Sudden death due to acute variceal bleeding: A review of diagnostic challenges and clinical presentation

Abstract

Acute variceal bleeding is one of the most fatal complications in patients with pre-existing portal hypertension secondary to chronic liver disease. Fatality from this condition is potentially preventable but requires urgent medical attention. The diagnosis is often challenging because patients may present with diverse manifestations, ranging from vague epigastric discomfort to haematemesis, melaena, syncope, haemorrhagic shock, or sudden cardiovascular collapse. In some cases, fatal haemorrhage may occur before a definitive diagnosis is established or therapeutic intervention can be initiated, making sudden death the first clinical manifestation. In the emergency setting, upper gastrointestinal endoscopy remains the diagnostic gold standard; however, its utility may be limited in rapidly deteriorating or out-of-hospital patients. This review summarises the current evidence on the pathophysiology of variceal rupture, emergency clinical presentation, diagnostic pitfalls, and factors associated with sudden death. It also highlights the importance of early recognition, prompt risk stratification, and timely intervention to reduce mortality, while emphasising the role of clinicopathological correlation in patients presenting with fatal or unexplained upper gastrointestinal haemorrhage.

Key Words: Acute variceal bleeding; Sudden death; Esophageal varices; Portal hypertension; Postmortem examination; Forensic pathology; Gastrointestinal hemorrhage

Core Tip: Sudden death due to acute variceal bleeding is an uncommon but potentially fatal complication of portal hypertension that may occur without a previously diagnosed liver disease. The diagnosis can be particularly challenging in out-of-hospital deaths, where clinical information and endoscopic confirmation are unavailable. This review highlights the varied clinical presentations, mechanisms of catastrophic haemorrhage, and important diagnostic pitfalls. Careful scene investigation, systematic autopsy examination, histopathology, and clinicopathological correlation are essential for establishing the cause of death and excluding alternative sources of gastrointestinal haemorrhage. Greater awareness among clinicians and forensic pathologists may improve diagnostic accuracy and cause-of-death certification.

INTRODUCTION

Acute variceal bleeding is one of the most serious complications of portal hypertension in patients with cirrhosis. The 6-week mortality following an episode of variceal bleeding is approximately 15%-25% in developed countries. Although advances in endoscopic screening, prophylactic strategies such as non-selective beta-blockers and endoscopic variceal ligation, and improvements in acute haemorrhage management have substantially reduced mortality from historical rates exceeding 50%, significant disparities persist according to healthcare infrastructure. In higher-resource settings, widespread screening and timely primary prophylaxis have reduced the incidence of first variceal haemorrhage, whereas in resource-limited settings, restricted access to diagnostic facilities, endoscopic services, prophylactic treatment, and specialised care may contribute to delayed presentation and poorer outcomes following the first bleeding episode. Therefore, the epidemiology and outcomes of variceal bleeding should be interpreted in the context of differences in healthcare infrastructure and access to preventive and emergency care[1,2].

Contemporary consensus emphasizes rapid recognition, haemodynamic stabilization, vasoactive therapy, antibiotic prophylaxis and timely endoscopy because treatment failure and early death are strongly influenced by the severity of hepatic dysfunction and systemic complications[3-5].

Most published literature focuses on patients who survive long enough to reach hospital and undergo endoscopy, whereas sudden out-of-hospital deaths from massive variceal haemorrhage remain less well characterised. In such cases, variceal bleeding may be the first manifestation of undiagnosed chronic liver disease, and haematemesis may be unwitnessed or absent. At autopsy, the presence of oesophageal varices alone does not establish the source of fatal haemorrhage; rupture must be demonstrated and alternative causes excluded. This mini-review focuses on sudden death due to acute variceal bleeding, with emphasis on clinical presentation, diagnostic pitfalls, and forensic clinicopathological correlation (Table 1).

Table 1 Comparison of acute variceal haemorrhage in in-hospital and out-of-hospital forensic presentations.

