 60.77%. The most common outcome was hospitalization (44.18%), followed by other serious (31.92%). The year with the largest number of reported cases was 2023, as shown in Fig. 3. Adverse reactions mainly occurred within less than 7 days and 7–28 days, as shown in Fig. 4.

Fig. 1
figure 1

Proportion of adverse drug outcomes.

Full size image
Fig. 2
figure 2

Top 3 country of adverse events.

Full size image
Fig. 3
figure 3

Distribution of adverse events due to LOLA reported annually.

Full size image
Fig. 4
figure 4

Time of adverse events due to LOLA reported annually.

Full size image

Disproportionality analysis

Through the screening of LOLA, a total of 323 positive risk signals for adverse events (AEs) were identified, covering 25 system organ classes (SOCs). When compared with the current LOLA product label, it was found that, in terms of the quantity of adverse event (AE) reports, 4 AEs were not listed. And considering the strength of the risk signals, 17 AEs were absent from the label, as shown in Supplementary Material 1. LOLA exhibited positive risk signals for AEs. According to the System Organ Class (SOC) classification, these AEs mainly impacted 4 organ systems. Among them, hepatobiliary disorders were at the top of the list, with a total of 84 positive risk signals detected for liver-related AEs. The positive signal for hepatitis fulminant was the strongest, with RR [82.13, 95% CI (40.97, 164.65)]. According to the ROR ranking, LOLA mainly accumulates in systems mainly involving hepatobiliary disorders, blood and lymphatic system disorders, metabolism and nutrition disorders, and investigations(abnormal lab findings such as elevated liver enzymes). The preferred terms (PTs) with higher intensities were central nervous system injury RR [4961.1, 95% CI (2411.89, 10,204.64)], hypochromasia RR [1142.75, 95CI% (561.17, 2327.04)], splenic injury RR [969.24, 95% CI (477.18, 1968.72)], hyperchloraemia RR [380.92, 95% CI (196.86, 737.06)], and anisocytosis RR [359.46, 95% CI (178.57, 723.61)], as shown in Table 2.

Table 2 Disproportionate analysis in top four soc.
Full size table

Discussion

Selection and validation of statistical methods

These four methods are commonly used statistical methods in adverse drug reaction (ADR) signal detection, and each method has its unique advantages and applicable scope: The ROR is based on the concept of the odds ratio and calculates the association strength between the target drug and the target adverse reaction. It is simple and easy to implement, making it suitable for preliminary signal detection. The PRR compares the reporting proportions of the target adverse reaction between the target drug and all other drugs. It is intuitive and easy to interpret, and is widely applied in pharmacovigilance systems. The BCPNN is based on Bayesian statistical methods and is capable of handling the problem of data sparsity. It is suitable for detecting signals of rare adverse reactions. The EBGM method can adjust the reporting frequencies of different combinations of drugs and adverse events to correct these reporting biases, so that the analysis results can better reflect the actual occurrence of adverse drug reactions. Reasons for not choosing other methods (such as FDR correction): The FDR (False Discovery Rate) correction is usually used for multiple testing correction to reduce false positive signals. However, the FDR correction may be too strict, resulting in the omission of some true signals. In preliminary signal detection, it may be more preferable to use methods with higher sensitivity (such as ROR and PRR), and then verify the results through other methods (such as BCPNN and EBGM).

Basic information on LOLA-related adverse event reports

The research results show that the number of adverse event reports of LOLA is related to gender. Among the LOLA reports with known genders, there was no significant difference in the number of patient reports between males and females. In terms of the reported age, the age of the population experiencing adverse events is mainly concentrated in the range of 41–65 years old. However, there are still 40 reports with unknown ages in this study, which may have a certain impact on the results. This database lists only the three countries with a clear number of reported cases, and the top three countries in terms of the number of reports are Poland, Germany, Korea. A significant proportion of reports originate from Poland, Germany, and Korea, which may reflect disparities in drug market distribution or variations in reporting culture. Poland may have a high prescription rate for specific drugs (e.g., antibiotics, psychotropic drugs) that are more likely to cause adverse reactions or are monitored more closely. For example, the problem of antibiotic abuse is more prominent in eastern European countries, which may lead to more reports of drug-related adverse events. Poland, Germany, and Korea may have established an efficient pharmacovigilance database or electronic health record system to facilitate the collection and aggregation of adverse reaction data. For example, the National Health Fund (NFZ) information system may be linked to the pharmacovigilance system to improve reporting efficiency. Check whether such geographic biases could amplify the observed signals (e.g., stricter monitoring in these countries for specific adverse reactions).

