Altered Mental Status Due to Hyperammonemia Syndrome and PRES in a 31-Year-old Bone Marrow Transplant Recipient

Abstract

Hyperammonemia syndrome and posterior reversible encephalopathy syndrome (PRES) are potentially devastating diagnoses in transplant patients. Their underlying etiologies and pathophysiologies remain incompletely understood, and while they are separately well-documented complications in posttransplant patients, they have not been described concurrently. Here we present a case of both hyperammonemia syndrome and PRES causing rapid mental status decline in a 31-year-old bone marrow transplant recipient. The patient had extensive testing to rule out other diagnoses and made a full recovery after correction of her hyperammonemia. Further research is needed to elucidate the underlying mechanisms of these disease processes; however, clinicians should keep both diagnoses in mind when treating transplant patients with acute neurologic changes.

History

A 31-year-old woman with a past medical history of sickle cell beta thalassemia postsibling matched bone marrow transplant (BMT) 1 year prior, presented to the emergency department (ED) with nausea, vomiting, fevers, headaches, and abdominal discomfort.

She was diagnosed with sickle cell beta thalassemia at age 1 and suffered from frequent complications related to her disease including need for blood transfusions, pain crises, pneumonia, acute chest syndrome, cytomegalovirus viremia, avascular necrosis, and transfusion-related iron overload. At age 30, 1 year prior, she received an allogenic BMT from her sister who had a medical history of sickle cell trait and reversible cerebral vasoconstriction syndrome (RCVS). Following BMT, she was taking tacrolimus for immunosuppression; however, this was discontinued 6 months before ED presentation. The patient had no history of alcohol use disorder, liver cirrhosis, or urea cycle disorders. She was not receiving parenteral nutrition. She was not taking steroids at the time of admission.

The patient developed mild encephalopathy during an admission 2 months prior to this presentation. Prior to BMT she had never suffered significant alterations in mental status or agitation during her frequent lifetime hospitalizations.

Physical Exam Findings

In the ED, the patient had a temperature of 37.4°C, heart rate of 151, blood pressure of 93/54, respiratory rate of 18, and oxygen saturation of 100% on room air. She was alert and fully oriented without neurologic deficits, her abdomen was soft and non-tender, and she appeared in no acute distress. Several hours after admission she was no longer oriented to time and had difficulty with attention. By the night of hospital day (HD) 0, she was unable to participate in conversations or follow directions. She became physically aggressive toward family members and the care team requiring physical and chemical restraints. Throughout her decline in mental status, the patient’s vital signs remained similar to those on admission. After discussion with her family, the patient was intubated and further sedated to safely complete the evaluation of her severe encephalopathy.

Diagnostic Studies

We had a broad initial differential diagnosis. Based on her initial vital signs and immunocompromised status, we were concerned for sepsis and infection, particularly meningitis and encephalitis. Our patient was prescribed outpatient narcotic pain medications, and we considered opioid withdrawal. She had received benzodiazepines and antipsychotics for agitation overnight and we thought about paradoxical reactions to these medications. We scrutinized her medication list for agents with known psychiatric side effects. The onset of altered mental status occurred very quickly after admission, making hospital-acquired delirium less likely, and we were hesitant to attribute her behavior to a primary psychiatric illness as she had no significant psychiatric history.

Blood work revealed a white blood cell count of 12.8 and hemoglobin of 12.1. Creatinine was mildly elevated and gradually increased during hospitalization, reaching a maximum of 2.7 mg/dL. Electrolytes were within normal limits. Serum ammonia was 169 μM/L (Figure 1). Liver function tests (LFTs) were normal. Blood cultures did not grow any organisms. Urine ureaplasma assay was negative.

Figure 1.

Ammonia levels (μM/L) throughout hospital course. Patient could not tolerate hyperammonemia treatment until intubated and sedated, resulting in a delay in lactulose initiation.

Lumbar puncture revealed an elevated opening pressure of 32 cm H₂O. She had a normal cerebrospinal fluid (CSF) cell count, glucose, and protein level. Extensive infectious testing of the CSF all resulted negative.

A liver ultrasound was unremarkable, including patency of the portal and hepatic veins. CT brain showed no acute findings. MRI of the brain demonstrated T2 FLAIR hyperintensities of the bilateral occipital and parietal cortex with contrast enhancement consistent with posterior reversible encephalopathy syndrome (PRES; Image 1).

Image 1.

Four millimeters axial T2 FLAIR MRI showing abnormal FLAIR hyperintensities in the bilateral occipital and parietal cortexes consistent with PRES. PRES, posterior reversible encephalopathy syndrome.

