Risk factors for acute fatty liver of pregnancy complicated by acute liver failure: a retrospective study of 133 cases

Abstract

Objective

To investigate the clinical manifestations and trends in laboratory indices in patients with acute fatty liver of pregnancy (AFLP) and identify risk factors for AFLP complicated by acute liver failure (ALF) to improve perinatal outcomes for these patients and their infants.

Methods

Clinical data, laboratory indices, and maternal and fetal outcomes for 133 patients with AFLP treated at our institution between January 1, 2010 and October 1, 2022 were retrospectively analyzed.

Results

The most frequent clinical manifestations of AFLP were anorexia (n = 59, 44.4%), nausea (n = 41, 30.8%), and vomiting (n = 35, 26.3%). Liver function tests were significantly elevated, including serum alanine aminotransferase (287.0 ± 280.0 U/L), aspartate aminotransferase (254.0 ± 245.9 U/L), total bilirubin (99.1 ± 7.4 µmol/L), and direct bilirubin (84.3 ± 5.9 µmol/L). There was also marked coagulopathy, with a fibrinogen level of 1.7 ± 1.0 g/L and a prothrombin time of 16.7 ± 7.5 s. The main complication was acute liver failure (n = 33, 24.8%). Multivariate logistic regression analysis identified total bilirubin (OR 1.01, 95% CI 1.01–1.02, P < 0.001), direct bilirubin (OR 1.01, 95% CI 1.01–1.02, P < 0.001), prothrombin time (OR 1.20, 95% CI 1.09–1.32, P < 0.001), and activated partial thromboplastin time (OR 1.02, 95% CI 1.00–1.04, P = 0.017) to be risk factors for acute liver failure in patients with AFLP.

Conclusions

Anorexia, nausea, and vomiting were the most common clinical manifestations of AFLP. Liver dysfunction and coagulopathy were indicated by changing trends in laboratory markers. Deterioration of liver function and coagulation parameters may be risk factors for acute liver failure in AFLP. Close monitoring of liver function and coagulation in patients with suspected or confirmed AFLP is essential to prevent this potentially lethal complication.

Graphical Abstract

Introduction

Acute fatty liver of pregnancy (AFLP) is a rare but potentially fatal obstetric emergency, which typically occurs at 28–40 weeks’ gestation [1], with the majority of cases occurring at approximately gestational week 35. AFLP is characterized by fatty infiltration of the liver that may be complicated by acute liver failure [2] (ALF) and has a reported prevalence of 1/16,000 to 1/6000 [3]. There are no typical early clinical signs of AFLP and no specific clinical manifestations. AFLP may present with anorexia, nausea, vomiting, epigastric pain, fatigue, and other symptoms. As the disease progresses, patients may develop coagulopathy and ALF or damage to other organs, including the kidneys, pancreas, digestive tract, and brain [4]. More serious complications include disseminated intravascular coagulation and multiple organ dysfunction syndrome, which are associated with high maternal and fetal mortality [5]. AFLP also has a detrimental impact on the infant in the perinatal period, causing premature birth, jaundice, hypoxic-ischemic encephalopathy, neonatal asphyxia, and death. The pathological features of AFLP are acute microvesicular steatosis in hepatocytes, which impedes synthesis of hepatic cholesterol, fibrinogen, and coagulation factors while also impairing binding and clearance of bilirubin [6]. The pathogenesis of AFLP remains unclear but may involve changes in maternal hormone levels during pregnancy and variations in genes for long-chain 3-hydroxyacyl-CoA dehydrogenase in the fetus [7, 8]. There is some research showing that estrogen plays a role in the pathogenesis of AFLP [9]. At present, there is no specific effective treatment for AFLP other than immediate delivery. However, supportive treatment can reduce the risk of maternal and fetal complications and mortality and improve the prognosis for both mothers and infants. In recent years, heightened awareness of AFLP among obstetricians has resulted in increased detection and incidence rates, leading to significant improvements in maternal and fetal outcomes.

The aim of this study was to investigate the clinical manifestations of AFLP and laboratory indices in patients with the condition as well as risk factors for ALF as a complication of AFLP to improve perinatal outcomes.

