Low plasma free choline is prevalent in patients receiving long term parenteral nutrition and is associated with hepatic aminotransferase abnormalities

Hepatic transaminase abnormalities have been previously reported in patients receiving long term total parenteral nutrition (PN). We sought to determine if such abnormalities are caused by choline deficiency-induced hepatocyte damage. In 41 subjects (19 male, 22 female) aged 45.1 +/- 24.3 years (range 0.1-79 years) who have received PN for 5.5 +/- 4.7 years (range 0.1-14.5 years). We determined plasma free and phospholipid bound choline levels, serum albumin, ALT and AST. We also determined the daily volume of intravenous lipid emulsion received by the patients as well as the concentration of free choline and phospholipid bound choline in the lipid emulsion. Plasma free choline was low in 33 41 subjects (mean 7.15 +/- 2.5 nmol/ml, range 3.3-15.6, normal 11.4 +/- 3.7). Phospholipid bound choline was normal in 34 41 subjects (mean 2157 +/- 620 nmol/ml, range 1026-3887, normal 2364 +/- 774). Elevations in ALT and AST were significantly correlated with plasma free choline (r = -0.34, p = 0.03, r = -0.37, p = 0.02 respectively) but not with phospholipid bound choline. No relationship was found between age, PN duration or daily volume of intravenous lipid and plasma free or phospholipid bound choline. The lipid emulsion contained 24 +/- 6 nmol/ml of free choline and 11 630 +/- 552 nmol/ml of phospholipid bound choline. We conclude that low plasma free choline is prevalent in patients receiving long term PN and this abnormality is associated with elevated serum aminotransferases. Furthermore, intravenous lipid emulsion is an inadequate source of choline for this patient group.     

Choline deficiency is associated with increased risk for venous catheter thrombosis

BACKGROUND: Patients with intestinal failure who require long-term parenteral nutrition (PN) develop catheter thrombosis as a complication. This patient group may also develop choline deficiency because of a defect in the hepatic transsulfuration pathway in the setting of malabsorption. This study was undertaken to determine whether choline deficiency is a risk factor for development of catheter thrombosis. METHODS: Plasma free and phospholipid-bound choline concentrations were measured in a group of 41 patients that required long-term PN. Episodes of catheter thrombosis from onset of PN to the time of blood testing were recorded. RESULTS: Sixteen (39%) patients developed catheter thrombosis, and 5 of these had recurrent catheter thrombosis. Plasma free choline was 7.7 +/- 2.7 nmol/mL in patients with no history of catheter thrombosis and 6.2 +/- 1.7 nmol/mL in patients with previous catheter thrombosis (p = .076 by Wilcoxon rank-sum test). The partial correlation between plasma free choline concentration and the frequency of clots after controlling for catheter duration was r = -0.33 (p = .038). The relative risk for catheter thrombosis in subjects with a plasma free choline concentration <8 nmol/mL was 10.0, 95% confidence interval (1.134-88.167). Plasma phospholipid-bound choline concentration was 2191.7 +/- 679.0 nmol/mL in patients with previous catheter thrombosis and 2103.3 +/- 531.2 nmol/mL in patients without history of catheter thrombosis (p = NS). CONCLUSION: Choline deficiency is a significant risk factor for development of catheter thrombosis in patients with intestinal failure who require PN.     

Plasma free, phospholipid-bound and urinary free choline all decrease during a marathon run and may be associated with impaired performance