Parameter	In-hospital presentation	Forensic presentation
Presentation	Haematemesis, melaena, altered sensorium or haemorrhagic shock	Sudden collapse or unexpected death
Past history	History of cirrhosis, portal hypertension	May be absent
Management	Assessment of airway, circulation and haemodynamic status, followed by resuscitation	Postmortem examination
Investigations	Upper gastrointestinal endoscopy to identify the bleeding source and provide haemostatic treatment	Gross examination for cirrhosis, splenomegaly and other features of portal hypertension, postmortem computed tomography/angiography, histopathology and toxicology

PATHOPHYSIOLOGICAL BASIS OF CATASTROPHIC VARICEAL HAEMORRHAGE

Portal hypertension, most commonly caused by cirrhosis, results from increased intrahepatic resistance and splanchnic vasodilatation, leading to the formation and enlargement of portosystemic collaterals, particularly gastro-oesophageal varices[6-10].

The risk of variceal rupture is determined by the interaction between portal pressure, variceal diameter and wall integrity[11]. A hepatic venous pressure gradient of approximately 10 mmHg defines clinically significant portal hypertension, while variceal bleeding is strongly associated with portal pressures exceeding approximately 12 mmHg[7,12]. However, portal pressure alone does not determine rupture risk. Increasing variceal diameter and thinning of the vascular wall increase wall tension and may predispose varices to rupture during transient haemodynamic stress. Endoscopic features such as large varices and red wale markings identify patients at increased risk of first bleeding[13,14].

Once rupture occurs, rapid blood loss causes reduced venous return, decreased cardiac output and systemic tissue hypoperfusion. Continued haemorrhage may overwhelm compensatory mechanisms, resulting in hypovolaemic shock, metabolic acidosis and multiorgan dysfunction[15].

The clinical consequences are amplified by advanced cirrhosis. Infection, renal dysfunction, hepatic encephalopathy and hepatocellular carcinoma are recognized adverse prognostic factors in patients with acute variceal bleeding. In one prognostic study, Child-Pugh score, creatinine and bacterial infection provided useful early discrimination of mortality risk[16,17]. The importance of renal dysfunction is consistent with the recognized vulnerability of patients with cirrhosis to acute kidney injury during haemorrhage[18].

Cirrhosis also causes complex alterations in haemostasis. Thrombocytopenia, reduced synthesis of coagulation factors, and abnormalities of fibrinolysis coexist with increased levels of von Willebrand factor and other compensatory changes. Thus, cirrhosis should not be regarded simply as a hypocoagulable state; rather, haemostasis is rebalanced and may be destabilised by infection, renal dysfunction, systemic inflammation, and acute blood loss[19-21].

The mechanisms of sudden death following variceal haemorrhage are multifactorial, including haemorrhagic shock, acute lethal arrhythmias induced by massive acute blood loss (severe myocardial hypoxia in underlying alcoholic cardiomyopathy), or rapid airway compromise/asphyxia from blood aspiration before exsanguination occurs.

CLINICAL PRESENTATION AND DIAGNOSTIC CHALLENGES

The classical presentation of acute variceal bleeding is haematemesis and/or melaena, with hypotension, tachycardia, pallor, altered sensorium and shock in severe cases[22]. Brisk upper gastrointestinal haemorrhage may also produce haematochezia[23,24].

The clinical presentation immediately preceding sudden death may be substantially less characteristic. Patients may initially report weakness, dizziness, nausea, abdominal or epigastric discomfort, presyncope or vague malaise. In patients with previously undiagnosed cirrhosis, there may be no known history of liver disease. Subtle clues such as alcohol use, previous jaundice, ascites, splenomegaly, thrombocytopenia or features of chronic liver disease may therefore be missed during an acute presentation. A major diagnostic challenge is that haematemesis is not invariably witnessed. A patient may vomit blood shortly before collapse, lose consciousness and subsequently be found dead. Alternatively, bleeding may be initially attributed to peptic ulcer disease, erosive gastritis, Mallory-Weiss tear, malignancy, Dieulafoy lesion or another cause of upper gastrointestinal haemorrhage[25-27]. The differential becomes particularly important in patients without an established diagnosis of portal hypertension.

