Effect of Ceftriaxone Dosage and Albumin–Bilirubin Score on the Risk of Ceftriaxone-Induced Liver Injury

Abstract

The albumin–bilirubin (ALBI) score is an index of hepatic functional reserve and is calculated from serum albumin and total bilirubin levels. However, the relationship between ceftriaxone (CTRX)-induced liver injury and ALBI score remains unknown. Therefore, we aimed to elucidate the risk of CTRX-induced liver injury based on the ALBI scores and CTRX dosage. This was a single-center, retrospective, case-control study of 490 patients and the primary outcome was CTRX-induced liver injury. We performed a COX regression analysis using age ≥75 years, male sex, alanine aminotransferase levels, ALBI score, and CTRX dosage regimen (4 ≥2 or 1 g/d) as explanatory factors. We also performed 1 : 1 propensity score matching between non-liver injury and liver injury groups. The incidence of liver injury was 10.0% (49/490). In COX regression analysis, CTRX 4 g/d was an independent risk factor for liver injury (95% coefficient interval: 1.05–6.96, p = 0.04). Meanwhile, ALBI score ≥−1.61 was an independent factor for liver injury (95% coefficient interval: 1.03–3.22, p = 0.04) with the explanatory factor of ≥2 and 1 g/d. The Kaplan–Meier curve indicated that the cumulative risk for CTRX-induced liver injury was significantly higher in the ALBI score ≥−1.61 group than in the ALBI score <−1.61 group before propensity score matching (p = 0.032); however, no significant differences were observed after propensity score matching (p = 0.791). These findings suggest that in patients treated with CTRX with ALBI score ≥−1.61, frequent liver function monitoring should be considered.

INTRODUCTION

Drug-induced liver injury is one of the most significant adverse reactions associated with antibiotic treatment. As the mortality rate of antibiotic-induced liver injury has been reported to be 26.3% in severe cases,¹⁾ risk management by healthcare workers is essential in clinical settings.

Ceftriaxone (CTRX) has activity against a wide range of Gram-positive and Gram-negative bacteria,²⁾ and third-generation cephalosporin antibiotics have been used for various infections, such as lower respiratory tract infection.³⁾ Data mining analysis suggested that CTRX administration may be involved in liver injury.⁴⁾ In fact, Nakaharai et al. reported that high-dosage CTRX (4 g/d) significantly increased the risk of liver injury compared with normal-dosage (2 g/d).⁵⁾ Meanwhile 4 g/d is used for meningitis and other severe infections⁶⁾ and patients with general infections are administered mainly 2 g/d,⁷⁾ indicating that a detailed investigation of risk factors for liver injury in normal-dosage of CTRX should be conducted.

The albumin–bilirubin (ALBI) score, calculated from serum albumin and total bilirubin levels,⁸⁾ is an objective index of hepatic functional reserve used for predicting the prognosis of hepatocellular carcinoma (HCC).⁹⁾ An increase in ALBI score is correlated with a decrease in hepatic functional reserve,¹⁰⁾ and may be associated with micafungin-,¹¹⁾ doripenem-,¹²⁾ and ampicillin/sulbactam-¹³⁾induced liver injury in non-HCC patients, suggesting that the ALBI score may be a predictive index for CTRX-induced liver injury. However, the relationship between the ALBI score and CTRX-induced liver injury remains unknown.

This study aimed to investigate the effects of CTRX dosage and ALBI score on the risk of CTRX-induced liver injury.

MATERIALS AND METHODS

Study Participants and Exclusion Criteria

This single-center, retrospective, case-control study was performed at the National Hospital Organization Mie Chuo Medical Center (Mie, Japan), using electronic medical records. Patients who were administered CTRX between July 1, 2021 and April 30, 2023, were enrolled, and baseline data were collected. The exclusion criteria of the present study were as follows: (I) multiple administrations of CTRX treatment in the same patient, (II) patients <18 years old, (III) duration of treatment <5 d, (IV) missing baseline data, and (V) baseline alanine aminotransferase (ALT) and/or alkaline phosphatase (ALP) levels more than twice the upper limit of normal (ULN) as in previous studies.^11–13) The DDW-J scale has been frequently used to diagnose drug-induced liver injury.¹⁴⁾ On this scale, drug-induced liver injury during treatment is defined as: 5–90 d is 2 points, and <5 and >90 d is 1 point.¹⁴⁾ Additionally, in a previous study on CTRX-induced liver injury, patients who received CTRX for less than 5 d were excluded.⁵⁾ Therefore, 5 d was used as the cut-off for effectively detecting liver injury in the present study.