BACKGROUND: Previous investigations have shown that plasma free choline decreases during long distance running. OBJECTIVE: This study was undertaken to determine if body choline status changes during a marathon run and whether performance is thereby adversely affected. DESIGN: Twenty-three accomplished marathon runners 25 to 49 years of age were studied before and after the 1997 Houston-Methodist Marathon. Fasting blood and five-hour urine samples were obtained in the morning, 14 days prior to the race, immediately after the race and approximately 48 hours after completion of the race. Runners were asked to predict their finish times two weeks prior to the race. Performance was indicated by the ratio of predicted to actual time. RESULTS: Both plasma free and phospholipid-bound choline concentrations as well as urinary free choline concentration decreased immediately following the race (19.2+/-4.5 to 14.6+/-4.2 nmol/mL, p=0.005, and 2565.2+/-516.4 to 2403.4+/-643.0 nmol/mL, p=0.068, respectively) and, except for the phospholipid-bound choline, rebounded towards baseline after 48 hours (15.6+/-3.2 and 2299.9+/-426.7 nmol/mL), although plasma concentrations remained significantly below baseline. Plasma free and phospholipid-bound choline concentrations were significantly correlated (r=0.46, p=0.0001), although urinary free choline concentration was not correlated with either. There was no correlation between plasma free, phospholipid-bound or urinary free choline concentration and actual finish time or the ratio of predicted to actual finish time. However, the percent decrease in urinary free choline concentration was significantly correlated with the ratio of predicted to actual time (r=0.47, p=0.036). No relationship was seen between this ratio and the percent decrease in either plasma free or phospholipid-bound choline concentrations immediately after the race. CONCLUSION: Our finding of both decreased free and phospholipid-bound choline suggests the decrease in choline status is related to accelerated choline metabolism or enhanced choline uptake by tissues rather than decreased hepatic choline release. The role of choline supplementation during endurance running requires further investigation.     

Plasma carnitine levels in patients receiving home parenteral nutrition

Patients on long-term home parenteral nutrition (HPN) are known to frequently develop hepatic steatosis or steatohepatitis. The etiology of this steatosis or steatohepatitis is unknown, but carnitine deficiency has been one of the postulated mechanisms. The importance of L-carnitine in hepatic fatty acid oxidation and the steatosis observed in primary and acquired carnitine deficiencies prompted us to determine plasma carnitine levels in 37 patients receiving long-term HPN. Thirteen patients (35%) had low total and free plasma carnitine levels. Fifteen of the 37 HPN patients were matched for age and sex with 15 patients with Crohn's disease who did not require HPN. Mean total and free plasma carnitine values were significantly lower (p less than 0.001) in these 15 HPN patients (32.2 +/- 11.9 and 28.4 +/- 10.8) when compared to Crohn's patients not requiring HPN (49.1 +/- 10.9 and 46.4 +/- 11.5). Associations were not detected between plasma carnitine and clinical or biochemical parameters that might have explained the low values.     

Choline deficiency causes reversible hepatic abnormalities in patients receiving parenteral nutrition: proof of a human choline requirement: a placebo-controlled trial

BACKGROUND: Previous studies have shown that plasma free choline concentrations are significantly decreased in many long-term home total parenteral nutrition (TPN) patients. Furthermore, low choline status has been associated with both hepatic morphologic and hepatic aminotransferase abnormalities. A preliminary pilot study suggested choline-supplemented TPN may be useful in reversal of these hepatic abnormalities. METHODS: Fifteen patients (10 M, 5 F) who had required TPN for > or =80% of their nutritional needs were randomized to receive their usual TPN (n = 8), or TPN to which 2 g choline chloride had been added (n = 7) for 24 weeks. Baseline demographic data were similar between groups. Patients had CT scans of the liver and spleen, and blood for plasma free and phospholipid-bound choline, alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase, gamma glutamyl transferase (GGT), bilirubin, serum lipids, complete blood count (CBC), and chemistry profile obtained at baseline, and weeks 2, 4, 6, 12, 16, 20, 24, and 34. CT scans were analyzed for Hounsfield unit (HU) densities. RESULTS: There were no significant differences in any measured parameters after 2 weeks. However, at 4 weeks, a significant difference in liver HU between groups was observed (13.3+/-5.0 HU [choline] vs 5.8+/-5.2 HU [placebo], p = .04). This significant trend continued through week 24. Recurrent hepatic steatosis and decreased HU were observed at week 34, 10 weeks after choline supplementation had been discontinued. A significant increase in the liver-spleen differential HU was also observed in the choline group (10.6+/-6.2 HU [choline] vs 1.3+/-3.3 HU [placebo], p = .01). Serum ALT decreased significantly (p = .01 to .05) in the choline group vs placebo at weeks 6,12, 20, and 24. Serum AST was significantly decreased in the choline group by week 24 (p = .02). The serum alkaline phosphatase was significantly reduced in the choline group at weeks 2, 12, 20, 24, and 34 (p = .02 to 0.07). Total bilirubin was normal in these patients and remained unchanged during the study. Serum GGT tended to decrease more in the choline group, but the greater decrease was not statistically significant. CONCLUSIONS: Choline deficiency is a significant contributor to the development of TPN-associated liver disease. The data suggest choline is a required nutrient for long-term home TPN patients.     