Laboratory investigations are useful for assessing severity but are not specific for the source of bleeding. Early haemoglobin concentration may underestimate the degree of acute blood loss because equilibration between intravascular and extravascular compartments takes time. Thrombocytopenia, hypoalbuminaemia, elevated bilirubin, prolonged international normalized ratio (INR) and renal dysfunction may instead provide indirect evidence of underlying hepatic disease and physiological deterioration[23].

Upper gastrointestinal endoscopy remains the definitive investigation in surviving patients because it can identify the bleeding source, demonstrate stigmata of recent haemorrhage and provide immediate haemostatic treatment[28]. Current consensus recommends endoscopy after initial haemodynamic stabilization, generally within 12 hours in suspected acute variceal bleeding[29]. However, this diagnostic pathway is inherently unavailable in patients who die before reaching hospital.

This distinction is important when interpreting the apparent epidemiology of sudden variceal death. Clinical cohorts are inherently enriched for patients who survive long enough to receive medical care, whereas medicolegal autopsy series identify a different population—individuals who die unexpectedly, frequently outside hospital and sometimes without a prior diagnosis of cirrhosis.

SUDDEN DEATH AND FORENSIC PRESENTATION

Sudden fatal oesophageal variceal haemorrhage is uncommon but well documented[30-33]. The most important forensic series remains the study by Tsokos and Türk[34], who reviewed 6038 medicolegal autopsies over five years and identified 45 deaths from fatal oesophageal variceal haemorrhage occurring outside hospital. The incidence within that medicolegal autopsy population was 0.75%; the mean age was 50.6 years, the male-to-female ratio was 1.6:1, and cirrhosis was present in every case. Forty-two individuals had alcoholic cirrhosis, hepatocellular carcinoma was present in three, and rupture was located in the lower third of the oesophagus in 44 cases. Alcohol was detected in femoral blood and/or urine in 30 cases. This study is particularly relevant because it demonstrates that fatal variceal haemorrhage can occur in relatively young individuals and may constitute an unexpected natural cause of death. The authors also highlighted a significant medicolegal problem: Blood staining at the scene and unusual body positions may create an initial suspicion of violent death. The presence of blood on clothing, bedding, bathroom surfaces or surrounding objects therefore requires interpretation in conjunction with the complete autopsy rather than being assumed to indicate trauma[34].

Another important diagnostic issue is aspiration. Massive haematemesis may result in blood entering the upper airway and lungs, particularly in individuals who become unconscious during haemorrhagic shock or have hepatic encephalopathy. Aspiration may contribute to airway obstruction and hypoxaemia and can become part of the terminal mechanism of death. Accordingly, examination of the larynx, trachea and bronchi is important in suspected fatal variceal haemorrhage[35].

The recent literature continues to contain relatively few dedicated autopsy studies of sudden variceal death. This is an important evidence gap because the clinical literature predominantly evaluates patients who reach hospital, while forensic series are generally small and retrospective[33,34,36].

AUTOPSY AND HISTOPATHOLOGICAL DIAGNOSIS

The oesophagus and stomach should be examined carefully after evacuation of their contents. Large quantities of fresh or clotted blood support massive upper gastrointestinal haemorrhage but are not, by themselves, diagnostic of variceal rupture. The distal oesophagus and gastro-oesophageal junction require particular attention because varices may collapse after death and the mucosal rupture can be small[37].

Gross findings may include tortuous, dilated submucosal veins, focal mucosal disruption, adherent clot or an identifiable rupture site (Figure 1). Histologically, oesophageal varices consist of markedly dilated submucosal venous channels. At rupture sites, focal disruption of the vessel wall and overlying mucosa, haemorrhage and thrombotic material may be identified. Serial histological sections may be necessary to identify predominantly subepithelial rupture sites. The liver should be examined for evidence of cirrhosis or another cause of portal hypertension. Gross nodularity should be correlated with histological evidence of fibrosis and architectural distortion. The spleen should be examined for congestive splenomegaly, while collateral vessels and other manifestations of portal hypertension should be documented. The respiratory tract should be inspected for aspirated blood. Blood within the tracheobronchial tree and pulmonary parenchyma may provide important evidence of aspiration during the terminal event. Other organs may demonstrate nonspecific consequences of prolonged shock, including acute tubular injury or hypoxic-ischaemic changes[33,34].

