Clinical and biological features of acute hepatic cytolysis in dengue fever: a study from Sourô Sanou University hospital, Bobo-Dioulasso

Highlights

• •In-depth analysis of dengue-related hepatic cytolysis at the Sourô Sanou University Hospital, Burkina Faso.
• •Aspartate aminotransferase 72% and alanine aminotransferase 46% were high; enzymes normalized in most cases within 13 days.
• •Hypertension, fever, dizziness, and anemia were linked to cytolysis as indicators.
• •Self-medication with antimalarials or paracetamol may worsen liver cytolysis.

Abstract

Objective

To describe hepatic cytolysis in patients hospitalized for dengue at the Sourô Sanou University Hospital in Bobo-Dioulasso.

Methods

This was a cross-sectional, descriptive, retrospective, and analytical study conducted over 3 months. All patients hospitalized for dengue during this timeframe were included. Data were collected using the KoBoCollect v1.30.1 application and analyzed with Epi Info software.

Results

A total of 256 patients were included. The mean age was 38.10 ± 16.28 years. Elevated aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels were observed in 72.26% and 46.85% of patients, respectively. A reduced prothrombin time was noted in 4.68% of cases, and unconjugated hyperbilirubinemia in 3.9%. Patients with hypertension had a 2.3-fold higher risk of developing hepatic cytolysis. The median length of hospital stay was five days. The case-fatality rate was 7.03%. Transaminase levels normalized in 95.5% of cases, with a mean recovery time of 13 days.

Conclusions

Hepatic cytolysis is a common complication of severe dengue. Its clinical course is generally favorable. Hypertension appears to be a significant risk factor for the development of this liver injury.

Introduction

Dengue virus is an arbovirus belonging to the Flaviviridae family. It is responsible for a re-emerging infectious disease that is endemic in many parts of the world. Although dengue fever most often follows a mild clinical course, certain forms can become severe or even fatal due to life-threatening multi-organ involvement [1,2]. It is now considered one of the leading causes of viral hemorrhagic fever worldwide. Despite its global burden, dengue remains largely underdiagnosed in several African countries, including Burkina Faso [3].

Since 2013, dengue has been circulating in Burkina Faso in the form of endemo-epidemic outbreaks, the most recent of which occurred in 2023, with a total of 68,346 probable cases reported nationwide [4]. During this outbreak, several severe cases, including those with dengue-associated hepatic cytolysis, were observed at the Sourô Sanou University Hospital (CHUSS). The occurrence of hepatic cytolysis during dengue infection may represent a marker of disease severity and a predictor of poor prognosis, although the exact pathophysiological mechanisms remain poorly understood [5]. To date, numerous studies have been conducted in Ouagadougou on the epidemiological and clinical features of dengue fever [3,5]. A 2022 study reported that 47.31% of patients presented with at least one elevated transaminase level, whereas 26.92% showed signs of hepatocellular insufficiency [6]. In contrast, in Bobo-Dioulasso, scientific data on dengue fever remain limited, and studies specifically addressing dengue-associated hepatic cytolysis are scarce. In this context, we undertook the present study, entitled ‘Hepatic cytolysis complicating dengue during the 2023 outbreak in the department of medicine of the CHUSS.’ The objective of this study was to describe the clinical, biological, and outcome-related characteristics of hepatic cytolysis in patients hospitalized for dengue fever between October 1 and December 31, 2023, at the CHUSS.

Patients and methods

Study design and setting

This was a cross-sectional study with an analytical component, based on the retrospective collection of medical data. It was conducted from October 1 to December 31, 2023, at the CHUSS in Bobo-Dioulasso. CHUSS is the main referral hospital in the Hauts-Bassins region and receives complex cases referred from peripheral health care facilities.

Study population

The study population consisted of all patients hospitalized in the department of medicine at CHUSS with confirmed dengue virus infection.

Inclusion criteria were patients hospitalized for dengue fever with a diagnosis of hepatic cytolysis, and patients who underwent at least two transaminase measurements during hospitalization.

Exclusion criteria included documented chronic liver disease prior to the infectious episode, co-infections with other causes of hepatic cytolysis, and incomplete medical records. Case recruitment was exhaustive and included all patients meeting the eligibility criteria during the study period.

Operational definitions

Confirmed dengue case

A suspected case with a positive rapid diagnostic test (NS1 antigen and/or IgM/IgG antibodies). Acute hepatitis associated with dengue is a confirmed dengue case presenting with severe hepatic cytolysis (transaminase levels > 500 IU/L), and/or hemodynamic instability (hypotension), and at least one warning sign or severity indicator such as thrombocytopenia or hemorrhagic manifestations. Rapid Diagnostic Test (RDT) for dengue is a simple and rapid test detecting either the viral NS1 antigen or dengue-specific IgM/IgG antibodies in blood.