Participants Data Collection

Baseline clinical laboratory data and medical histories of the participants were collected by referring to their medical records at the start of CTRX administration. Concomitant antibiotics were defined as those administered during the same period as CTRX. Concomitant total parenteral nutrition and fat emulsion were defined as at least one day of CTRX administration. Concomitant acetaminophen was defined as 1500 mg/d administered for three consecutive days, according to a previous study.⁵⁾ ALBI score was calculated using the following formula: [log10 total bilirubin (µmol/L) × 0.66] + [albumin (g/L) × −0.085].¹⁵⁾

Definition of CTRX-Induced Liver Injury

The primary outcome of this study was CTRX-induced liver injury. The ULN of ALT and ALP levels was determined based on the following facility criteria: male (ALT, 42 IU/L and ALP, 113 IU/L) and female (ALT, 23 IU/L and ALP, 113 IU/L). In the present study, “baseline” refers to the ALT and ALP values immediately before CTRX administration. If the patients had baseline ALT and/or ALP levels below the ULN (ALT/ALP < ULN), CTRX-induced liver injury was defined as a 2-fold increase from the ULN for ALT and/or ALP following CTRX administration.¹¹⁾ When the baseline ALT and/or ALP levels were below 2 × ULN (ULN ≤ ALT/ALP < 2 × ULN), liver injury was identified when these values increased 2-fold from the baseline following CTRX administration.¹¹⁾

According to a previous study,⁵⁾ we monitored ALT and ALP levels during CTRX administration, and until 14 d after the end of administration. The pattern of liver injury was assessed by the ratio of ALT/ULN and ALP/ULN as follows: cholestatic-type injury is ALT/ALP <2, mixed cholestatic-hepatocellular injury is 2≤ ALT/ALP ≤5, and hepatocellular injury is ALT/ALP >5.¹⁶⁾

Statistical Analysis

The Mann–Whitney U test was used to compare continuous variables between the non-liver injury and liver injury groups after confirming a non-normal distribution. Fisher’s exact test was used to compare categorical variables. Cutoff value was evaluated based on the sensitivity, specificity, and area under the curve (AUC) using receiver operating characteristic analysis. In COX regression analysis, CTRX-induced liver injury was used as an independent variable. Because ALT,¹²⁾ ALBI score,^11–13) and CTRX dosage⁵⁾ may be involved in liver injury, we selected these as explanatory factors. The association between ALBI score and liver injury was examined with adjustment for (model 1) except the dosage of CTRX, (model 2) ALBI score and 4 g/d of CTRX, (model 3) ALBI score and ≥2 g/d of CTRX, and (model 4) ALBI score and 1 g/d of CTRX. The cumulative risk of CTRX-induced liver injury was calculated using the Kaplan–Meier curve and compared using the log-rank test. We also performed 1 : 1 propensity score matching between the non-liver injury and liver injury groups. The propensity score was calculated from the liver injury-related factors, such as sex, age, serum creatinine, ALT, AST, ALP, C-reactive protein, white blood cell, platelet, CTRX dosage, concomitant acetaminophen, clindamycin, ampicillin, total parenteral nutrition, fat emulsion, history of hepatitis B, hepatitis C, cholecystitis, cholangitis, cholangiocarcinoma, liver cancer, liver cirrhosis, alcoholic liver disease, and antibiotic allergy. The non-liver injury and liver injury groups were caliper-matched 1 : 1 (0.2 of the standard deviation of the logit of the propensity score). After propensity score matching, the respective baselines of patient characteristics were evaluated, and the standardized difference was calculated. All statistical analyses were performed using SPSS Statistics version 28 (IBM Japan, Tokyo, Japan), and statistical significance was set at p < 0.05.

Ethics Approval Statement

This study was conducted in accordance with the Ethical Guidelines for Medical and Health Research Involving Human Subject. This study design was approved by the National Hospital Organization Mie Chuo Medical Center (Approval Ref. MCERB-202261). Due to the retrospective study design, oral agreement was not required. Therefore, informed consent was obtained using an opt-out document posted on the website of the National Hospital Organization Mie Chuo Medical Center.