Figure 1 Postmortem photograph showing dilated and tortuous gastro-oesophageal varices. The varices involve the lower oesophagus and extend into the gastric fundus, with multiple small punctate haemorrhagic areas suggestive of recent bleeding.

Postmortem computed tomography (PMCT) and postmortem computed tomography angiography (PMCTA), particularly multi-phase PMCTA, can be useful in diagnosing fatal variceal haemorrhage. PMCT can demonstrate hyperdense material within the gastrointestinal lumen, suggesting the presence of blood. PMCTA can demonstrate focal contrast leakage during the arterial or venous phases, which is more suggestive of true haemorrhage, whereas diffuse leakage or findings limited to the dynamic phase may represent postmortem artefacts. Therefore, PMCT and PMCTA can complement autopsy findings in cases of fatal variceal bleeding, although their findings should be interpreted in conjunction with autopsy findings and other forensic evidence[38].

Toxicological examination is not required in every case but should be considered when circumstances are unclear, particularly where alcohol or sedative drugs may have contributed to impaired consciousness, aspiration or delayed recognition of haemorrhage[33,34].

DIFFERENTIAL DIAGNOSIS AT AUTOPSY

The discovery of oesophageal varices does not automatically establish them as the cause of death. Alternative sources of upper gastrointestinal haemorrhage must be systematically excluded. Peptic ulcer disease, Mallory-Weiss tears, erosive gastritis, gastric malignancy, Dieulafoy lesions, vascular malformations and portal hypertensive gastropathy may produce substantial haemorrhage. Portal hypertensive gastropathy itself may cause acute or chronic gastrointestinal blood loss and should be distinguished from gastric varices[39].

Blood in the stomach may also originate from extra-gastrointestinal sources, including severe epistaxis or oral trauma. Examination of the oral cavity, pharynx and nasal passages is therefore appropriate in unexplained gastrointestinal blood loss[28].

The forensic diagnosis should consequently be based on a combination of gross findings, histopathology, toxicology where indicated, circumstances of death and available clinical information. This approach minimizes the risk of incorrectly attributing death to the incidental presence of varices.

CLINICOPATHOLOGICAL CORRELATION

In a hospitalized patient, clinical records may provide evidence of haematemesis, haemodynamic collapse, cirrhosis, endoscopic bleeding and response to therapy. In an out-of-hospital death, the autopsy and scene examination may provide the only evidence of the disease. However, each link requires supporting evidence. The immediate cause of death should describe the terminal event, while the underlying cause should identify the disease responsible for portal hypertension whenever it can be established. This distinction has both clinical and medicolegal significance. For example, “massive upper gastrointestinal haemorrhage due to ruptured oesophageal varices” describes the immediate pathological event, whereas “portal hypertension due to cirrhosis” identifies the underlying disease. Additional contributing conditions such as alcohol-associated liver disease, hepatocellular carcinoma or aspiration may be recorded where supported.

RISK FACTORS FOR SUDDEN FATAL OUTCOME

Several characteristics identify patients in whom variceal bleeding is more likely to be severe or fatal. These include advanced hepatic dysfunction, large varices, high-risk endoscopic stigmata, active bleeding at endoscopy, infection, renal dysfunction, hepatocellular carcinoma and haemodynamic shock[40-44].

Child-Pugh class remains a clinically useful measure of hepatic reserve, while Model for End-Stage Liver Disease provides an objective estimate of short-term mortality risk based principally on bilirubin, creatinine and INR[45,46]. Prognostic studies in acute variceal bleeding have demonstrated that creatinine and infection add important information beyond hepatic scores alone[18,40,47].