Discussion

PRES and RCVS

PRES was first described in 1996. ¹ It is considered a diagnosis of exclusion in patients with acute, reversible neurologic changes and typical neuroimaging findings. Seizures have been described as the most common presenting symptom, but neurologic changes can also include headache, somnolence, altered mental status, and visual changes. MRI classically demonstrates occipital lobe involvement. Several theories exist regarding the pathogenesis of PRES. The most widely held hypothesis is that hyperperfusion due to elevated blood pressure causes vascular leak, loss of autoregulation, and vasogenic edema. Endothelial dysfunction caused by toxins is another proposed mechanism. Patients at risk of developing PRES include those with hypertensive emergency, large fluctuations in blood pressure, pre-eclampsia, sepsis, and those taking immunosuppressive medications. Treatment is primarily supportive and prioritizes blood pressure control. Patients typically have a full recovery, although there are cases of persistent neurologic disorder after the acute phase. ²

RCVS has similar clinical, imaging, and epidemiological features to PRES. Like PRES, the pathophysiology is unclear but is thought to be secondary to dysfunction of cerebral vasculature autoregulation and changes in the blood brain barrier. ³

PRES in Transplant

PRES has been well documented in transplant patients on immunosuppressive medications and is associated with calcineurin inhibitors (CNI).^4-7 In transplanted patients on tacrolimus, incidence of PRES has been reported from 0.34% to 1.6%. ⁶ Interestingly, there is no correlation between CNI level and development of PRES; most patients have therapeutic CNI levels at onset. No difference in incidence has been found between patients who discontinue the drug and those who do not. There is little information in the literature regarding the risk of PRES in patients who have previously taken a CNI. However, time between transplant and onset of PRES is usually within months. One retrospective chart review identified 19 stem cell transplant patients with a median time to PRES of 85 days (maximum of 248 days). ⁸ Genetic researchers have suggested polymorphisms that predispose patients on CNIs to neurotoxicity. ⁶

Hyperammonemia Syndrome

Isolated hyperammonemia is rare, but often fatal when it occurs in immunocompromised patients. Most often described following solid organ transplant, hyperammonemia syndrome is characterized by elevated serum ammonia levels with clinical symptoms and no alternative etiologies identified. Patients generally have normal or mildly abnormal LFTs and new onset neurologic symptoms including confusion, agitation, seizures, and coma. Levels greater than 200 µM/L are associated with poor neurologic outcomes, cerebral edema, and brainstem herniation.^9,10 MRI findings include involvement of the cingulate and insular cortex with additional variable cortical involvement that may be misinterpreted as PRES. ¹⁰ Several theories exist regarding the etiology of isolated hyperammonemia in transplant recipients. The most widely accepted theory involves development of a systemic ureaplasma infection due to donor colonization. ⁹

Hyperammonemia Syndrome in BMT

This phenomenon is most well studied in lung transplant patients in whom the incidence is estimated to be between 1% and 4.1%. Mortality rates have been reported to be as high as 50% to 75%. ¹¹ While significantly less common, case reports and small studies have described hyperammonemia syndromes in patients with hematologic disorders after high-dose chemotherapy, hematopoietic stem cell transplant, and BMT.^9,12-14 Literature has typically described these patients presenting earlier than our patient did; Davies et al found a median time to presentation posttransplant of 25 days (maximum of 106 days). ¹³ Empiric treatment for ureaplasma species in stem cell transplant patients with elevated ammonia levels has been reported but literature is limited. ¹⁴

Clinical Course

The patient’s ammonia levels began decreasing shortly after the initiation of lactulose, protein restriction, rifaximin, and doxycycline for empiric ureaplasma coverage (Figure 1). By HD 6 hyperammonemia and encephalopathy had resolved. She was extubated with only minor residual confusion and returned to neurologic baseline by HD 10. She continued to recover and was eventually discharged home after a 20-day hospitalization.

The diagnosis in this case was reached through a process of exclusion. The occurrence of these events in hospitalizations following BMT implies an etiology related to the transplant itself over other epidemiologic risk factors. This is further supported by the donor’s history of RCVS as well as the patient’s history of treatment with tacrolimus, a known risk factor for PRES. It is worth noting that the donor was diagnosed with RCVS without MRI findings and may have been PRES misidentified. While overlap does exist between the radiographic features of RCVS, PRES, and hyperammonemia syndrome, this patient’s MRI was primarily isolated to the parietal and occipital cortex with contrast enhancement that is mainly consistent with PRES.

It is unclear to what extent each disease process contributed to her presentation. However, the timeline of her clinical improvement correlated with the correction of serum ammonia. The etiology of hyperammonemia remains uncertain in light of the negative ureaplasma assay, although this mechanism is less established when related to BMT. Our patient also developed both hyperammonemia and PRES beyond the expected time frame, suggesting the importance of considering these diagnoses in patients with more remote BMT histories.

This is a case of 2 rare and poorly understood transplant-related causes of encephalopathy occurring simultaneously. While thought to be pathophysiologically distinct, further investigation is warranted to identify a potential link.