RESULTS

A total of 1657 patients were administrated CTRX during the study period (Fig. 1). A total of 1167 patients were excluded based on the exclusion criteria, and 490 patients were eligible for the present study. The non-liver injury and liver injury groups comprised 441 and 49 patients, respectively. In the liver injury group, the most common type was hepatocellular injury (29/49, 59.2%), followed by mixed cholestatic-hepatocellular injury (11/49, 22.4%) and cholestatic-type injury (9/49, 18.4%). The baseline characteristics and clinical laboratory data of the non-liver injury and liver injury groups are presented in Table 1. As for the CTRX dosage, 2 g/d (85.1%, 417/490) was the most common, followed by 1 g/d (11.2%, 55/490) and 4 g/d (3.7%, 18/490). Univariate analysis indicated that the administration of 4 g/d (p = 0.026) may contribute to the development of CTRX-induced liver injury. There were no significant differences in other clinical laboratory data, concomitant medications, or medical histories between the non-liver injury and liver injury groups. After 1 : 1 propensity score matching, both the non-liver injury and liver injury groups had 46 patients each. Table 1 shows the baseline characteristics of patients in the non-liver injury and liver injury groups after propensity score matching. No significant differences were observed between the non-liver injury and liver injury groups. However, there were several factors with a standardized difference greater than 0.1. The cut-off values for ALT levels and ALBI score were 24 IU/L (sensitivity: 0.388, specificity: 0.744, AUC: 0.555) and −1.61 (sensitivity: 0.429, specificity: 0.729, AUC: 0.501), respectively. As shown in Table 2, the COX regression analysis revealed that the adjusted hazard ratio (HR) for CTRX-induced liver injury in patients with ALBI score ≥−1.61 was 1.83 (95% coefficient interval (CI): 1.03–3.22, p = 0.04) in model 1. In model 2, ≥4 g/d was an independent risk factor for liver injury with adjusted HR at 2.70 (95% CI: 1.05–6.96, p = 0.04). In model 3, the adjusted HR for ALBI score ≥−1.61 was 1.82 (95% CI: 1.03–3.22, p = 0.04). In model 4, the adjusted HR for ALBI score ≥−1.61 was 1.82 (95% CI: 1.03–3.22, p = 0.04).

Fig. 1. Flow Diagram Illustrating the Patient Recruitment Process

Table 1. Baseline of Patient Characteristics between Non-liver Injury and Liver Injury Groups