Material and methods

Study population

A retrospective search of the medical records at our institution identified 277 patients with a suspected diagnosis of AFLP between January 1, 2010 and October 1, 2022. After exclusion of 144 patients in whom the final diagnosis was not AFLP and those with incomplete data, 133 patients with a confirmed diagnosis of AFLP and essentially complete clinical and laboratory data were included in the study.

The study protocol was approved by our hospital ethics committee. Informed consent for participation in the study was obtained from patients and their families after they had received a detailed explanation of the purpose of the study and the methodology used.

Diagnostic criteria

In all cases, AFLP was diagnosed using the Swansea criteria introduced in 2002. According to these criteria, a diagnosis of AFLP is made if six or more of the following 14 features are present in the absence of other explanations: vomiting, abdominal pain, polyuria/polydipsia, hepatic encephalopathy, elevated bilirubin (> 14 µmol/L), hypoglycemia (< 4 mmol/L), elevated uric acid (> 340 µmol/L), an elevated white blood cell count (> 11 × 10⁹/L), ascites or “bright liver” on an ultrasound scan, elevated transaminases (> 42 U/L), an increased ammonia level (> 7 µmol/L), impaired renal function (creatinine > 150 µmol/L), coagulopathy (prothrombin time [PT] > 14 s or activated partial thromboplastin time [APTT] > 34 s), and liver tissue biopsy showing microvesicular steatosis. Patients with hepatitis, hepatotoxicity, biliary tract disease, or cholestasis in pregnancy were excluded. All patients underwent systematic exclusion of viral hepatitis, including hepatitis E virus (HEV) testing via IgM/IgG serology (Wantai ELISA) and HEV RNA PCR (Artus HEV RT-PCR Kit). Patients with positive HEV markers were excluded from the AFLP cohort.

The diagnosis of AFLP complicated by ALF was supported by: (1) third-trimester onset with concurrent preeclampsia features (blood pressure > 140/90 mmHg and proteinuria ≥ 300 mg/24 h), (2) characteristic biochemical profile (uric acid > 8 mg/dL, AST/ALT ratio < 1.5), and (3) absence of HEV/other hepatotropic virus markers [10].

The clinical suspicion for ALF development in AFLP patients was triggered by any of the following:

1. Progressive coagulopathy (INR increase ≥ 0.3 within 24 h),

2. Neurocognitive impairment (disorientation or asterixis),

3. Metabolic acidosis (pH < 7.3 with lactate > 4 mmol/L),

4. Hepatorenal syndrome (serum creatinine doubling within 48 h) [11].

Methodology

Background information was collected, including patient age, number of pregnancies, parity, and week of pregnancy at onset of clinical symptoms of AFLP. Clinical symptoms included fatigue, headache, anorexia, nausea, vomiting, thirst, polyuria, epigastric pain, and epigastric tenderness. Mean values for laboratory indices, namely, white blood cells, total bilirubin, direct bilirubin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), uric acid, serum creatinine, blood glucose, PT, APTT, prothrombin time activity (PT%), and fibrinogen before and after delivery were calculated. The methods used to image the liver during the study period were b-ultrasonography and computed tomography (CT). Information was collected on duration and efficacy of treatment and on maternal complications, including acute kidney injury, ALF, postpartum hemorrhage, respiratory distress, pulmonary infection, heart failure, acute pancreatitis, and gastrointestinal bleeding. Pregnancy outcomes included gestational age at delivery, mode of delivery, and mortality. Perinatal outcomes included sex, birth weight, Apgar scores at 1, 5, and 10 min, and perinatal mortality.

Statistical analysis

Continuous data that were normally distributed are shown as the mean ± standard deviation and were compared between before and after treatment using the paired t-test. Data that were not normally distributed are shown as the median (interquartile range) and were compared using a nonparametric test. Categorical data are summarized as the number (percentage) and were compared using the chi-squared test. Multivariate logistic regression models were used to adjust for the effects of covariates on the odds ratio (OR). The statistical analyses were performed using SPSS version 23.0 software (IBM Corp., Armonk, NY, USA). A P-value < 0.05 was considered statistically significant and a P-value < 0.01 was considered highly statistically significant.