Classification of dengue severity

This included dengue with warning signs, defined as the presence of one or more of the following: severe abdominal pain, persistent vomiting, edema, ascites, pleural effusion, asthenia, lethargy, orthostatic hypotension, hepatomegaly, moderate mucosal bleeding, or hemoconcentration with thrombocytopenia.

Severe dengue was defined by the occurrence of one or more of the following: shock, oligoanuria, neurological impairment, frank jaundice, severe hemorrhage, respiratory distress, or hepatic cytolysis with transaminase levels > 1000 IU/L.

Data sources and types

Data were extracted from hospitalization records. The variables collected included sociodemographic variables (age, sex, medical history), clinical variables (presenting symptoms and severity signs at admission), laboratory variables (complete blood count, transaminases [aspartate aminotransferase(AST) and alanine aminotransferase(ALT)], creatinine, thick blood smear, and dengue RDT detecting NS1 antigen and/or dengue-specific IgM/IgG antibodies), and outcome variables (length of hospital stay, clinical outcome [discharge or death], and time to normalization of transaminase levels).

Statistical analysis

Data were entered and analyzed using Epi Info version 7.2.2.16. Quantitative variables were expressed as means ± standard deviations, while qualitative variables were described using absolute frequencies and percentages. Before performing Student’s t-test for comparison of means, the normality of distributions and the homogeneity of variances were assessed. Comparisons of qualitative variables were performed using the chi-square test when expected frequencies were ≥ 5 and Fisher’s exact test when expected frequencies were < 5. Bivariate analyses were conducted to identify factors associated with hepatic cytolysis. Variables found to be significant in bivariate analysis were included in a multivariate binary logistic regression model. Variable selection for the final model was carried out using bidirectional stepwise selection based on the Akaike Information Criterion (AIC), calculated as: AIC = 2p − 2ln(L), where p is the number of parameters and L is the likelihood function. The model with the lowest AIC was retained. Statistical significance was set at P <0.05

General data

During the study period, 635 patients were hospitalized in the department of medicine, including 391 for dengue fever. Among them, 256 met the inclusion criteria (Figure 1). According to the World Health Organization (WHO) classification, Group C dengue cases were the most frequent, representing 57.03% of the study population (Figure 2). The mean age of patients was 38.1 ± 16.3 years, with a range from 16 to 92 years. The most represented age group was 20–30 years, accounting for 91 cases. Women made up 53.12% of the study population (136 patients), resulting in a sex ratio of 0.9. Among them, 11 women were pregnant, representing 23.4% of the female participants. With regard to education level, 127 patients (49.6%) had received formal education, with secondary education being the most common (62 patients or 24.2%). Most patients (87.1%) resided in urban areas.

Figure 1.

Flow diagram of patients included in the study.

Figure 2.

Distribution of patients according to the WHO classification.

Clinical features

Medical history revealed high blood pressure in 15.6% of patients and sickle cell disease in 5.5%. The use of traditional treatments was common and reported by 179 patients (85.5%). These treatments primarily consisted of herbal remedies (62.2%) and decoctions (37.7%) (Table 1).

Table 1.

Clinical and biological features of acute hepatic cytolysis in dengue fever: a study from Sourô Sanou University hospital, Bobo-Dioulasso

Therapeutic history	Actual	Percentage
Self-medication
Yes	179	85.5
No	77	14.5
Traditional treatment type	77
Phytotherapy	48	62.2
Herbal decoction	29	37.7
Taking paracetamol
Yes	45	17.6
No	211	82.4
Taking non-steroidal anti-inflammatory drug
Yes	4	1.6
No	252	98.4
Antimalarial
Yes	79	30.8
No	177	69.1

The median time to consultation was five days, with a range of 1 to 77 days. Most patients (175 cases) sought medical care within seven days of symptom onset. The most frequently reported symptoms were headache (92.9%), fever (80.5%), and asthenia (70.3%).

Biological data

Upon admission, the median AST level was 640 IU/L (range: 507–8752), and the median ALT level was 802 IU/L (range: 623–2151). A decreased prothrombin level was observed in 73.8% of cases (Table 2).

Table 2.

Distribution of patients according to biological data.