Factors	Before propensity score matching				After propensity score matching
	Non-liver injury	Liver injury	p-Value	Std diff	Non-liver injury	Liver injury	p-Value	Std diff
	n = 441	n = 49			n = 46	n = 46
Basic property
Sex (male/female)	246/195	25/24	0.547a)	0.094	20/26	22/24	0.834a)	0.101
Age (years)	83 (74, 89)c) [24–102]d)	84 (72, 87)c) [30–99]d)	0.626b)	0.082	83 (70, 88)c) [25–96]d)	84 (74, 88)c) [30–99]d)	0.662b)	0.101
Body weight (kg)	49.0 (41.0, 59.8)c) [24.1–134]d)	49.0 (41.6, 58.9)c) [23.7–84.0]d)	0.623b)	0.038	51.5 (43.0, 60.9)c) [29.2–94.1]d)	48.4 (41.6, 59.2)c) [23.7–84.0]d)	0.482b)	0.135
Height (cm)	157 (150, 165)c) [125–186]d)	155 (150, 164)c) [132–178]d)	0.592b)	0.069	155 (148, 161)c) [125–175]d)	155 (150, 165)c) [132–178]d)	0.827b)	0.071
Body surface area (m2)	1.45 (1.32, 1.63)c) [0.94–2.55]d)	1.47 (1.33, 1.63)c) [0.95–1.96]d)	0.646b)	0.048	1.50 (1.34, 1.63)c) [1.03–2.21]d)	1.45 (1.33, 1.64)c) [0.95–1.96]d)	0.788b)	0.065
Clinical laboratory data
Scr (mg/dL)	0.89 (0.66, 1.31)c) [0.32–11.44]d)	0.99 (0.77, 1.60)c) [0.33–5.77]d)	0.076b)	0.303	1.05 (0.64, 1.82)c) [0.33–6.20]d)	1.01 (0.79, 1.65)c) [0.33–5.77]d)	0.359b)	0.217
Ccr (mL/min)	40.4 (25.5, 59.4)c) [4.5–193.8]d)	40.4 (19.4, 56.9)c) [3.9–151.3]d)	0.384b)	0.103	37.6 (22.5, 59.2)c) [7.75–147.7]d)	36.9 (18.6, 52.3)c) [3.9–151.3]d)	0.421b)	0.181
BUN (mg/dL)	20.2 (13.9, 30.7)c) [4.3–147.0]d)	20.9 (13.3, 39.8)c) [6.6–167.5]d)	0.470b)	0.194	20.9 (11.6, 31.2)c) [5.9–86.3]d)	21.4 (13.5, 40.9)c) [6.6–167.5]d)	0.353b)	0.265
ALT (IU/L)	16 (11, 24)c) [3–82]d)	17 (12, 27)c) [7–55]d)	0.130b)	0.103	15 (12, 20)c) [6–45]d)	17 (12, 25)c) [7–55]d)	0.106b)	0.370
AST (IU/L)	24 (18, 34)c) [7–317]d)	24 (19, 32)c) [9–125]d)	0.757b)	0.034	23 (18, 33)c) [8–61]d)	24 (19, 31)c) [9–125]d)	0.494b)	0.192
ALP IFCC (IU/L)	84 (66, 108)c) [29–225]d)	89 (72, 110)c) [48–194]d)	0.431b)	0.081	90 (74, 112)c) [45–161]d)	89 (72, 109)c) [48–194]d)	0.553b)	0.103
Alb (g/dL)	3.1 (2.7, 3.6)c) [1.3–4.7]d)	3.1 (2.5, 3.6)c) [1.6–4.5]d)	0.856b)	0.020	2.9 (2.3, 3.5)c) [1.7–4.1]d)	3.3 (2.5, 3.6)c) [1.6–4.5]d)	0.150b)	0.336
T-Bil (mg/dL)	0.7 (0.5, 1.0)c) [0.2–4.1]d)	0.7 (0.4, 1.0)c) [0.3–3.4]d)	0.524b)	0.015	0.7 (0.5, 1.0)c) [0.2–2.7]d)	0.7 (0.4, 1.0)c) [0.3–3.4]d)	0.944b)	0.037
ALBI score	−1.93 (−2.28, −1.55)c) [−3.38–−0.99]d)	−1.89 (−2.28, −1.48)c) [−3.02–−0.61]d)	0.985b)	0.023	−1.70 (−2.26, −1.32)c) [−2.87–−0.75]d)	−2.02 (−2.40, −1.48)c) [−3.02–−0.61]d)	0.177b)	0.325
