Severe red kidney beans toxicity in an 8-year-old girl: a rare case of hypovolemic shock and prerenal acute kidney injury

Abstract

Background

Red kidney beans (Phaseolus vulgaris) contain high levels of the lectin phytohaemagglutinin (PHA). Ingestion of raw or insufficiently cooked beans can cause acute toxicity, with symptoms typically arising 1–3 h after consumption. While most cases produce transient nausea, vomiting, and diarrhea, severe outcomes including hypovolemic shock and acute kidney injury (AKI) are rare.

Case presentation

An 8-year-old girl developed loss of consciousness, profound hypotension, and bradycardia after ingesting a home-cooked red kidney bean dish. She required cardiopulmonary resuscitation (CPR) and intensive fluid resuscitation. Laboratory findings indicated dehydration and prerenal azotemia (BUN/creatinine ratio ≈ 33:1). A follow-up serum creatinine 24 h after admission was 0.5 mg/dL. After aggressive supportive care, she regained consciousness, and renal function recovered. Treatment was entirely supportive, consisting of airway management, intravenous crystalloids (20 mL/kg boluses), and gastrointestinal protection. No specific antidote exists for PHA. The patient improved over the course of hours, with normalization of vital signs and renal parameters.

Conclusion

This case highlights an unusual presentation of red kidney bean lectin poisoning manifesting as shock and prerenal AKI. Clinicians should consider undercooked bean ingestion in children with sudden gastroenteritis and shock. Prompt fluid resuscitation is effective, and complete recovery is expected with supportive care.

Background

Red kidney beans (Phaseolus vulgaris) contain high levels of the lectin phytohaemagglutinin (PHA), which binds intestinal epithelial cells and causes profound gastrointestinal toxicity [1]. PHA is heat-labile but persists in undercooked beans. Proper cooking inactivates most lectin: raw beans can contain 20,000–70,000 hemagglutination units (HAU)/g, whereas cooked beans are reduced to < 400 HAU/g [2–4]. Even the ingestion of a few raw or improperly cooked beans (4–5 beans) can trigger severe symptoms [1]. The toxic lectin disrupts the gut mucosa and releases enterotoxins or hormones that induce rapid-onset vomiting and diarrhea [1]. Symptoms typically begin 1–3 h after ingestion and often include abdominal pain, nausea, forceful vomiting, and watery diarrhea [1]. Recovery is usually spontaneous within 3–10 h [2, 4].

Outbreaks of kidney bean poisoning have been reported globally. In a large series among French military personnel, 200 cases of food poisoning occurred from a single meal of undercooked chili con carne [1]. Fifty UK incidents (1976–1989) and a Czech series with hospitalization of children have been documented [5, 6]. In China (2004–2013), thousands of cases and dozens of outbreaks were linked to fresh kidney beans [7]. These poisonings illustrate that while red bean lectin toxicity is well-established, severe complications like shock or acute kidney injury (AKI) are unusual and under recognized [1, 3].

Diagnosis relies on history and exclusion of other causes; stool cultures and viral panels are typically negative. There is no specific antidote for lectin poisoning. Management is entirely supportive, focusing on aggressive fluid resuscitation to reverse hypovolemia and correct electrolyte imbalances [8]. Prophylactic antiemetics or histamine blockers may reduce gastric irritation, though evidence is anecdotal. Intravenous isotonic fluids are indicated when dehydration is moderate to severe [8]. In children, prompt restoration of intravascular volume is critical, as hypovolemic shock can cause end-organ injury, including pre-renal AKI [8]. Clinicians should maintain suspicion for lectin poisoning in any child with an acute onset of vomiting and diarrhea after bean ingestion, to ensure timely rehydration and monitoring for complications.

Case presentation

An 8-year-old previously healthy girl presented with loss of consciousness, persistent vomiting, diarrhea, and respiratory grunting 6 h after eating a home-cooked red kidney bean dish. The family reported the beans were “thoroughly cooked,” but soon after the meal, the patient vomited several times and developed watery diarrhea. Within hours, she became drowsy and unresponsive. Her father, who ate the same meal, had mild nausea and vomiting but recovered without medical care; no other household members were ill.

On arrival at the emergency department, the child was unconscious (Glasgow Coma Scale 8) and hemodynamically unstable. She was found to have a profound bradycardia (20 beats/min), which rapidly deteriorated to pulseless electrical activity, necessitating CPR. Her blood pressure was unrecordable, and respirations 12/min with gasping effort. Oxygen saturation was 82% on room air. Examination revealed clear lungs and normal heart sounds, with no murmurs. The abdomen was soft without distension or organomegaly. There was no rash or trauma. Pupils were mid-sized and reactive to light. Neurologic exam showed generalized hypotonia, symmetric deep tendon reflexes (2+), and downgoing plantar responses. On presentation, the patient was clinically oliguric.