Plasma choline in normal newborns, infants, toddlers, and in very-low-birth-weight neonates requiring total parenteral nutrition

Choline deficiency is associated with hepatic abnormalities in adult volunteers and patients administered total parenteral nutrition (TPN). Preliminary investigation has suggested that plasma-free choline concentration (PFCh) is greater in neonatal animals, including humans, than in adults. The aims of this study were to determine the normal PFCh and phospholipid-bound choline concentration (PPLBCh) for newborns, infants, and toddlers and to determine the change during TPN. We also sought to determine the degree of fetal choline extraction, the relation between maternal and newborn plasma choline concentrations, and the relation between plasma choline status and normal newborn length, weight, and gestational age. Blood samples were obtained from 104 full-term newborns in two centers (Ben Taub and Maimonides), 25 mothers, 21 normal infants aged 20.3 +/- 11.8 wk, 12 normal infants aged 62.4 +/- 3.9 wk, and 14 preterm infants (gestational age = 28.9 +/- 2.2 wk) who required TPN. The vein PFChs were 28.1 +/- 13.0 nmol/mL (Ben Taub) and 68.1 +/- 16.9 nmol/mL (Maimonides). The artery PFChs were 27.1 +/- 13.0 nmol/mL (Ben Taub) and 57.9 +/- 11.6 nmol/mL (Maimonides). The vein PPLChs were 1004.7 +/- 246.6 nmol/mL (Ben Taub) and 1121.2 +/- 289.6 nmol/mL (Maimonides). The artery PPLChs were 1065.7 +/- 469.3 nmol/mL (Ben Taub) and 1106.9 +/- 285.8 nmol/mL (Maimonides). The vein-minus-artery differences for PFCh were 1.0 +/- 9.7 nmol/mL (Ben Taub) and 10.2 +/- 10.9 nmol/mL (Maimonides). The vein-minus-artery differences for PPLCh were -51.9 +/- 398.2 nmol/mL (Ben Taub General Hospital, Houston, Texas) and 14.4 +/- 254.3 nmol/mL (Maimonides, New York, New York). Maternal venous PFCh was 8.4 +/- 3.1 nmol/mL. Maternal venous PPLCh was 2592.1 +/- 584.0 nmol/mL (range = 1227.8-3729.0). Maternal venous PFCh correlated with newborn arterial PFCh (r = 0.53, P < 0.05) but not with newborn venous PFCh. No correlation was seen between maternal venous and newborn PPLCh. No significant differences were seen in PPLCh or choline extraction in Ben Taub versus Maimonides patients, although PFCh was significantly greater in the newborns from Maimonides (P < 0.05). The mean venous PFCh and PPLCh in the preterm infants before beginning TPN was 21.2 +/- 6.3 and 1366.8 +/- 339.1 nmol/mL, respectively. Just before initiation of tube feeding (4.0 +/- 2.7 d after TPN had been started), mean venous PFCh and PPLCh was 18.4 +/- 5.3 and 2251.8 +/- 686.9 nmol/mL, respectively. When TPN was discontinued and tube feeding increased to goal, after 10.8 +/- 10.4 d, venous PFCh and PPLCh was 22.6 +/- 8.7 and 2072.5 +/- 540.6 nmol/mL, respectively. Venous PFCh and PPLCh was 13.4 +/- 2.5 and 1827.5 +/- 327.0 nmol/mL, respectively in the older infant group. In conclusion, newborn PFCh is significantly greater than PFCh in adults but falls to adult levels within the first year of life. Low maternal PFCh may be associated with low newborn PFCh. Normal newborn plasma choline status has no bearing on intrauterine growth, although the role of maternal choline deficiency in underweight newborns is unknown. Newborn PPLCh is substantially below that of adults, which suggests its use in membrane synthesis during growth.     