CRP (mg/dL)	4.9 (1.9, 9.9)c) [0.1–41.0]d)	5.3 (1.5, 10.7)c) [0.1–27.7]d)	0.929b)	0.064	6.3 (2.4, 13.6)c) [0.1–41.0]d)	5.0 (1.3, 12.2)c) [0.1–27.7]d)	0.410b)	0.168
WBC (×102/µL)	87.7 (62.2, 122.4)c) [8.8–1125.0]d)	93.3 (64.3, 136.2)c) [26.7–237.4]d)	0.361b)	0.052	89.3 (64.3, 125.0)c) [29.6–346.9]d)	90.1 (65.3, 135.2)c) [26.7–237.4]d)	0.922b)	0.120
PLT (×104/µL)	20.0 (14.9, 27.0)c) [0.7–67.2]d)	22.0 (17.2, 26.9)c) [3.7–58.8]d)	0.338b)	0.134	21.8 (17.9, 28.6)c) [0.7–45.3]d)	22.3 (17.8, 28.2)c) [3.7–58.8]d)	0.873b)	0.056
Administration of CTRX
Duration of administration (d)	7 (6, 10)c)	7 (5, 10)c)	0.732b)	0.073	8 (6, 10)c)	7 (5, 10)c)	0.458b)	0.023
Total dosage (g)	14.4 (10.7, 19.2)c) [5.0–96.0]d)	14.9 (10.5, 21.0)c) [4.0–96.0]d)	0.518b)	0.157	14.9 (11.4, 20.0) c) [5.0–66.0]d)	15.0 (10.5, 21.0)c) [5.0–68.0]d)	0.919b)	0.005
Dosage of CTRX (g/d)	2 (2, 2)c)	2 (2, 2)c)	0.320b)	0.217	2 (2, 2)c)	2 (2, 2)c)	0.992b)	0.030
4 g /d, n (%)	13 (2.9)	5 (10.2)	0.026a)	0.292	5 (10.9)	4 (8.7)	1.000a)	0.074
2 g /d, n (%)	379 (85.9)	38 (77.6)	—	0.216	34 (73.9)	36 (78.3)	—	0.094
1 g /d, n (%)	49 (11.1)	6 (12.2)	—	0.031	7 (15.2)	6 (13.0)	—	0.058
Concomitant medications, n (%)
Acetaminophen	0 (0.0)	1 (2.0)	0.100a)	0.202	0 (0.0)	0 (0.0)	—	—
Clindamycin	29 (6.6)	1 (2.0)	0.344a)	0.022	0 (0.0)	1 (2.2)	1.000a)	0.202
Ampicillin	1 (0.2)	0 (0.0)	1.000a)	0.063	0 (0.0)	0 (0.0)	—	—
Total parenteral nutrition	10 (2.3)	2 (4.1)	0.341a)	0.102	3 (6.5)	2 (4.3)	1.000a)	0.112
Fat emulsion	2 (0.5)	1 (2.0)	0.271a)	0.135	1 (2.2)	1 (2.2)	1.000a)	<0.001
Site of infection, n (%)
Central nerve system	6 (1.4)	2 (4.1)	—	—	1 (2.2)	1 (2.2)	—	—
Respiratory tract	226 (51.2)	20 (40.8)	—	—	21 (45.7)	19 (41.3)	—	—
Gastrointestinal tract	24 (5.4)	2 (4.1)	—	—	2 (4.3)	1 (2.2)	—	—
Urinary tract	64 (14.5)	8 (16.3)	—	—	6 (13.0)	7 (15.2)	—	—
Others	26 (5.9)	4 (8.2)	—	—	1 (2.2)	6 (13.0)	—	—
Unknown	95 (21.5)	13 (26.5)	—	—	15 (32.6)	12 (26.1)	—	—
Medical history, n (%)
Hepatitis B	0 (0.0)	1 (2.0)	0.100a)	0.202	0 (0.0)	0 (0.0)	—	—
Hepatitis C	1 (0.2)	1 (2.0)	0.190a)	0.173	0 (0.0)	0 (0.0)	—	—
Cholecystitis	2 (0.5)	0 (0.0)	1.000a)	0.100	0 (0.0)	0 (0.0)	—	—
Cholangitis	4 (0.9)	1 (2.0)	0.411a)	0.135	0 (0.0)	1 (2.2)	0.100a)	0.063
Cholangiocarcinoma	0 (0.0)	1 (2.0)	0.100a)	0.092	0 (0.0)	0 (0.0)	—	—
Liver cancer	2 (0.5)	0 (0.0)	1.000a)	0.100	0 (0.0)	0 (0.0)	—	—
Liver cirrhosis	2 (0.5)	0 (0.0)	1.000a)	0.100	0 (0.0)	0 (0.0)	—	—
Alcoholic liver disease	2 (0.5)	0 (0.0)	1.000a)	0.100	0 (0.0)	0 (0.0)	—	—
Antibiotics allergy	14 (3.2)	1 (2.0)	1.000a)	0.075	1 (2.2)	1 (2.2)	1.000a)	<0.001