Results

Clinical, laboratory, and imaging findings

Table 1 summarizes the patients’ pre-delivery clinical, laboratory, and imaging data based on the Swansea criteria. The most common clinical manifestations were anorexia (n = 59, 44.4%), nausea (n = 41, 30.8%), and vomiting (n = 35, 26.3%). The most significant changes in laboratory parameters were in bilirubin (n = 119, 89.5%) and transaminases (n = 114, 85.7%). Abnormalities were seen on liver CT scans in four patients and on ultrasound scans in 20.

Table 1.

Clinical manifestations and laboratory and imaging findings that met the Swansea criteria in patients with acute fatty liver of pregnancy

Variate	Research standards	Quantities	Percentage
Clinical symptom
Nausea	Positive	41	30.8%
Vomiting	Positive	35	26.3%
Epigastric pain	Positive	19	14.3%
Parched	Positive	11	8.3%
Polyuria/drinking	Positive	8	6.0%
Hepatic encephalopathy	Positive	2	1.5%
Laboratory indicators
Elevated bilirubin	> 14 μmol/L	119	89.5% ↑↑↑
Glucose reduction	< 4 mmol/L	56	42.1% ↑↑
Elevated uric acid	> 340 μmol/L	115	86.5% ↑↑
Leucocytosis	> 11 × 10 L9	88	66.2% ↑↑
Elevated transaminases	> 42 U/L	114	85.7% ↑↑↑
Elevated blood ammonia	> 47 μmol/L	89	66.9% ↑↑
Elevated creatinine	> 150 μmol/L	43	32.3% —
PT extension	> 14 s	86	64.7% ↑↑↑
APTT prolongation	> 34 s	106	79.7% ↑↑
Imaging and Pathologic Manifestations
Liver biopsy	microvesicular cell degeneration	0	0
CT scan of the liver	fatty liver	4	3.0%
Ultrasonic scan	hyperechoic liver on ultrasound	20	15.0%
Other clinical symptoms
Anorexia	Positive	59	44.4%
Fatigue	Positive	26	19.6%
Headache	Positive	2	1.5%

Liver biopsy has potential hazards so is not recommended in patients with AFLP, all of whom have various degrees of coagulation damage

APTT activated partial thromboplastin time, CT computed tomography, PT prothrombin time

The main laboratory results are shown in Table 2. The mean white blood cell count was 15.0 ± 8.8 (5.1–86.6), ALT was 287.0 ± 280.0 U/L (14.0–1568.1), AST was 254.0 ± 245.9 U/L (5.6–1832.3), total bilirubin was 99.1 ± 7.4 µmol/L (4.7–401.5), and direct bilirubin was 84.3 ± 5.9 µmol/L (4.1–288.1). The PT increased to 16.7 ± 7.5 (8.5–80.0), the APTT increased to 45.6 ± 45.6 s (18.4–138.5), and the plasma fibrinogen level decreased to 1.7 ± 1.0 g/L (0.2–6.8).

Table 2.

Results of laboratory function tests in 133 patients with acute fatty liver of pregnancy

Variate	Realm	Mean (x ± s)	Reference interval	Exceptions (percentage)
WBC (× 109 L)	5.1–86.6	15.0 ± 8.8	3.5–9.5	103 (77.4)
Alanine aminotransferase (U/L)	14.0–1568.1	287.0 ± 280.0	7–40	119 (89.5)
Aspartate aminotransferase (U/L)	5.6–1832.3	254.0 ± 245.9	7–40	126 (94.7)
Blood ammonia(μmol/L)	8.6–168.8	66.6 ± 23.6	11–51	86 (64.7)
Blood glucose(mmol/L)	2.1–15.6	4.6 ± 1.8	3.9–6.1	77 (57.9)
Total bilirubin(μmol/L)	4.7–401.5	99.1 ± 7.4	0–21.0	114 (85.7)
Direct bilirubin (μmol/L)	4.1–288.1	84.3 ± 5.9	0–8.0	127 (95.5)
Serum creatinine(μmol/L)	34.0–823.0	144.7 ± 82.4	41–73	125 (94.0)
Urea nitrogen(mmol/L)	0.7–19.3	7.2 ± 3.2	2.6–7.5	52 (39.1)
Uric acid(μmol/L)	86.0–1123.0	500.5 ± 147.1	155–357	118 (88.7)
Prothrombin time (s)	8.5–80.0	16.7 ± 7.5	8.8–13.8	90 (67.7)
Activated Partial Thromboplastin Time (s)	18.4–138.5	45.6 ± 45.6	28–42	66 (49.6)
Fibrinogen (g/L)	0.2–6.8	1.7 ± 1.0	2.0–4.0	98 (73.7)