Changes in plasma free and phospholipid-bound choline concentrations in chronic hemodialysis patients.

BACKGROUND: Choline deficiency may develop in malnourished patients, those with cirrhosis, and those who require total parenteral nutrition. Previous data has suggested an important role for the kidneys in the maintenance of choline homeostasis. OBJECTIVE: This study was undertaken to determine the change in plasma choline during hemodialysis and to determine if it was lost in the dialysate. DESIGN: Thirteen adult patients (10 men, 3 women) who had required hemodialysis for a mean of 10.8 years were studied. Dialysis was performed 3 times weekly for 4 hours using either a cellulose acetate or polysulfone dialyzer membrane. Venous and arterial blood, and dialysate samples were taken for measurement of plasma free and phospholipid-bound choline concentration before beginning dialysis and after each hour of dialysis. An in vitro system was devised to determine if choline could bind to a significant degree to the dialysis membrane. RESULTS: Plasma free choline concentration was increased above normal (11.7 +/- 3.7 nmol/mL) at baseline and declined progressively during dialysis. In contrast, plasma phospholipid-bound choline concentration increased progressively during dialysis. The decrease in plasma free choline (-1.8 +/- 0.3 nmol/mL(-1)/h(-1); P = 1.6 x 10(-6)) was almost entirely related to that which was removed during dialysis, although the magnitude of the loss was not correlated with the increase in plasma phospholipid-bound choline concentration (125 +/- 20.5 nmol/mL(-1)/h(-1); P < 1.2 x 10(-6)). Patients lost a mean of 246 pmol of free choline during hemodialysis. Choline did not bind to the dialysis membrane. CONCLUSION: Plasma free choline concentration is elevated before dialysis, and choline is lost to a significant degree in the dialysate. Further investigation is necessary to determine whether a transient, dialysis-induced choline deficiency develops, and whether there is a role for choline supplementation in these patients. The choline homeostatic mechanism requires further investigation in renal failure patients.     

The effect of lecithin supplementation on plasma choline concentrations during a marathon

BACKGROUND: Previous studies have shown that plasma and urinary free choline concentrations decrease significantly during a marathon, and that these decreases may be associated with decreased performance. OBJECTIVE: In a pilot study, we sought to determine whether lecithin supplementation prior to a marathon would maintain plasma free and urinary choline concentrations and improve performance versus placebo. METHODS: 12 accomplished marathon runners, males (7) and females (5), 21 to 50 years of age were randomized to receive lecithin (4 capsules BID; PhosChol 900) or placebo beginning one day prior to the 2000 Houston-Methodist Health Care Marathon. The lecithin supplement provided approximately 1.1 g of choline on a daily basis (2.2 g total). Runners estimated finish time based on recent performance and training. Fasting, pre- and post-marathon plasma and a five-hour urine collection were analyzed for free choline and plasma for phospholipid-bound choline. Pre-race predicted, as well as the actual finish time, were recorded. RESULTS: All subjects completed the marathon. Plasma free choline decreased significantly in the placebo group and increased significantly in the lecithin group (9.6 +/- 3.6 to 7.0 +/- 3.6 nmol/mL vs. 8.0 +/- 1.2 to 11.7 +/- 3.6 nmol/mL, p = 0.001 for the delta between groups). No significant changes in plasma phospholipid-bound choline concentration were observed. There was a non-significant decrease in urine free choline in both groups. Actual finish time was 256.3 +/- 46.3 minutes for the lecithin group vs. 240.8 +/- 62.0 for the placebo group and the actual:predicted time was 1.03 +/- 0.06 (lecithin) and 1.07 +/- 0.08 (placebo), p = 0.36. CONCLUSION: Short-term lecithin supplementation prior to a marathon maintains normal plasma free choline concentration during the race, but failed to improve performance.     

