Choline may be an essential nutrient in malnourished patients with cirrhosis.

Elemental diets designed for nutritional support in protein-calorie malnutrition are often deficient in choline, a nonessential nutrient. Previously, malnourished patients on these diets were found to be at risk of developing plasma choline deficiency. We have now estimated the prevalence of this deficiency by determining fasting plasma levels of choline among cirrhotic and noncirrhotic malnourished male subjects maintained on regular hospital mixed food or elemental parenteral and enteral formulas. Plasma choline concentrations (microM, average +/- SD) were as follows: (i) mixed foods, 11.3 +/- 4.3 for cirrhotic (n = 22) and 9.3 +/- 2.4 for noncirrhotic (n = 12) patients; (ii) parenteral formula, 5.3 +/- 1.6 for cirrhotic (n = 5) and 8.6 +/- 5.2 for noncirrhotic (n = 16) subjects; and (iii) enteral formula, 6.1 +/- 1.2 for cirrhotic (n = 5) and 11.7 +/- 1.9 for noncirrhotic (n = 4) subjects. The level for healthy normal subjects eating mixed foods was 12.0 +/- 2.2. The prevalence of plasma choline deficiency, i.e., plasma levels greater than or equal to 2 SD below the normal average, was as follows: parenteral formula, all cirrhotic and 10 of 16 noncirrhotic subjects; enteral formula, all cirrhotic and none of the noncirrhotic subjects. The reversibility of choline deficiency was examined in a longitudinal study of three phases involving 10 patients--5 with alcoholic cirrhosis (all on enteral formula); 5 noncirrhotic (1 on enteral and 4 on parenteral formula). During phase 1 (3-day equilibration period; ad libitum regular hospital diet), plasma choline levels were within the normal range for all subjects. During phase 2 (2 wk, choline depletion phase, elemental formulas), choline levels were subnormal in all cirrhotic subjects (5.1 +/- 2.0 microM) on enteral formula and all noncirrhotic patients on parenteral formula (5.9 +/- 1.3 microM). During phase 3 (2 wk, choline repletion phase, elemental formula + 6 g choline/day), the levels normalized in all patients (cirrhotic 11.4 +/- 3.1 microM and noncirrhotic 11.9 +/- 3.2 microM). Analyses of abdominal computed tomographic scans and plasma liver chemistries in the cirrhotic subjects during the three phases suggested a correlation between plasma choline deficiency and hepatic steatosis and abnormal liver enzyme levels in some patients. Therefore, choline may be an essential nutrient in malnourished cirrhotic patients and its deficiency may be associated with adverse hepatic effects.