Abstract
Yellow Fever (YF) is a vector-borne flavivirus infection that can rapidly evolve into fulminant hepatic failure (FHF), a condition associated with high mortality and severe neurological complications, including intracranial hemorrhage (ICH). Early identification of such complications remains a clinical challenge, especially in critically ill patients where the neurological examination is often limited. We present the case of a 60-year-old male with laboratory-confirmed YF by reverse transcription polymerase chain reaction, who developed ICH during the course of his illness. The implementation of a multimodal neuromonitoring approach—including near-infrared spectroscopy, transcranial Doppler, and jugular venous oxygen saturation—enabled the early detection of cerebral hemodynamic and perfusion changes suggestive of intracranial pathology. Despite the timely recognition of ICH, the patient's rapidly deteriorating coagulopathy, driven by FHF, precluded successful intervention, culminating in brain death. This case highlights the value of integrating advanced neuromonitoring modalities in the intensive care setting to enable prompt identification of life-threatening neurological complications. Although not sufficient to alter the fatal trajectory in this instance, early neuromonitoring may offer a critical window for therapeutic decision-making in patients with severe YF. Future research is needed to define standardized neuromonitoring protocols and to evaluate their impact on outcomes in this high-risk population.
Keywords
Introduction
Yellow Fever (YF) is a mosquito-borne viral disease, primarily transmitted by Aedes species, that has re-emerged in several tropical regions of Africa and America despite the existence of an effective vaccine. This illness is caused by the Yellow Fever Virus, a flavivirus transmitted to humans through the bite of infected mosquitoes.1,2 Although most infections remain asymptomatic, approximately 10–15% of infected individuals develop severe forms of the disease, which include fulminant hepatic failure (FHF) and intracranial hemorrhage (ICH).2,3
The progression of YF to severe forms can be remarkably rapid, with clinical deterioration observed within hours. This accelerated progression is particularly pronounced in patients who develop neurological complications. It has been documented that up to 80% of patients with severe forms of YF exhibit neurological alterations, and among these, 30% develop cerebral edema and intracranial hypertension. 4 Rapid deterioration of organ function presents a serious challenge to patient management, as multi-organ failure often reduces the patient's capacity to benefit from medical treatment.2,5,6
Multimodal neurological monitoring has emerged as a pivotal tool for the early identification of complications, enabling real-time assessment of cerebral perfusion, oxygenation, and cerebral blood flow. 7 Techniques such as near-infrared spectroscopy (NIRS), transcranial Doppler (TCD), jugular venous oxygen saturation (JvO2) measurement are essential for detecting hemodynamic alterations before they clinically manifest as overt symptoms of intracranial hypertension or ICHs. 7
The rapid progression of the disease continues to pose a significant challenge, as hemodynamic and neurological alterations can remain undetected until they become irreversible. We report the case of a patient with YF in whom the application of a multimodal neuromonitoring methodology allowed for the early suspicion and subsequent confirmation of ICH, and furthermore, guided the management of cerebral hemodynamics.
Case report
The patient was a male in his early 60 s residing in a rural area of Ataco municipality, Tolima, Colombia. He was admitted in March 2025 to the Clínica Keralty, Ibagué, Tolima, Colombia. He had no relevant medical history and had not received prior vaccination for YF. He initially presented with a 5-day history of unquantified fever, general malaise, abdominal pain, nausea, and emesis. The diagnosis of YF was corroborated by reverse transcription polymerase chain reaction for the YF virus.