WBC white blood cells

Complications and perinatal outcomes

Table 3 shows the demographic characteristics, complications, and perinatal outcomes. The most frequent complication was ALF (n = 33, 24.8%). There were two cases of multiple organ dysfunction syndrome (1.5%). The average patient age was 30.16 ± 5.00 years. Fifty-three (39.9%) of the 133 patients were pregnant for the first time and 67 (50.4%) were giving birth for the first time. Ninety-seven patients (72.9%) were treated for AFLP for more than 7 days. The mean gestational age at delivery was 36.27 ± 3.10 weeks and the average interval between symptom onset and delivery was 6.02 ± 8.03 days. Delivery was by cesarean section in 106 patients (79.3%) and by vaginal delivery in 27 (20.3%). All 33 AFLP patients complicated by ALF underwent cesarean section for pregnancy termination, while none of the patients who delivered via the vaginal mode developed ALF. There were 112 singleton pregnancies (84.2%) and 21 multiple pregnancies (15.8%). Five of the 133 patients died (maternal mortality, 3.8%). The average birth weight was 2501.4 ± 675.3 g, and the mean Apgar scores at 1, 5, and 10 min were 9.1 ± 2.3, 8.6 ± 2.5, and 7.5 ± 2.8, respectively. There were 85 male neonates (59.9%) and 57 female neonates (40.1%). Nine of the 142 fetuses died (fetal mortality, 6.3%).

Table 3.

Complications and perinatal outcomes in 133 patients with acute fatty liver of pregnancy and their fetuses

Variate	Average value/number	Percentage
Demographic characteristics
Age (years)	30.16 ± 5.00
Gestation period
1	53	39.9
2	33	24.8
≥ 3	47	35.3
The time of birth
0	67	50.4
1	49	36.8
≥ 2	17	12.8
Number of days of treatment (days)
< 7	36	27.1
≥ 7	97	72.9
Complications
Liver failure	33	24.8
Acute kidney injury	27	20.3
Lung infection	13	9.8
Postpartum hemorrhage	12	9.0
Heart failure	7	5.3
Gastrointestinal bleeding	3	2.3
Respiratory distress	2	1.5
Acute pancreatitis	2 -	1.5
MODS	2	1.5
Pregnancy outcome
Average week of pregnancy termination (weeks)	36.27 ± 3.10
Average time from onset of symptoms to termination of pregnancy (days)	6.02 ± 8.03
Mode of termination of pregnancy
Cesarean section	106	79.7
Give birth without complications	27	20.3
Type of pregnancy
Singleton pregnancy	112	84.2
Multiple pregnancy	21	15.8
Survive	128	96.2
Dead	5	3.8
Fetal outcome
Distinguishing between the sexes
Male	85	59.9
Women	57	40.1
Average birth weight (g)	2501.4 ± 675.3
Mean 1 min Apgar score	9.1 ± 2.3
Average 5 min Apgar score	8.6 ± 2.5
Mean 10 min Apgar score	7.5 ± 2.8
Survive	133	93.7
Dead	9	6.3

MODS multiple organ dysfunction syndrome

Univariate analysis

The results of the univariate analysis of risk factors for ALF in AFLP are shown in Table 4. Thirty-three (24.8%) of the 133 patients developed ALF. Gestational week at onset (P = 0.030), fatigue (P = 0.032), ALT (P < 0.001), AST (P = 0.002), total bilirubin (P < 0.001), direct bilirubin (P < 0.001), PT (P = 0.005), APTT (P = 0.009), and PT% (P < 0.001) were identified to be potential prognostic factors. Box plots for ALT, AST, PT, APTT, PT%, total bilirubin, and direct bilirubin are shown in Fig. 1.

Table 4.