These variables are particularly relevant when attempting to understand sudden death. A patient with severe portal hypertension and advanced hepatic dysfunction may have limited physiological reserve, so that even a short period of uncontrolled bleeding can result in irreversible shock.

CLINICAL IMPLICATIONS

Suffocation is one of the fatal complications of variceal bleeding. This complication occurs when severe massive hematemesis causes blood to be accidentally aspirated into the tracheobronchial tree, which immediately triggers acute airway obstruction that cannot be relieved spontaneously, followed by complete oxygen deprivation and subsequent fatal asphyxia. Timely and prompt establishment of a fully patent airway through standardized endotracheal intubation is therefore critical and indispensable during the initial resuscitation phase to effectively prevent this catastrophic, life-ending outcome. Furthermore, early intubation facilitates subsequent diagnostic and therapeutic procedures, such as upper gastrointestinal endoscopy, by ensuring patient stability and minimizing the risk of further aspiration during sedation. Protective airway management should be prioritized even in patients with altered mental status or ongoing active bleeding, as these factors significantly elevate the likelihood of aspiration events[6].

Transjugular intrahepatic portosystemic shunt (TIPS) is an important therapeutic option for patients with severe portal hypertension and uncontrolled or recurrent variceal haemorrhage despite standard pharmacological and endoscopic treatment. By creating a low-resistance channel between the portal and hepatic veins, TIPS reduces portal pressure and thereby decreases the pressure within oesophageal and gastric varices. It may serve as rescue or salvage therapy in selected patients when conventional haemostatic measures are unsuccessful. However, TIPS is associated with important complications, particularly hepatic encephalopathy, which may result from diversion of portal blood away from hepatic detoxification. Shunt dysfunction, including stenosis or thrombosis, may also occur and can compromise its therapeutic effect. Therefore, careful patient selection and appropriate post-procedural surveillance are essential to balance the benefits of effective portal decompression against the potential complications of portosystemic shunting[44-46].

The possibility of acute variceal haemorrhage should be considered early in any patient with haematemesis or melaena who has known or suspected chronic liver disease. The initial clinical priority is recognition of haemorrhage and circulatory compromise, followed by rapid stabilization and definitive endoscopic evaluation. Contemporary consensus recommends early vasoactive therapy and antibiotic prophylaxis, with endoscopy after stabilization[3,4,23,29].

Emergency clinicians should look for subtle evidence of portal hypertension, while forensic pathologists should consider variceal rupture in unexplained sudden deaths accompanied by substantial upper gastrointestinal blood loss. In a sudden out-of-hospital death, the combination of massive upper gastrointestinal blood, cirrhotic liver morphology, portal hypertensive changes and a demonstrable variceal rupture provides a strong clinicopathological basis for diagnosis. Conversely, varices without demonstrable rupture should prompt continued examination for alternative causes of haemorrhage[48].

CONCLUSION

Sudden death due to acute variceal bleeding represents an under-recognized intersection between hepatology, emergency medicine and forensic pathology. Although most acute variceal bleeding occurs in patients with recognized chronic liver disease, catastrophic haemorrhage may occasionally be the first clinical manifestation of previously unsuspected portal hypertension. The clinical presentation may range from classical haematemesis and melaena to nonspecific symptoms followed by abrupt collapse. The diagnosis in survivors depends principally on timely endoscopy, whereas in out-of-hospital deaths it may depend almost entirely on careful scene investigation, autopsy, histopathology and clinicopathological correlation. Demonstration of a ruptured oesophageal or gastric varix, substantial gastrointestinal blood loss and underlying portal hypertension is essential, while alternative sources of haemorrhage must be excluded. Recognition of sudden fatal variceal haemorrhage should therefore extend beyond the gastroenterology unit. Greater awareness among emergency physicians and forensic pathologists may improve diagnostic accuracy, appropriate cause-of-death certification and understanding of this potentially preventable cause of sudden natural death.