Immediate resuscitation was initiated. Cardiopulmonary resuscitation was performed with intubation; spontaneous circulation returned after brief CPR. Intravenous normal saline boluses (20 mL/kg) were given rapidly. Hemodynamic stability was achieved solely with crystalloid fluid resuscitation; no vasoactive or inotropic support was required. Due to the profound, life-threatening nature of her presentation with circulatory collapse, and a history strongly suggestive of a specific, potentially persistent gastrointestinal toxin (phytohaemagglutinin), the attending physician made the decision to perform gastric lavage following stabilization of her airway and hemodynamics. This intervention was undertaken as a precautionary measure to remove any potential residual unabsorbed toxin from the stomach, despite the 6-hour delay, given the extreme clinical severity and the theoretical risk of ongoing absorption from the gut lumen. The lavage yielded a small amount of gastric contents. The patient was kept NPO (nil per os). Cimetidine was administered intravenously for gastric protection. Over the next two hours, the patient’s blood pressure and consciousness gradually improved. She regained full consciousness and began drinking fluids with good tolerance. Following aggressive fluid resuscitation, her urine output normalized within the first 12 h of admission.

Initial laboratory studies showed evidence of dehydration and prerenal azotemia. Complete blood count was unremarkable (White blood cells 8,500/mm^3, neutrophils 58%, hemoglobin 14 g/dL, platelets 332 × 10^3/mm^3). Serum creatinine was 1.2 mg/dL and blood urea nitrogen 40 mg/dL (BUN/Cr ≈ 33:1). The elevated serum creatinine (1.2 mg/dL), representing a significant acute increase from a presumed normal baseline for her age, was consistent with KDIGO Stage 1 acute kidney injury in the setting of profound volume depletion (prerenal azotemia). Liver enzymes and electrolytes were within normal limits (sodium 138 mEq/L, potassium 4.3 mEq/L, ionized calcium 1.8 mmol/L). Random glucose was 108 mg/dL. Urinalysis revealed a high specific gravity (1.030) with no proteinuria, hematuria, or casts, consistent with prerenal physiology. Arterial blood gas (ABG) and serum lactate levels, while clinically indicated, are not routinely available at our facility and could not be obtained during the acute resuscitation.

To exclude other etiologies, stool studies for common bacterial and viral pathogens were negative. Blood cultures showed no growth. There was no history of exposure to other nephrotoxins, herbal products, or medications. After fluid resuscitation, urine output normalized and azotemia resolved. A follow-up serum creatinine 24 h after admission was 0.5 mg/dL, with a BUN of 12 mg/dL, confirming rapid resolution of the prerenal azotemia[Table 1]. Renal ultrasound was not performed given the rapid clinical and biochemical improvement with volume resuscitation. The patient’s symptoms (vomiting and diarrhea) also subsided with conservative management. No further complications occurred. By discharge on hospital day 2, vital signs and renal function were normal, and the patient was discharged home with complete recovery.

Table 1.

Key laboratory parameters at presentation and after 24 h

Parameter	At presentation	After 24 h	Reference range
Renal function
Serum creatinine	1.2 mg/dL	0.5 mg/dL	0.3-0.7 mg/dL
Blood urea nitrogen	40 mg/dL	12 mg/dL	5-18 mg/dL
Electrolytes
Sodium	138 mEq/L	140 mEq/L	135–145 mEq/L
Potassium	4.3 mEq/L	4.1 mEq/L	3.5-5 mEq/L
Urinalysis
Specific Gravity	1.030	Not repeated	1.005–1.030
Protein / Blood / Casts	Negative	Not repeated

Discussion

Severe hypovolemic shock due to kidney bean lectin is rare. Most published cases of red bean poisoning describe self-limited gastroenteritis [1]. In a French military outbreak, symptoms were intense but self-resolving: 69.6% had diarrhea and 65.2% abdominal pain, with spontaneous recovery by 10 h [1]. The incidence of shock or AKI in such cases is not reported, implying they are exceptional. Our patient’s presentation with bradycardia (pulse 20), unmeasurable blood pressure, and decreased perfusion indicates profound volume loss. The profound bradycardia observed is most consistent with a terminal pre-arrest rhythm in severe shock. However, a direct vagotonic effect of PHA, potentially mediated through intense nausea and vomiting or direct interaction with gastrointestinal neural plexuses, cannot be entirely ruled out as a contributory mechanism, though this remains speculative. The family history supported dose dependence: the father ate the same beans with milder illness. Severity of toxicity correlates with lectin dose [1], explaining why this child developed shock.