Alb: albumin. ALBI: albumin–bilirubin. ALP: alkaline phosphatase. ALT: alanine aminotransferase. AST: aspartate aminotransferase. BUN: blood urea nitrogen. Ccr: creatinine clearance. CRP: C-reactive protein. CTRX: ceftriaxone. PLT: platelet. Scr: serum creatinine. Std diff: standardized difference. T-Bil: total bilirubin. WBC: white blood cell. a) Fisher's exact test. b) Mann–Whitney U test. c) Each value represents the median (25, 75% percentile). d) Each value represents the minimum and maximum.

Table 2. Factors Affecting the Development of CTRX-Induced Liver Injury Using COX Regression Analysis

Factors	Model 1			Model 2			Model 3			Model 4
	AdjustedHR	95% CI	p-Value	AdjustedHR	95% CI	p-Value	AdjustedHR	95% CI	p-Value	AdjustedHR	95% CI	p-Value
Age (≥75 years)	0.87	0.46–1.65	0.68	0.93	0.49–1.76	0.81	0.87	0.46–1.64	0.67	0.87	0.46–1.64	0.67
Sex (male)	0.79	0.45–1.40	0.43	0.79	0.45–1.40	0.42	0.79	0.45–1.41	0.43	0.79	0.45–1.41	0.43
ALT (≥24 IU/L)	1.70	0.94–3.08	0.08	1.67	0.92–3.03	0.09	1.70	0.94–3.08	0.08	1.70	0.94–3.08	0.08
ALBI score (≥ −1.61)	1.83	1.03–3.22	0.04	1.73	0.98–3.07	0.06	1.82	1.03–3.22	0.04	1.82	1.03–3.22	0.04
Dosage (4 g/d)	—	—	—	2.70	1.05–6.96	0.04	—	—	—	—	—	—
Dosage (≥2 g/d)	—	—	—	—	—	—	0.93	0.39–2.18	0.86	—	—	—
Dosage (1 g/d)	—	—	—	—	—	—	—	—	—	1.08	0.46–2.55	0.86

ALBI: albumin–bilirubin. ALT: alanine aminotransferase. 95% CI: 95% coefficient interval. HR: hazard ratio.

The Kaplan–Meier curve analysis indicated that the cumulative risk for CTRX-induced liver injury was significantly higher in the ALBI score ≥−1.61 group (n = 142) than in the ALBI score <−1.61 group (n = 348) (p = 0.032 by log-rank test) (Fig. 2A). Conversely, after propensity score matching (p = 0.791) (Fig. 2B), there were no statistically significant differences between the 4 g/d (p = 0.163) (Fig. 3A), 2 g/d (p = 0.097) (Fig. 3B), and 1 g/d (p = 0.995) (Fig. 3C) dosage regimen groups.

Fig. 2. Kaplan–Meier Curves Showing the Cumulative Incidence of CTRX-Induced Liver Injury in Patients with Baseline ALBI Score ≥−1.61 and ALBI Score <−1.61

A: Before propensity score matching. B: After propensity score matching.

Fig. 3. Kaplan–Meier Curves Showing the Cumulative Incidence of CTRX-Induced Liver Injury in Patients with Baseline ALBI Scores ≥−1.61 and <−1.61 According to the CTRX Dosage

A: 4 g/d. B: 2 g/d. C: 1 g/d.

DISCUSSION

This study results revealed that the risk of CTRX-induced liver injury may differ depending on CTRX dosage.

The incidence of CTRX-induced liver injury was 10.0% (49/490), and it was consistent with that of a previous study (13.2%, 62/471).⁵⁾ In the present study, we collected data on concomitant common drugs and medical histories because they may contribute to liver injury development without CTRX administration. Although no significant differences were observed between the non-liver injury and liver injury groups for each factor (Table 1), further studies with larger sample sizes are needed because of the small sample size of these patients in the present study.

A high dose of CTRX may be an independent risk factor for liver injury, and this phenomenon was observed in the present study.⁵⁾ Although the ALBI score was not a statistically independent risk factor for liver injury in model 2 (p = 0.06), an ALBI score ≥−1.61 exhibits a tendency towards liver injury development at a dosage of 4 g/d.

The ALBI score may be involved in liver injury in patients treated with CTRX at ≥2 g/d in model 3 (Table 2), suggesting that decreased hepatic functional reserve may be associated with the development of liver injury at normal dosage. Similarly, the ALBI score may be involved in liver injury in patients treated with CTRX at 1 g/d in model 4 (Table 2). The most of the patients with pneumonia or urinary tract infections were administrated 2 g/d (78.6%),⁷⁾ and most patient in this study were administered 2 g/d. Therefore, risk assessment of liver injury using the ALBI score may be useful for patients treated with 2 g/d CTRX. Biliary excretion of CTRX is approximately 45%,²⁾ and non-renal clearance of CTRX is reduced by approximately 66% in patients with cirrhosis without ascites, compared with those with normal liver function.²⁾ Thus, it was speculated that in patients, the hepatic functional reserve decreases as a result of decreased biliary excretion of CTRX. Because there was no significant difference in creatinine clearance between the non-liver injury and liver injury groups (Table 1), the serum CTRX concentration in patients with low hepatic functional reserve treated with 1g/d and 2 g/d might have increased to the same level as that at 4 g/d. Therefore, further investigations are necessary using a COX regression analysis using the peak CTRX concentration as an explanatory factor.