Involvement of SOC by LOLA-related AEs

LOLA is an essential substrate for the synthesis of urea and glutamine8. Ornithine can activate specific enzymes in urea synthesis, thereby promoting ammonia metabolism, reducing blood ammonia levels, and achieving the conversion and detoxification of blood ammonia9. The results show that the adverse events reported for LOLA involve 25 organ systems in the Med classification, which may be related to its wide-ranging mechanism of action. Among them, hepatobiliary disorders, blood and lymphatic system disorders, metabolism and nutrition disorders, and investigations have a relatively large number of reports and strong signals, which is generally consistent with the description in the instruction manual. The risk of hepatobiliary disorders was not described in the manual, but this study found that the positive signal risk for the hepatobiliary system is the highest. The hepatobiliary system exhibits the strongest signal (ROR = 6.68). LOLA is metabolized in the body into ornithine and aspartate, both of which participate in the urea cycle and glutamine synthesis, aiding in the clearance of ammonia. However, in patients with severely impaired liver function, the liver’s metabolic capacity may already be at its limit, and additional metabolic burdens could lead to hepatocyte injury. High doses of LOLA may result in elevated amino acid levels, disrupting the normal metabolic functions of the liver. LOLA is primarily used to treat patients with liver diseases, who may already be at risk for hepatobiliary injury. Therefore, the observed hepatobiliary damage may not be entirely caused by LOLA but could be related to the underlying liver disease state of the patients. LOLA may interact with other drugs, affecting their metabolism or toxicity. For example, when used in combination with other hepatotoxic drugs, the risk of hepatobiliary injury may be increased.

LOLA and other suspected PT signals

The research results show that the PTs indicating that LOLA may cause hepatobiliary disorders include hepatitis fulminant, hepatic failure, liver injury, hepatic cirrhosis, cholangitis, acute hepatic failure, hyperbilirubinaemia, autoimmune hepatitis, drug-induced liver injury, hepatotoxicity, and hepatic steatosis. This suggests that clinicians should be vigilant against adverse reactions in the hepatobiliary system when applying LOLA10. Considering that the indications of LOLA are somewhat similar to the symptoms of some PTs in hepatobiliary disorders, central nervous system diseases, and infectious and infestational diseases, and the changes in relevant examination indicators may be caused by the diseases treated by LOLA itself, or by the action of LOLA or subsequent infections, clinical judgment is required during application11. Therefore, during the treatment of hepatobiliary disorders and central nervous system diseases, clinicians should carefully identify whether there are adverse events12.

Mechanism of LOLA

LOLA can provide substrates for the synthesis of urea and glutamine13. Glutamine serves as both the detoxification outcome of ammonia and the storage and transport form of it. In physiological and pathological states, the synthesis of urea and glutamine can be influenced by ornithine, aspartate, and other dicarboxylic compounds14. LOLA participates in nearly the whole process of urea-cycle activation and ammonia detoxification15. In this process, arginine is formed, and then urea is split off to form ornithine. Aspartate participates in the synthesis of nucleic acids in liver cells, which is beneficial for the repair of damaged liver cells16. In addition, due to the indirect promoting effect of aspartate on the tricarboxylic acid cycle metabolism process in liver cells, the energy production in liver cells is promoted, enabling the restoration of various functions of the damaged liver cells17. LOLA plays a significant role in the medical field, particularly in the treatment of liver diseases. Below is an analysis of its benefits and risks: LOLA can directly participate in hepatocyte metabolism, activating key enzymes in liver detoxification, thereby assisting in the clearance of harmful free radicals from the body and enhancing liver detoxification function. LOLA can increase the activity of carbamoyl phosphate synthetase and ornithine carbamyltransferase, promoting the utilization of ammonia by the brain, liver, and kidneys to synthesize urea and glutamine, thereby lowering blood ammonia levels. LOLA may also have potential benefits such as nourishing brain cells, promoting growth and development, enhancing immunity, and improving memory. However, these benefits require further research for confirmation18. Although LOLA has remarkable effects in the treatment of liver diseases, it is not suitable for all patients with liver disease19. Furthermore, LOLA may also induce a range of other adverse reactions, including but not limited to central nervous system injury, hypochromasia, splenic injury, hyperchloraemia, and anisocytosis. Therefore, in clinical practice, it is necessary to strengthen the monitoring of patients’ blood routine, liver and kidney functions, and other relevant biochemical indicators to promptly detect and address any potential abnormalities, ensuring the safety and effectiveness of the treatment20,21. Before use, doctors should comprehensively assess the patient’s condition and physical status to determine suitability for LOLA22.

Research limitations

Most of the reporting countries are in Europe and America, with relatively few in Asia. Since the FAERS database is a voluntary reporting system and most of the reporters in this study are consumers, the quality of the reports varies. There are often deficiencies such as missing reported data, under-reporting, lack of actual clinical information of patients, and difficulty in excluding other risk factors. Therefore, the data may not be accurate. Whether it has clinical significance requires further clinical trials for exploration. Competing Risks: Patients may concurrently use other hepatotoxic drugs (e.g., acetaminophen), and the impact of co-medications has not been addressed. Despite potential underreporting, FAERS remains a valuable signal detection tool; the use of multiple disproportionality analyses strengthens reliability.

Conclusion

Based on data from the FAERS database, this study employed the ROR, PRR, EBGM, and BCPNN methods to detect adverse event signals associated with LOLA, package insert must be based on comprehensive information—including robust epidemiological studies, clinical trials, and thorough risk–benefit evaluations—and must strictly adhere to pharmacovigilance guidelines and risk-minimization strategies. Our study’s signals, derived from the FAERS database using ROR, PRR, EBGM, and BCPNN, should be regarded as preliminary hypotheses that can inform. Medical staff should work together to judge the changes in relevant laboratory indicators during clinical application. When treating patients with hepatobiliary diseases, blood diseases, and central nervous system diseases, LOLA should be used with caution. At the same time, vigilance should be maintained against adverse reactions related to the hepatobiliary system.