Analysis of potential risk factors for acute liver failure in women with acute fatty liver of pregnancy

Variate	No acute liver failure	With acute liver failure	P-value
Demographic characteristics
Number of patients	100	33
Patient’s age	30.34 ± 4.98	29.61 ± 5.09	0.466
Gestational week at onset	35.78 ± 3.17	34.32 ± 3.75	0.030*
Gestational week for labor	36.54 ± 2.92	35.44 ± 3.52	0.077
Gestation period			0.951
1	40	13
≥ 2	60	20
The time of birth			0.392
0	48	19
1	40	9
≥ 2	12	5
Days from first symptom to delivery	5.35 ± 6.33	7.88 ± 11.66	0.242
Type of delivery			0.618
Cesarean section	81	28
Give birth without complications	19	5
Number of pregnancies			0.224
Singleton	82	30
Multiple births	18	3
Sex of the unborn child			0.411
A male	44	11
Females	32	16
Male/male	10	2
Female/female	0	0
Male/female	5	1
Clinical manifestation
Fatigue	16	11	0.032*
Headache	2	0	0.561
Anorexia	43	17	0.394
Nauseating	30	12	0.495
Vomiting	23	13	0.066
Parched	10	1	0.363
Polyuria	7	1	0.674
Abdominal pain	16	3	0.474
Laboratory tests
WBC	14.23 ± 9.18	17.52 ± 6.96	0.061
Alanine aminotransferase	328.31 ± 294.89	155.80 ± 163.77	< 0.001***
Aspartate aminotransferase	276.36 ± 254.28	163.50 ± 148.30	0.002 **
Total bilirubin	79.86 ± 76.10	157.59 ± 82.54	< 0.001***
Direct bilirubin	68.83 ± 61.44	131.42 ± 60.36	< 0.001***
Blood glucose	4.64 ± 1.78	4.51 ± 1.90	0.729
Serum creatinine	141.02 ± 85.95	155.70 ± 67.55	0.373
Uric acid	509.83 ± 133.89	471.30 ± 176.86	0.190
Urea nitrogen	7.05 ± 2.86	7.62 ± 4.15	0.466
Prothrombin Time	15.19 ± 4.32	21.42 ± 11.74	0.005 **
Activated Partial Thromboplastin Time	43.04 ± 18.20	53.66 ± 24.40	0.009 **
Prothrombin Time Activity	70.46 ± 26.77	45.91 ± 16.94	< 0.001***
Fibrinogen	1.78 ± 1.05	1.46 ± 0.85	0.112

WBC white blood cells

Fig. 1.

Box plots for the various laboratory indicators.. ALT, alanine aminotransferase; APTT, activated partial thromboplastin time; AST, aspartate aminotransferase; Dbil, direct bilirubin; Tbil, total bilirubin; PT, prothrombin time; PT%, prothrombin time activity

Multivariate logistic regression analysis

Multivariate logistic regression analysis identified total bilirubin (OR 1.01, 95% CI 1.01–1.02, P < 0.001), direct bilirubin (OR 1.01, 95% CI 1.01–1.02, P < 0.001), PT (OR 1.20, 95% CI 1.09–1.32, P < 0.001), and APTT (OR 1.02, 95% CI 1.00–1.04, P = 0.017) to be risk factors for ALF in patients with AFLP (Table 5).

Table 5.

Risk factors for acute liver failure in patients with acute fatty liver of pregnancy identified by multivariable logistic regression analysis

Variate	B	SE	Wald	Df	OR	95% CI		P
Gestational week at onset	−0.12	0.06	4.42	1.00	0.88	0.79	0.99	0.036*
Fatigue	−0.97	0.46	4.42	1.00	0.38	0.15	0.94	0.036*
Alanine aminotransferase	0.00	0.00	8.82	1.00	1.00	0.99	1.00	0.003**
Aspartate aminotransferase	0.00	0.00	5.89	1.00	1.00	0.99	1.00	0.015*
Total bilirubin	0.01	0.00	17.05	1.00	1.01	1.01	1.02	< 0.001***
Direct bilirubin	0.01	0.00	18.13	1.00	1.01	1.01	1.02	< 0.001***
Prothrombin time	0.18	0.05	14.11	1.00	1.20	1.09	1.32	< 0.001***
Activated partial thromboplastin time	0.02	0.01	5.69	1.00	1.02	1.00	1.04	0.017*
Prothrombin time activity	−0.06	0.01	17.46	1.00	0.94	0.91	0.97	< 0.001***