Multimodal neuromonitoring included serial TCD, continuous NIRS, and intermittent JvO₂ assessments. TCD was performed through bilateral transtemporal acoustic windows using a GE HealthCare Vivid™ S70N ultrasound system with a 2-MHz probe, insonating the middle cerebral arteries (MCAs) at a depth of 50 mm. Mean flow velocity and pulsatility index (PI) were measured at least three times per day, and more frequently when clinical deterioration was suspected. NIRS monitoring was performed continuously using bilateral ForeSight™ tissue oximetry sensors (Edwards Lifesciences, USA) placed on the frontal regions. Proper placement was verified, and the potential effect of cervical flexion was assessed to exclude artifacts related to carotid compression. Representative values were selected to reflect clinically significant trends. JvO₂ was obtained intermittently via a 5 Fr jugular bulb catheter (Arrow®, Teleflex Medical, USA), and simultaneous arterial and jugular venous blood gases were collected to ensure accuracy in cerebral oxygen extraction calculations. Of note, the left-sided drop in regional cerebral oxygen saturation (rSO₂) measured by NIRS showed temporal correlation with an increase in the PI in the left MCA, supporting the presence of evolving intracranial hypertension
The patient exhibited a gradual worsening of his general condition, characterized by persistent fever, progressive jaundice, and notable hepatomegaly. Laboratory assessments revealed a severe coagulopathy and acute hepatic dysfunction, evidenced by a significant increase in transaminases, hypoproteinemia, and related coagulopathy (Table 1). Initial fluid resuscitation was performed using volume-sparing strategies, including 3% hypertonic saline administered in 2 cc/kg intravenous boluses every 6 h, and 20% human albumin at a dose of 20 grams IV every 8 h. Subsequently, he progressed to West Haven Grade IV hepatic encephalopathy.
Clinical and laboratory evolution of the patient during hospitalization.
Notes: TCD: transcranial Doppler; TAMAX: time averaged maximum velocity of middle cerebral artery; PI: pulsatility index of middle cerebral artery; NIRS: near-infrared spectroscopy; JvO2: jugular venous oxygen saturation.
The patient exhibited progression of multi-organ involvement, with hepatic, hematological, and neurological damage being particularly pronounced. Given the presence of hepatic encephalopathy, the patient required invasive mechanical ventilation and sedoanalgesia. Sedation and analgesia were achieved with continuous intravenous infusions of propofol (2–4 mg/kg/h) and remifentanil (0.1–0.2 mcg/kg/min), targeting deep sedation. These agents were maintained from day 4 of hospitalization until the time of death. Initially, a non-contrast computed tomography (CT) scan of the brain was performed, which revealed no pathological findings, and no changes in pupillary diameters were observed in the clinical exam. Subsequently, a multimodal neuromonitoring strategy was initiated. Twelve hours after the initial non-contrast brain CT scan, TCD revealed an increased PI of the left MCA, suggesting elevated intracranial pressure (ICP) in the left hemisphere. At this point, no alterations in cerebral oxygenation by JvO2 or asymmetry/decrease in cerebral oxygenation values NIRS were observed. Rigorous medical management of ICH was initiated based on the TCD findings. As the pulsatility indices on TCD increased, suggesting elevated ICP, 7% hypertonic saline was administered in intravenous boluses of 2 cc/kg. Two boluses were given before the patient's clinical deterioration, after which a non-contrast brain CT scan was ordered (Figure 1).
Serial non-contrast computed tomography (CT) scans of the brain in a Yellow Fever patient. Notes: Left image demonstrates a neuroimaging pattern consistent with intact skull bones, cerebral parenchyma exhibiting normal tissue structure and density, ventricles within normal limits for size and shape, and absence of midline shift. Right image, obtained approximately 12 h after, due to severe neurological deterioration on examination, reveals left frontal and occipital involvement characterized by intraparenchymal hemorrhage and vasogenic edema with mass effect, along with marked midline deviation compatible with cerebral herniation.
The non-contrast brain CT scan revealed a left occipital-parietal intraparenchymal hematoma, approximately 30 cc in volume, accompanied by an extensive left subdural hematoma. These findings were associated with a midline shift and evidence of subfalcine herniation (Figure 2).