Liver test	Actual	Percentage
Prothrombin levels
Normal	55	21.5
Not completed	12	4.7
Low	189	73.8
Gamma GT
Normal	-	-
High	20	7.8
Not completed	236	92.2
Total bilirubin
Normal	10	3.9
High	16	6.3
Not completed	230	89.8
Unconjugated bilirubin
Normal	16	6.3
High	10	3.9
Not completed	230	89.8
Alkaline Phosphatase
Normal	-	-
High	17	6.6
Not completed	239	93.4

GT : Gamma-Glutamyl Transferase.

Anemia was observed in 92 patients (35.9%), with mild anemia in 37 cases (40.2%) and severe anemia in 33 cases (35.9%). Thrombocytopenia was present in 191 patients (74.6%), classified as mild in 101 cases (52.9%), moderate in 64 cases (33.5%), and severe in 26 cases (13.6%). A thick blood smear was performed in 217 patients (84.8%), of which 65 (29.9%) were positive. Elevated serum creatinine was noted in 80 patients (31.3%). Abdominal ultrasound, performed in selected cases, revealed hepatomegaly in 14 patients (26.4%).

Clinical course

The median length of hospital stay was five days, with a range of 1 to 51 days. Among the 85 patients with complete follow-up, the mean time to normalization of transaminase levels was 13.85 ± 7.1 days, ranging from 1 to 30 days. The case-fatality rate was 7%, corresponding to 18 deaths.

Bivariate analysis

Statistically significant associations were observed between: a history of high blood pressure and hepatic cytolysis (P = 0.02); fever (P = 0.001) and dizziness (P = 0.02) with hepatic cytolysis; and management of bleeding (P = 0.004) and red blood cell transfusion (P = 0.01) with transaminase normalization. In contrast, no significant associations were found between self-medication, mortality, and the occurrence of hepatic cytolysis.

Multivariate analysis

Multivariate logistic regression identified dizziness as an independent factor significantly associated with hepatic cytolysis (Table 3).

Table 3.

Multiple logistic regression of factors associated with hepatic cytolysis.

Associated factors	Coefficient	IC à 95%	P
High blood pressure (yes/no)	0.1	0.3-0.6	0.6
Dizziness (yes/no)	0.1	0.01-0.2	0.02
Fever (yes/no)	0.1	0.3-0,02	0.1
Treatment of bleeding (yes/no)	0.1	0.6-0.4	0.8
Red blood cell concentrate transfusion (yes/no)	0.1	0.6-0.4	0.8

CI, confidence interval.

Discussion

General data

This retrospective study, conducted at the CHUSS during the 2023 dengue outbreak, provides one of the first detailed descriptions of hospitalized dengue cases in Bobo-Dioulasso, with a particular focus on hepatic cytolysis. Although the study was limited by incomplete medical records and a diagnostic approach primarily based on rapid diagnostic tests (in the absence of reverse transcription polymerase chain reaction (RT-PCR) confirmation), it nonetheless highlights a substantial burden of liver involvement among hospitalized patients. Loss to follow-up for patients who left the hospital without formal discharge also represents a methodological limitation. Despite these constraints, the clinical and biological trends observed are consistent with findings reported in previous national and international studies. The mean age of the study population (38 years) aligns with observations from prior investigations in Burkina Faso: 35 years in the study by Guingané et al. [6] and 38 years in the cohort described by Sondo et al. [2]. Similarly, Khalil et al. [7] reported a mean age of 37 years in Pakistan. The predominance of young adults is likely attributable to their increased exposure to Aedes mosquito bites during work-related outdoor activities, which typically coincide with peak vector activity.

Clinical data

High blood pressure was the most common comorbidity in our cohort (15.62%), followed by sickle cell disease (5.46%) and diabetes mellitus (3.12%). These conditions have been widely recognized as factors predisposing to severe dengue. Sondo et al. [2] similarly reported a prevalence of 15% for high blood pressure but a higher proportion of diabetes (11%) in 2017. The presence of such comorbidities underscores the need for close monitoring of seemingly moderate dengue cases, as these patients may deteriorate rapidly. A notable feature of this outbreak was the extremely high rate of self-medication: more than 80% of patients reported taking medications prior to admission. This aligns with the estimate by Ouoba et al. [8] that approximately 85% of the Burkinabè population practices self-medication. Several factors contribute to this trend, including the easy availability of medications, both through pharmacies and informal markets, cultural norms of home-based treatment, and the absence of a curative therapy for dengue. Although free hospital care was provided during the outbreak, this measure did not fully counter entrenched self-treatment practices. Importantly, self-medication may delay hospital presentation, obscure early symptoms, and induce hepatotoxicity, complicating both diagnosis and management. The predominant clinical manifestations, which include headache (92.96%), fever (80.46%), vomiting (65.65%), and myalgia (53.51%), were consistent with data from Guingané et al. [6] and previous classical dengue descriptions by Hober et al. [9]. Hemorrhagic manifestations, including hematemesis (14.84%), gingivorrhagia (13.67%), and epistaxis (9.76%), reflected the multifactorial disruption of hemostasis, involving thrombocytopenia, coagulation abnormalities, platelet dysfunction, vascular fragility, and possible drug-induced hepatic injury. Signs suggestive of liver involvement, which include jaundice, right hypochondrial pain, and pruritus, were less frequent but are known manifestations of dengue-associated hepatocellular injury. Variability between studies may reflect differences in demographic structure; for example, pediatric cohorts, such as the series reported by Smith and Khakpoor [10], often exhibit different clinical patterns.