CI confidence interval, OR odds ratio

Receiver-operating characteristic curves for laboratory indicators

The receiver-operating characteristic curves for ALT, AST, PT, APTT, PT%, total bilirubin, and direct bilirubin are shown in Fig. 2. In the total study population, the area under the curve for diagnosis of AFLP complicated by ALF was 0.80 (95% CI 0.72–0.88; P < 0.001) for total bilirubin, 0.79 (95% CI 0.70–0.88; P < 0.001) for direct bilirubin, and 0.77 (95% CI 0.68–0.86; P < 0.001) for PT. The optimal cut-off point was 61.58 for total bilirubin (sensitivity 0.97, specificity 0.62, Youden index 0.58), 87.20 for direct bilirubin (sensitivity 0.76, specificity 0.76, Youden index 0.52), and 16.70 for PT (sensitivity 0.69, specificity 0.76, Youden index 0.45).

Fig. 2.

Results of receiver-operating characteristic curve analysis of laboratory indicators. ALT, alanine aminotransferase; APTT, activated partial thromboplastin time; AST, aspartate aminotransferase; Dbil, direct bilirubin; Tbil, total bilirubin; PT, prothrombin time; PT%, prothrombin time activity

Discussion

The initial clinical manifestations of AFLP typically appear in late pregnancy in the form of digestive symptoms, primarily anorexia, nausea, vomiting, and epigastric pain [12]. However, a minority of pregnant patients may present with diarrhea, epigastric discomfort, and pruritus [13, 14]^. Although all the patients in our study met the Swansea criteria for AFLP, relatively few had clinical manifestations of nausea, vomiting, epigastric pain, thirst, polydipsia polyuria, and hepatic encephalopathy. Anorexia and fatigue are not included in the Swansea criteria but were common in our patients, especially anorexia, which was the first symptom in approximately half of the cases. Recent research has demonstrated that anorexia is becoming an increasingly common first symptom in patients with AFLP [15]. These observations indicate that the Swansea criteria have limitations; however, larger studies are needed to determine whether anorexia should be included among the common clinical symptoms of AFLP [16]. The clinical manifestations of AFLP show considerable variability and lack specificity, with new precursor symptoms continually being identified [17]. For example, Al-Husban et al. [18] noted that rash may be an initial symptom of AFLP. Therefore, a high degree of clinical suspicion is needed for timely identification and intervention in a parturient with a series of non-specific clinical symptoms, such as anorexia, fatigue, and rash, in the third trimester of pregnancy [19].

Liver function and coagulation tests should be performed for prevention and diagnosis of AFLP and to protect the life and health of the mother and fetus [20]. In this study, we focused on 133 patients with laboratory data available for two reasons. First, the Swansea criteria mention that patients with AFLP show particular trends in laboratory markers. Second, the diagnosis of AFLP was based on clinical manifestations combined with abnormal laboratory data. All our patients had laboratory abnormalities to some extent; 103 (77.4%) had increased white blood cells, 119 (89.5%) had increased ALT, 126 (94.7%) had increased AST, 114 (85.7%) had increased total bilirubin, 127 (95.5%) had increased direct bilirubin, 125 (94.0%) had increased serum creatinine, 118 (88.7%) had increased uric acid, 90 (67.7%) had a prolonged PT, and 98 (73.7%) had decreased plasma fibrinogen. According to the Swansea criteria, patients with AFLP tend to have abnormal liver enzyme levels, with markedly elevated total and direct bilirubin, decreased blood glucose, elevated creatinine, prolonged clotting time, and decreased plasma fibrinogen, which is very similar to our data. Therefore, this recommendation is helpful for diagnosis and prophylactic treatment of patients suspected to have AFLP and patients with non-specific symptoms.

