Serum ammonia levels: An overused test in the evaluation of altered mental status in patients without chronic liver disease

From the 2019 HVPAA National Conference

Dr. Matthew Schroeder (University of Texas Southwestern), Ms. Punya Chittajallu (University of Texas Southwestern), Dr. Deepak Agrawal (University of Texas Southwestern), Dr. Arjmand Mufti (University of Texas Southwestern), Mr. Chris McKintosh (Parkland Health & Hospital System), Dr. Kavita Bhavan (University of Texas Southwestern)

Background

Serum ammonia level is a frequently utilized laboratory test in the evaluation of altered mental status. While elevations are more commonly seen in patients with cirrhosis, hyperammonemia can be seen in non-hepatic etiologies including drug toxicities (anti-epileptics), aggressive hematologic malignancies (multiple myeloma), malnutrition, gastrointestinal bleeding, organ transplantation, and inborn errors of metabolism (urea cycle disorders). Despite its frequent use, the utility of routine testing of serum ammonia in patients without liver disease has not been clearly demonstrated.

Objective

To evaluate the use of serum ammonia and associated clinical characteristics of patients with hyperammonemia without cirrhosis.

Methods

A retrospective observational cohort study was performed using a query all adult patients in which a venous serum ammonia level was drawn at Parkland Memorial Hospital between January 1, 2012 and November 4, 2018. History of advanced chronic liver disease was established by using ICD-10 codes in the patient’s medical history or calculation of non-invasive indices (FIB-4 <3.25 and APRI <1.00). Manual chart reviews were performed on randomly selected patients in order to validate the study definition of cirrhosis (n=100) and to evaluate provider response to elevated ammonia levels (n=100).

Results

During the study period, 28,827 ammonia levels were drawn on 12,415 patients, of which 6,223 (50.1%) did not have cirrhosis. Manual chart review validated the cohort (98% patients did not have cirrhosis). Baseline characteristics of patients included 3,788 (60.9%) males, median age 53 years, median BMI 26.4, median INR 1.1, and median total bilirubin 0.4 mg/dL . Initial levels were ordered most often in the Emergency Department (4,311, 69.3%) and levels were ordered a mean 1.25 times (range 1-39) per encounter. Ammonia was elevated to ≥50 µ/dl in 1,094 (17.6%) encounters, with ≥100 µ/dl in 97 (1.6%). Lactulose was administered to 250 (22.8%) patients with elevated ammonia. Elevated levels were acknowledged by providers in only 39% of manually reviewed records. Possible non-hepatic etiologies of elevated ammonia included history of seizure disorders and presumed antiepileptics 225 (20.6%), hematologic malignancies in 13 (1.2%), gastrointestinal bleeding in 12 (1.1%), and urea cycle disorders in 3 (0.3%) of patients.

Discussion

Serum ammonia is checked over 1,300 times a year for workup of altered mental status in patients without advanced liver disease, mostly in the Emergency Department. About one in six patients had elevated ammonia levels with many having conditions or medications which are known to cause mild, clinically insignificant, increase in ammonia levels. Elevated levels of ammonia were not acknowledged in a majority of cases and were managed with lactulose in approximately one-fourth of patients, which is often not clinically indicated.

Clinical Implication

The use of serum ammonia levels in the diagnosis of encephalopathy in patients without liver disease is common but should be limited to high risk populations and only after preliminary workup has been completed.

Figures