Paraclinical data

Hepatic cytolysis was prominent, with AST elevation in 72.26% of patients and ALT elevation in 46.85%. Median values at admission were markedly high (640 IU/L for AST and 802 IU/L for ALT), with extreme values reaching more than 10 times the upper limit of normal. More than half of the cohort met WHO criteria for severe dengue, indicating a significant burden of severe forms during this outbreak. Although Guingané et al. [6] reported a lower proportion of severe cytolysis (15.05%), differences in inclusion criteria and outbreak dynamics may explain this discrepancy. Interestingly, while 189 patients exhibited reduced prothrombin time and 10 had indirect hyperbilirubinemia, these abnormalities did not proportionally reflect the extent of cytolysis, suggesting that severe hepatocellular injury can occur with relatively preserved synthetic hepatic function. This decoupling is consistent with observations in dengue-related liver injury described in the literature. Thrombocytopenia was the most frequent hematological anomaly (74.60%), followed by anemia (35.93%). These rates are comparable to those reported by Malhi et al. [11] and Khalil et al. [7]. The high anemia rate may be attributable to occult or overt bleeding, compounded by impaired coagulation due to hepatic dysfunction. Renal impairment was also frequent (31.25%), consistent with earlier findings by Coulibaly et al. [12] and Malhi et al. [11]. The pathophysiology of dengue-associated acute kidney injury is complex, involving dehydration, renal hypoperfusion, capillary leak, direct viral nephrotoxicity, and drug-induced injury linked to self-medication [13,14]. These results highlight the need for routine renal monitoring and, ideally, collaborative studies with nephrology departments to better characterize dengue-associated renal damage.

Evolutionary aspects

Outcome and prognosis

The median length of hospitalization (five days) was similar to that reported by Malhi et al. [11]. However, adherence to post-discharge follow-up was poor, with more than half of the patients missing scheduled consultations. This may reflect both rapid clinical improvement and socioeconomic constraints, notably transportation costs. These findings underscore the importance of reinforcing patient education regarding the need for monitoring potential late complications. The case-fatality rate in our study (7.03%) was higher than that reported by Guingané et al. [6] (4.30%). This difference may reflect a higher proportion of secondary dengue infections, as outbreaks in 2022 and 2023 could have involved different circulating serotypes. The phenomenon of antibody-dependent enhancement, which increases the risk of severe disease during heterologous secondary infection, has been well described in the literature [15]. The larger sample size and the high prevalence of severe forms may also explain this elevated mortality. Normalization of transaminase levels occurred in 95.50% of patients within a mean of 13.85 ± 7.09 days, consistent with the transient nature of dengue-associated hepatic injury reported in other studies. Multivariate analysis identified hypertension, fever, and vertigo as significant factors associated with hepatic cytolysis. Hypertension may contribute through chronic endothelial dysfunction and microvascular disturbances that predispose the liver to inflammatory or ischemic stress. Fever likely reflects a higher viral burden or amplified immune response, whereas vertigo may indicate systemic deterioration, hypovolemia, or metabolic imbalance. Interestingly, red blood cell transfusion was positively associated with faster transaminase normalization, suggesting that correction of anemia and improved oxygen delivery may promote hepatic recovery. These results reinforce the importance of comprehensive supportive management, particularly prevention of bleeding and optimization of organ function

Conclusion

Dengue is a viral infection that primarily affects the liver, kidneys, and muscles. Hepatic cytolysis is a frequent but generally mild complication, with liver enzymes typically normalizing over time. Comorbidities, particularly hypertension, increase the risk of liver injury, underscoring the need for close monitoring. Prospective studies are needed to better identify prognostic factors and optimize patient management.

Declaration of competing interest

The authors have no competing interests to declare.