Because an ALBI score ≥−1.61 was significantly associated with a cumulative risk of liver injury in all patients (Fig. 2A), the dose-dependent effect on this risk was elucidated. Consequently, although no significant difference in cumulative risk was observed in the 2 g/d group (Fig. 3B), an ALBI score ≥−1.61 was associated with a high risk of CTRX-induced liver injury. However, it was not fully clarified whether the 1 g/d and 4 g/d dosage regimens affect the cumulative risk of liver injury due to the small sample size.

We considered the possibility that confounding factors related to liver injury might not have been adequately corrected in the Cox regression analysis. Therefore, we attempted to correct them using propensity score matching. Although several factors had a standardized difference greater than 0.1, no significant difference in the cumulative risk of CTRX-induced liver injury between the ALBI score ≥−1.61 and ALBI score <−1.61 groups (log-rank test p = 0.791) was observed. However, further study is needed because the accuracy of propensity score matching remains an issue in small population sample sizes.

Our previous reports have revealed that the ALBI score may be risk factor for micafungin-,¹¹⁾ doripenem-,¹²⁾ ampicillin/sulbactam-¹³⁾ induced liver injury. Moreover, the ALBI score is used for the prediction of mortality in valvular surgery,¹⁷⁾ heart failure,¹⁸⁾ and idiopathic dilated cardiomyopathy.¹⁹⁾ With these evidences into consideration, ALBI score can be applied to non-HCC patients. The Child–Pugh score has been widely used for the assessment of prognosis with the combination of liver cirrhosis, which contains ascites, hepatic encephalopathy, prothrombin time, albumin, and bilirubin.²⁰⁾ Because the degree of ascites and hepatic encephalopathy is assessed by a subjective assessment index, there is a possibility that the grade evaluation may differ depending on the evaluator. In general, it is difficult for healthcare workers, such as pharmacists, to accurately evaluate the degree of ascites and hepatic encephalopathy except for physicians. Therefore, the ALBI score is a simpler index for evaluating hepatic functional reserve than the Child–Pugh score and is potentially useful for the risk assessment of CTRX-induced liver injury by all healthcare workers in clinical settings.

This study had some limitations. First, this study had a single-center observational design, and biases existed owing to the clinical departments and treatment policies. Second, as 1167 patients were excluded from this study, there may have been a large selection bias. Third, since 760 patients who received treatment for less than 5 d were excluded from the study, drug-induced liver injury that developed in less than 5 d may have been overlooked. Thus, further studies with a wider observation period are needed. Fourth, because older patients were the majority of participants in this study, further studies with younger patients are needed. Fifth, liver injury was caused by the patient’s medical history and concomitant drugs could not be completely ruled out. Sixth, none of the patients in the liver injury group were diagnosed with CTRX-induced liver injury by a physician. Finally, we could not evaluate hepatic functional reserve indices other than the ALBI score.

CONCLUSION

The ALBI score (≥−1.61) may be a risk factor for the development of liver injury in patients treated with CTRX. Therefore, frequent monitoring of the liver function is necessary in these patients.

Acknowledgments

The present study was funded by Grants from the Japan Society for the Promotion of Science (JSPS) Grants-in-Aid for Scientific Research (KAKENHI) (Grant No. 22K15331).

