Current thoughts on UBA test as a diagnostic for portosystemic shunt

Do I really need to do a bile acid stimulation test to diagnose a portosystemic shunt, or can I use urine bile acids?

For decades, clinicians have clamored for testing to help determine liver function that is sensitive, minimally influenced by extra hepatic disorders, convenient, safe, affordable, easy-to-interpret, widely available and has a one day turn-around time.

We know that persistent elevations of liver enzymes identify liver disease and direct us to go further to identify the type of disease present, but many times the decision to proceed with more invasive testing, such as biopsy, is pursued sooner in the face of an abnormal function test.

Recently, the urine bile acid (UBA) test has been shown to be a convenient means of detecting acquired liver disease causing functional, cholestatic or perfusion abnormalities. Normally, only a small amount of bile acids are eliminated in the urine. However, in both dogs and cats, high total serum bile acids (TSBA) can lead to urinary excretion of water-soluble conjugate and sulfated forms. Two prospective studies proved the UBA in randomly collected urines can effectively screen for abnormally increased TSBA in dogs and cats, and therefore can be used to test for liver dysfunction. Because the UBA is convenient, safe to obtain in very small patients, easy to interpret, sensitive, and not complicated by issues of hemolysis and lipemia, it is gaining wide popularity among clinicians.

It has been noted that animals with portosystemic vascular anomalies (PSVA) do not develop UBA values as high as anticipated considering their dynamically increased postprandial serum bile acids (PSBA). It was hypothesized that collection of random urine samples failed to capture the dynamics of the enterohepatic bile acid challenge as reflected in the PSBA test. In a poster at the 2004 ACVIM meeting, Dr. Center, et al, investigated the UBA test in dogs with PSVA's more fully and particularly the temporal association of UBA with meal consumption. Randomly collected urine samples from dogs (n=71) were entered prospectively into this study after a diagnosis of PSVA was confirmed using Doppler-assisted abdominal ultrasonography and colorectal scintigraphy or radiographic portography; 68 of the 71 dogs had liver biopsies performed.

In a separate study, serial UBA was measured in 14 normal dogs and in 22 dogs with PSVA before and 1, 2, 3, 4, 6, 8, 10 and 12 hours after feeding. Dogs with SBA greater than or equal to 25umol/L or UBA/UCr greater than or equal to 7.3 were considered to have abnormal tests.

In the 71 dogs with PSVA, test sensitivity (percentage with abnormal tests) for FSBA (fasting serum bile acids) was 89 percent, 96 percent for PSBA and 92 percent for UBA/UCr. Consistent with the initial investigation, sensitivity of the UBA/UCr in random urine was between that of the FSBA and PSBA tests. This observation suggested potential value in synchronizing urine sample collection with meal ingestion. This hypothesis was supported by the outcome of the serial UBA/UCr study where abnormally elevated UBA/UCr values were consistently encountered at the four-hour to eight-hour collection intervals (100-percent sensitivity) in dogs with PSVA. The UBA/UCr values in urine collected at these intervals were 100-percent reliable for detecting increased PSBA concentrations and therefore show great value as a clinical tool. Serial UBA/UCr and SBA samples collected in normal dogs clearly defined a lag time between meal induced SBA challenge and increase in UBA values. This data substantiates that timed urine collections have value in reflecting the dynamics of the enterohepatic SBA flux.

These further investigations of the UBA/UCr test confirm its usefulness as a substitute for TSBA values. Specifically, these studies clearly demonstrate that urine samples for UBA/UCr should be collected four to eight hours after feeding for best test performance in detecting portosystemic shunting. Similar testing strategy might optimize test performance for other forms of liver disease. However, the UBA/UCr already rivals the SBA tests in most other disorders. Because this study involved complete urinary bladder voiding every two hours, the dynamics of peak test performance might differ somewhat in clinical patients voiding less frequently (high values may persist in samples collected later).

Expanded information regarding the UBA/UCr provided by these studies brings further confidence in applying this testing strategy to clinical patients. Based on these findings, timed urine samplings could be performed by an informed pet owner at home. This urine test appears to be a useful means for detecting liver dysfunction or portosystemic shunting in canine patients, and especially where there is difficulty in obtaining paired TSBA samples.