Liver tissue biopsy and imaging examinations are mentioned in the Swansea criteria, but are difficult to perform in everyday clinical work. Liver tissue biopsy was not performed in our series for two reasons. First, there is an elevated risk of disseminated intravascular coagulation following liver tissue biopsy, potentially leading to subcapsular and intraperitoneal hemorrhage [21]. Second, AFLP typically presents as a life-threatening condition with a sudden onset, making reliance on liver tissue biopsy for pathological confirmation impractical and potentially delaying treatment. In our study, all 133 patients exhibited varying degrees of impaired liver function and coagulation at the time of disease onset. Therefore, none were diagnosed by liver tissue biopsy. Although liver imaging can aid in the diagnosis, AFLP has a rapid onset, its findings are usually not obvious, and there are certain limitations to imaging. Only twenty of the 133 patients with AFLP in this study had evidence of fatty liver changes on ultrasonography, with a further four patients showing only slight enhancement of the parenchyma on CT scans of the liver. Consequently, the clinical diagnosis of AFLP depends mainly on clinical manifestations and laboratory results with less reliance on liver tissue biopsy and imaging [22]. However, when clinically indicated and feasible, we recommend the utilization of CT and ultrasonography to both assess hepatic fat deposition and exclude space-occupying lesions, thereby enhancing the diagnostic accuracy of AFLP.

In this study, there were no significant demographic abnormalities. There is some evidence suggesting that first delivery, a male fetus, and a multiple pregnancy are risk factors for AFLP [23]. In our study, more than half of the patients with AFLP were delivering for the first time, and their fetuses were significantly more likely to be male than female. If further studies confirm this finding, first-time birth and a male fetus may have important implications for prevention of AFLP. In terms of perinatal outcomes, maternal mortality and fetal mortality were low, with five (3.8%) and nine (6.3%) cases, respectively. Currently, there is no specific treatment protocol for AFLP. Timely delivery is the most effective strategy for safeguarding the life and health of both mother and baby. In terms of the mode and timing of delivery, 106 of our patients (79.7%) delivered by cesarean section and 27 (20.3%) delivered naturally. The average interval between onset of symptoms and delivery was 6.02 ± 8.03 days. The method and timing of delivery in patients with AFLP are influenced by various factors, including fetal maturity and maternal condition. The pregnancy is typically ended by cesarean section rather than natural labor. Vaginal delivery is likely to increase energy expenditure in patients with AFLP, leading to further increases in fatty acid levels and worsening the burden on the liver. Consequently, rapid decompensation may occur in the woman or the fetus, further supporting the clinical preference for cesarean section.

AFLP poses a significant risk of severe and potentially fatal complications. In our study, the complications included acute liver failure (n = 33), acute kidney injury (n = 27), lung infection (n = 13), postpartum hemorrhage (n = 12), heart failure (n = 7), gastrointestinal bleeding (n = 3), respiratory distress (n = 2), acute pancreatitis (n = 2), and multiple organ dysfunction syndrome (n = 2). Among the five fatal cases in this study, two deaths were directly attributed to ALF, while the remaining three cases involved severe comorbidities, including severe eclampsia, gestational heart disease, and pulmonary infections. ALF is the most lethal complication of AFLP and was the most common complication in our study. Therefore, we conducted univariate and multiple logistic regression analyses, which identified elevated ALT, AST, total bilirubin, and direct bilirubin and prolonged PT to be potential risk factors for development of ALF. An early gestational week at onset (P = 0.036) and fatigue (P = 0.036) may also be associated with ALF in AFLP. Objectively, patients with an early gestational week at onset are more likely to have impaired liver function and coagulation. Recent research has shown that ALF is associated with reduced production of clotting factors and a reduction in procoagulant proteins, leading to coexistence of coagulopathy and hypercoagulability. This could be a plausible explanation for earlier onset of the disease and weeks of pregnancy being a risk factor for ALF. Further studies in larger samples are needed to confirm our findings. Liver support therapy is recognized as a simplified yet effective approach for managing ALF [24]. In this study, all 33 AFLP patients complicated by ALF showed clinical improvement following aggressive medical therapy, which may be attributed to the multidisciplinary collaboration (involving obstetrics, internal medicine, hepatobiliary surgery, and the Intensive Care Unit) implemented at our center. These findings underscore the critical role of multidisciplinary team (MDT)-based interventions in optimizing outcomes for AFLP patients complicated by ALF. We therefore strongly advocate for the systematic integration of MDT strategies in clinical protocols for such cases. Current international guidelines have established liver transplantation as the most effective treatment for ALF. For AFLP patients complicated by ALF who fail to respond to conservative medical management, prioritization of liver transplantation is strongly recommended in clinical settings with transplantation capabilities, provided the patient meets eligibility criteria. This approach has been demonstrated to significantly improve survival outcomes in critically ill patients with irreversible hepatic dysfunction [25].