Neuromonitoring with transcranial Doppler (TCD) and intracranial hemorrhage progression in a Yellow Fever patient. Notes: Image A displays the middle cerebral artery (MCA) insonated via the temporal window, showing a normal flow pattern; however, a slightly elevated pulsatility index (PI) is present, indicating early changes in intracranial pressure (ICP). Image B demonstrates a worsening Doppler pattern, characterized by a decrease in both systolic and diastolic velocities and a progressive increase in PI, confirming the diagnosis of intracranial hypertension. Image C illustrates a pattern with systolic predominance and minimal diastolic flow, along with a significantly elevated PI, consistent with impending cerebral circulatory arrest. Finally, Image D exhibits further lesion progression with the presence of a biphasic or reverberating flow pattern and systolic spikes of less than 200 milliseconds duration, which correlated clinically with the patient's absence of brainstem reflexes. In all instances, measurements were performed at a depth of 50 mm.
Despite the implementation of rigorous ICH management measures, including deep sedation, hypertonic solutions, protective ventilatory parameters, hemodynamic augmentation with an inodilator, and neuroprotective ventilatory strategies, TCD documented a progressive increase in the bilateral pulsatility indices of the MCA. This was associated with an asymmetric decline in NIRS values for the left cerebral hemisphere, without significant changes in JvO2 values at that point. Hemodynamic support consisted of low-dose norepinephrine (up to 0.05 mcg/kg/min) and continuous infusion of milrinone (0.375 mcg/kg/min) to augment cerebral perfusion pressure.
In response to the progressive increase in pulsatility indices on TCD, consistent with intracranial hypertension, and the concomitant paradoxical elevation in SjvO₂ suggesting impaired cerebral oxygen extraction due to neuronal injury, a comprehensive intracranial hypertension management protocol was reinforced. This included deep sedation, administration of hypertonic saline solutions, optimization of protective ventilatory parameters, vasoactive support adjustment, and hemodynamic augmentation strategies. These interventions, already detailed with specific dosages in the pharmacological management section, were temporally aligned with the neuromonitoring changes to mitigate secondary cerebral injury
This paradoxical increase in SjvO₂ was interpreted as “Class B hypoxia,” a condition characterized by impaired cerebral oxygen extraction due to severe mitochondrial dysfunction and a marked reduction in the cerebral metabolic rate of oxygen (CMRO₂), rather than improved cerebral oxygen delivery
Given these findings, neurosurgical consultation for surgical drainage of the hematomas was obtained. However, the patient subsequently exhibited progressive bilateral pupillary dilation, a marked increase in JvO2 values suggestive of Class B cerebral hypoxia due to mitochondrial neurological damage, and a concurrent decline in bilateral NIRS values. Furthermore, evaluation of cerebral vasculature revealed reverberant flow, indicating progression to cerebral circulatory arrest. Consequently, neurosurgical management was deemed futile, and the diagnostic protocol for brain death was initiated. After a 6-day intensive care unit stay, the patient was declared deceased. Refer to Table 2 for the timeline of the clinical case.
Clinical case timeline.
Notes: YF: Yellow Fever; RT-PCR: reverse transcription polymerase chain reaction; ICU: intensive care unit; TCD: transcranial Doppler; MCA: middle cerebral artery; PI: pulsatility index; ICP: intracranial pressure; NIRS: near-infrared spectroscopy; JvO2: jugular venous oxygen saturation; CT: computed tomography.
The reporting of this study conforms to the CARE guidelines. 8 Written informed consent for medical treatment and clinical management was obtained from the patient's legal representative, who also provided written informed consent for the publication of this case report and any accompanying images. Finally, we have de-identified all patient details to ensure confidentiality.
Discussion
The incidence of neurological complications in YF cases complicated by FHF has been documented in the literature. For instance, a study by Ho et al. during the YF outbreak in Brazil reported an increase in ICP through optic nerve sheath diameter measurement by ultrasound in only 2% of cases. 4 However, no prior reports have documented the use of real-time multimodal neuromonitoring—combining TCD, NIRS, and jugular oximetry—in YF patients. This low percentage likely reflects the inherent limitations in detecting subtle, yet significant, clinical changes in critically ill YF patients. In contrast, the implementation of a multimodal neurological monitoring protocol in the present case enabled the early identification of cerebral hemodynamic alterations secondary to ICH. Nevertheless, the rapid progression of the underlying disease ultimately limited the effectiveness of therapeutic interventions.