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Treeprasertsuk S, Huntrakul J, Ridtitid W, Kullavanijaya P, Björnsson ES. The predictors of complications in patients with drug-induced liver injury caused by antimicrobial agents. Aliment. Pharmacol. Ther., 31, 1200–1207 (2010).
• 2) Stoeckel K, Tuerk H, Trueb V, McNamara PJ. Single-dose ceftriaxone kinetics in liver insufficiency. Clin. Pharmacol. Ther., 36, 500–509 (1984).
• 3) Richards DM, Heel RC, Brogden RN, Speight TM, Avery GS. Ceftriaxone: a review of its antibacterial activity, pharmacological properties and therapeutic use. Drugs, 27, 469–527 (1984).
• 4) Akimoto H, Nagashima T, Minagawa K, Hayakawa T, Takahashi Y, Asai S. Signal detection of potential hepatotoxic drugs: case-control study using both a spontaneous reporting system and electronic medical records. Biol. Pharm. Bull., 44, 1514–1523 (2021).
• 5) Nakaharai K, Sakamoto Y, Yaita K, Yoshimura Y, Igarashi S, Tachikawa N. Drug-induced liver injury associated with high-dose ceftriaxone: a retrospective cohort study adjusted for the propensity score. Eur. J. Clin. Pharmacol., 72, 1003–1011 (2016).
• 6) Raemy S, Casanova C, Baldan R, Barreto E, Tande AJ, Endimiani A, Leib SL, Fischer U, Sendi P. Penicillin-susceptible Streptococcus pneumoniae meningitis in adults: does the ceftriaxone dosing matter? Antibiotics. (Basel), 12, 878 (2023).
• 7) Muhammed OS, Nasir BB. Drug use evaluation of ceftriaxone in Ras-Desta Memorial General Hospital, Ethiopia. Drug. Healthc. Patient. Saf., 12, 161–168 (2020).
• 8) Kawata T, Ikeda A, Masuda H, Komatsu S. Association between albumin–bilirubin score at admission and in-hospital mortality in patients with acute heart failure. Int. Heart J., 62, 829–836 (2021).
• 9) Toyoda H, Johnson PJ. The ALBI score: from liver function in patients with HCC to a general measure of liver function. JHEP. Rep, 4, 100557 (2022).
• 10) Johnson PJ, Berhane S, Kagebayashi C, Satomura S, Teng M, Reeves HL, O’Beirne J, Fox R, Skowronska A, Palmer D, Yeo W, Mo F, Lai P, Iñarrairaegui M, Chan SL, Sangro B, Miksad R, Tada T, Kumada T, Toyoda H. Assessment of liver function in patients with hepatocellular carcinoma: a new evidence-based approach—the ALBI grade. J. Clin. Oncol., 33, 550–558 (2015).
• 11) Asai Y, Yamamoto T, Sato Y. Risk assessment of micafungin-induced liver injury using spontaneous reporting system data and electronic medical records. J. Infect. Chemother., 28, 690–695 (2022).
• 12) Asai Y, Ooi H, Sato Y. Risk evaluation of carbapenem-induced liver injury based on machine learning analysis. J. Infect. Chemother., 29, 660–666 (2023).
• 13) Ooi H, Asai Y, Sato Y. Risk evaluation of ampicillin/sulbactam-induced liver injury based on albumin–bilirubin score. J. Infect. Chemother., 29, 900–904 (2023).
• 14) Watanabe M, Shibuya A. Validity study of a new diagnostic scale for drug-induced liver injury in Japan-comparison with two previous scales. Hepatol. Res., 30, 148–154 (2004).
• 15) Ma T, Li QS, Wang Y, Wang B, Wu Z, Lv Y, Wu RQ. Value of pretransplant albumin–bilirubin score in predicting outcomes after liver transplantation. World J. Gastroenterol., 25, 1879–1889 (2019).
• 16) deLemos AS, Ghabril M, Rockey DC, Gu J, Barnhart HX, Fontana RJ, Kleiner DE, Bonkovsky HL. Amoxicillin–clavulanate-induced liver injury. Dig. Dis. Sci., 61, 2406–2416 (2016).
• 17) Duman ZM, Timur B. Albumin–bilirubin score: a novel mortality predictor in valvular surgery. Braz. J. Cardiovasc. Surg., 38, 271–277 (2023).
• 18) Han S, Wang C, Tong F, Li Y, Li Z, Sun Z, Sun Z. Prognostic impact of albumin–bilirubin score on the prediction of in-hospital mortality in patients with heart failure: a retrospective cohort study. BMJ Open, 12, e049325 (2022).
• 19) Jiang M, Wei XB, Huang JL, Su ZD, Lin YW, Yu DQ. Albumin–bilirubin score to predict outcomes in patients with idiopathic dilated cardiomyopathy. Arq. Bras. Cardiol., 118, 1108–1115 (2022).
• 20) Peng Y, Qi X, Guo X. Child–Pugh versus MELD score for the assessment of prognosis in liver cirrhosis: a systematic review and meta-analysis of observational studies. Medicine. (Baltimore), 95, e2877 (2016).