Our ROC curve analysis showed that total bilirubin, direct bilirubin, and PT have some clinical value in the diagnosis of AFLP complicated by ALF and that fluctuation in total bilirubin has the highest accuracy for prevention of ALF in AFLP. Therefore, close monitoring of these laboratory indices, particularly liver function and coagulation, is imperative in these patients. Deterioration in these parameters may herald the onset or progression of ALF in AFLP and a need for timely intervention and management to mitigate adverse outcomes.

This study provides valuable insights into maternal and perinatal outcomes in AFLP patients complicated by ALF, offering potential strategies for optimized clinical management to mitigate ALF risks. A key strength lies in the exclusive inclusion of cases from a single tertiary care center, effectively reducing institutional bias associated with heterogeneous clinical protocols across multiple facilities.

However, several limitations should be acknowledged. First, the sample size remains constrained by the single-center design, potentially limiting statistical power for rare outcomes. Second, the data spanning a 12-year period (2010–2022) may have introduced heterogeneity in maternal and neonatal outcomes due to advancements in medical technology and evolving diagnostic/therapeutic protocols during this timeframe. Most notably, AFLP and hemolysis, elevated liver enzymes and low platelet count syndrome (HELLP syndrome) can both complicated by ALF. We have not to perform comparative analyses of ALF pathophysiology between AFLP and HELLP syndrome. Identification of the risk factors for AFLP complicated by ALF remains challenging, and multicenter studies in larger samples are needed to resolve this problem.

Conclusion

In summary, AFLP severely affects the health of both mother and fetus and is challenging in terms of diagnosis. Moreover, it has some serious and potentially fatal complications, especially ALF. Recognizing non-specific symptoms in the third trimester of pregnancy and detecting abnormal laboratory indices can aid recognition of AFLP and prevent its complications. When the diagnosis is confirmed, the clinician should intervene on the pregnancy in a timely manner to protect the life and health of both mother and infant and ultimately improve maternal and fetal outcomes.

Abbreviations

AFLP

Acute fatty liver of pregnancy

ALF

Acute liver failure

ALT

Alanine aminotransferase

APTT

Activated partial thromboplastin time

AST

Aspartate aminotransferase

CI

Confidence interval

CT

Computed tomography

Dbil

Direct bilirubin

HELLP syndrome

Hemolysis, elevated liver enzymes and low platelet count syndrome

HEV

Hepatitis E virus

MDT

Multidisciplinary team

MODS

Multiple organ dysfunction syndrome

OR

Odds ratio

PT

Prothrombin time

PT%

Prothrombin time activity

Tbil

Total bilirubin

WBC

White blood cells

Authors’ contributions

LJH, FWY and LSN contributed to data enrollment, statistical analysis, study design, manuscript writing, and proofreading; LXJ, HW, HZY, JSJ, LKL and CDJ contributed to the statistical analysis and provided methodological consultation; and CYH contributed to the project design and administration. All authors have granted their approval for the manuscript.

Funding

National Natural Science Foundation of China (81830045).

Data availability

The data described in the manuscript will be made available upon request pending application.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Guangzhou Medical University Affiliated Third Hospital (Approval No.: 医伦会审[2024]第074号| Medical Ethics Committee Approval No. (2024) 074; Date of Approval: April 15, 2024). All procedures complied with the Declaration of Helsinki. Informed consent exemption was granted by the ethics committee due to the retrospective nature of the study.

Written informed consent was exempted by the ethics committee for the following reasons: the research involved retrospective analysis of anonymized medical records; all patient identifiers (names, ID numbers, admission dates) were permanently removed; the study protocol posed no more than minimal risk to subjects; all patients had signed broad consent forms during hospitalization permitting use of their anonymized data for research purposes.

Consent for publication

The author confirms: that the work described has not been published before; that it is not under consideration for publication elsewhere; that its publication has been approved by all co-authors; that its publication has been approved by the responsible authorities at the institution where the work carried out. The author agrees to publication in the Journal BMC Pregnancy and Childbirth. The authors warrants that our contribution is original and that we have full power to make this grant.

Competing interests

The authors declare no competing interests.