The utility of non-invasive monitoring techniques such as TCD for identifying hemodynamic alterations in patients with FHF of various etiologies is well established. Abdo et al. conducted TCD monitoring in FHF patients and observed a trend toward lower mean velocity in the right MCA (56.1 vs. 58.1 cm/s) and a higher PI (1.71 vs. 1.41) among those who died compared with patients who successfully underwent transplantation. Although these differences did not reach statistical significance (p = 0.800 and p = 0.787, respectively), they suggest a potential association with elevated ICP and should be interpreted with caution. 9 In the present case, TCD proved crucial for detecting an increased PI in the left MCA, indicating elevated ICP within the left hemisphere, which was subsequently confirmed by non-contrast brain CT.
JvO2 monitoring has been implemented as a valuable tool for estimating changes in cerebral perfusion. A case report by Kazuo et al. documented a correlation between declines in JvO2 and reductions in mean arterial pressure (correlation coefficient: 0.894; p < 0.05) in a patient with FHF. 10 In the present case, although JvO2 values initially did not show significant alterations, as the ICP estimated by TCD progressively increased and led to a reduction in cerebral blood flow, JvO2 values paradoxically increased. This elevation is not indicative of improved cerebral oxygenation but rather represents a profound failure of cerebral oxygen extraction at the cellular level. As ICP reached critical levels, causing widespread neuronal injury, the CMRO2 plummeted due to severe mitochondrial dysfunction.10,11
In patients with FHF, NIRS has demonstrated utility in detecting changes in cerebral oxygenation prior to the manifestation of overt clinical symptoms. A case report by Henning et al. documented that variations in mean arterial pressure, secondary to vasoactive agent titration in FHF patients monitored with NIRS, correlated with changes in total cerebral hemoglobin oxygen concentration of 2 (0.3–14.8) mmol/l and p < 0.05. 11 In the present case, a progressive decline in cerebral oxygenation was associated with an increase in the MCA pulsatility indices, further supporting the suspicion of cerebral tissue hypoperfusion.
In the presented case, the rapid progression of severe YF underscores the critical need for improved treatment protocols and earlier intervention to avert progression to more advanced stages of the disease. Furthermore, the existing literature suggests that the combined application of TCD, NIRS, and JvO2, while promising, necessitates further validation through prospective studies, particularly within the context of YF, to definitively confirm its utility in routine clinical practice. This case highlights the potential role of multimodal neuromonitoring for the early detection of intracranial complications in YF, but its findings are limited by the single-patient design and the inability to modify the fatal outcome due to severe coagulopathy. Further prospective research is needed to validate these observations and to establish standardized monitoring protocols in this setting.
Conclusion
This case illustrates the potential value of multimodal neuromonitoring for the early detection of intracranial complications in severe YF. Although the outcome was fatal, the observations emphasize the importance of early recognition and may help guide clinical decision-making in similar cases.
Footnotes
Acknowledgements
We thank the intensive care team at Clínica Keralty, Ibagué, Colombia for their clinical support. We also acknowledge the use of AI-assisted language editing (ChatGPT, OpenAI) for improving the clarity and grammar of the manuscript.
Ethics approval and consent to participate
The authors declare that no experiments involving human subjects or animals were conducted in the present study. The use of clinical history information was granted with the express consent of the patient's family, in accordance with Colombian regulations. Furthermore, the information contained within this manuscript was previously reviewed and approved by the Ethics Committee of Fundación Universitaria Sanitas.
Consent for publication
Written informed consent was obtained from the patient's legal guardians for the publication of this case report and its accompanying images.
Author contributions
JAMO contributed to conceptualization, formal analysis, methodology, project administration, resources, supervision, visualization, writing—original draft, and writing—review & editing. AJFD, JFCG, and GAGV contributed to formal analysis, methodology, resources, supervision, visualization, writing—original draft, and writing—review & editing. HRA and AQA contributed to resources, visualization, writing—original draft, and writing—review & editing.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data and materials availability
Not applicable.