Diagnostic challenges are significant. Clinically, bean lectin poisoning mimics acute food poisoning or viral gastroenteritis (onset with profuse vomiting/diarrhea). Key distinguishing features are rapid onset after known bean ingestion and lack of fever or inflammatory markers. In our case, the six-hour delay was longer than the typical 1–3 h [1], but still compatible. Further history revealed the beans were soaked for only a few hours and boiled within a stew for approximately 15–20 min, which is insufficient to fully degrade PHA compared to guidelines recommending > 12 h of soaking and vigorous boiling for at least 10 min in water. Laboratory tests for PHA do not exist in routine labs, so the diagnosis was presumptive by history. The combination of the temporal link to bean ingestion, the milder illness in a co-consumer, negative infectious workup, and rapid recovery with fluids strongly supports the diagnosis of phytohaemagglutinin toxicity. Investigators note that foodborne outbreaks due to PHA are often “poorly documented” because PHA is not routinely tested [1]. In practice, a careful history of undercooked legume ingestion is crucial. In any child with acute profuse vomiting/diarrhea and rapid dehydration, this cause should be considered, especially if multiple cases occur or if family members report similar symptoms.

The cornerstone of treatment is aggressive intravascular volume repletion [8]. Our management followed pediatric guidelines for hypovolemic shock: a 20 mL/kg bolus of 0.9% saline, repeated as needed [9, 10]. This is consistent with recommendations that isotonic crystalloid is first-line in pediatric dehydration and shock [8]. Early fluid resuscitation likely reversed her prerenal azotemia; her BUN/creatinine ratio (~ 33:1) was consistent with volume-depletion AKI. We monitored electrolytes and kidney function; no dialysis was needed. Supportive measures (antiemetics, histamine blockade) were applied, though their efficacy in lectin toxicity is unproven. Although anaphylaxis to legumes is described, it was considered unlikely in this case due to the absence of mucocutaneous or respiratory signs and the dominant presentation of massive gastrointestinal fluid loss. The rapid response to fluid resuscitation alone further supports hypovolemia as the primary driver of shock. The rapid recovery of consciousness and normalization of renal parameters within hours demonstrate that intensive supportive care is effective, as noted in other outbreaks.

In published pediatric gastroenteritis cohorts, AKI is common (≈ 25% incidence) and related to dehydration [11]. This underscores that any severe diarrheal illness can cause AKI; lectin poisoning is simply a specific trigger. In contrast to intrinsic renal toxins, prerenal AKI from hypoperfusion usually resolves with fluids [9]. Our case had no permanent sequelae, aligning with literature that most patients recover fully once volume is restored. A specific limitation in the acute management was the unavailability of point-of-care lactate and ABG testing, which would have provided valuable objective measures of shock severity and acid-base status.

Comparing our case to reports, red bean poisoning severe enough to cause shock is exceedingly unusual. We found no other pediatric case of cardiac arrest from kidney bean ingestion. This highlights the importance of awareness: even home-cooked dishes must be prepared with adequate soaking and boiling to deactivate PHA [12–14]. Public health guidelines recommend soaking dried beans for at least 12 h, discarding the water, and then boiling them vigorously in fresh water for a minimum of 10 min to ensure lectin inactivation [13]. Our patient’s family claimed thorough cooking, but they may have under-soaked or not boiled long enough, as stiff beans (despite cooking) can retain lectin [3]. Clinicians should counsel caregivers on safe bean preparation to avoid such poisonings.

Conclusion

This life-threatening case emphasizes that ingestion of undercooked red kidney beans can precipitate severe hypovolemic shock and prerenal AKI in children. Early recognition, based on a careful dietary history, and prompt, aggressive fluid resuscitation are paramount for full recovery. Public education on proper bean preparation, including soaking for over 12 h and boiling vigorously for at least 10 min, is essential to prevent such poisoning.

Author contributions

Asteway M. Haile: Writing: original draft, Resources, Data curation, Conceptualization. Biruk T. Mengistie: Writing: original draft, review & editing. Alazar (A) Teshager: Writing –original draft, Resources. Telila K. Belisa: Writing – review & editing, Elezer (B) Zewde: Writing – review & editing, Software, Data curation. Andebet S. Deress: Writing – review & editing, Software.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The data underlying the results presented in this work are available within the manuscript.

Declarations

Ethical approval

IRB review and approval were waived for this case report.

Consent to participate

Not applicable.

Consent for publication

Written informed consent was obtained from the patient for publication of the case details and accompanying images.

Competing interests

The authors declare no competing interests.