Plasma choline concentration in humans fed parenterally

Choline is an essential nutrient for some mammals; it is used for membrane and neurotransmitter synthesis. We analyzed plasma samples, obtained periodically during TPN therapy, for choline concentration. Malnourished patients referred to a nutrition support service were prospectively assigned to be treated with daily infusions of amino acids with, and without, supplemental daily infusions of lipid emulsion for a period of 1 wk. After the first week, all subjects received intravenous lipid, and most were offered enteral food supplements. Initial plasma choline concentrations in the 25 malnourished patients were significantly lower than those measured in plasma samples from 23 hospitalized patients known to be eating well (6.5 +/- 0.6 vs 9.7 +/- 0.7 nmol/ml; mean +/- SEM; p less than 0.001). During the first week of TPN therapy, plasma choline concentrations in the lipid-restricted group tended to decrease (from 7.3 +/- 1.0 to 4.7 +/- 0.5 nmol/ml; mean +/- SEM; p less than 0.05), while in the lipid-supplemented group plasma choline tended to increase (from 5.6 +/- 0.5 to 6.2 +/- 0.7 nmol/ml; mean +/- SEM; p less than 0.05). Plasma choline concentration increased during wk 2-4, when all patients were treated with lipid emulsions, and some were offered enteral foods. We conclude that malnourished humans who eat no choline have diminished stores of plasma (and possibly tissue) choline.     

Subject Index to Volume 5

Background: Previous investigations have shown that plasma free choline decreases during long distance running.

Objective: This study was undertaken to determine if body choline status changes during a marathon run and whether performance is thereby adversely affected.

Design: Twenty-three accomplished marathon runners 25 to 49 years of age were studied before and after the 1997 Houston-Methodist Marathon. Fasting blood and five-hour urine samples were obtained in the morning, 14 days prior to the race, immediately after the race and approximately 48 hours after completion of the race. Runners were asked to predict their finish times two weeks prior to the race. Performance was indicated by the ratio of predicted to actual time.

Results: Both plasma free and phospholipid-bound choline concentrations as well as urinary free choline concentration decreased immediately following the race (19.2±4.5 to 14.6±4.2 nmol/mL, p=0.005, and 2565.2±516.4 to 2403.4±643.0 nmol/mL, p=0.068, respectively) and, except for the phospholipid-bound choline, rebounded towards baseline after 48 hours (15.6±3.2 and 2299.9±426.7 nmol/mL), although plasma concentrations remained significantly below baseline. Plasma free and phospholipid-bound choline concentrations were significantly correlated (r=0.46, p=0.0001), although urinary free choline concentration was not correlated with either. There was no correlation between plasma free, phospholipid-bound or urinary free choline concentration and actual finish time or the ratio of predicted to actual finish time. However, the percent decrease in urinary free choline concentration was significantly correlated with the ratio of predicted to actual time (r=0.47, p=0.036). No relationship was seen between this ratio and the percent decrease in either plasma free or phospholipid-bound choline concentrations immediately after the race.

Conclusion: Our finding of both decreased free and phospholipid-bound choline suggests the decrease in choline status is related to accelerated choline metabolism or enhanced choline uptake by tissues rather than decreased hepatic choline release. The role of choline supplementation during endurance running requires further investigation.     

Plasma vitamin E and selenium and breath pentane in home parenteral nutrition patients

Because both vitamin E and selenium protect against lipid peroxidation, we evaluated the relationship between breath pentane, evolved from the peroxidation of linoleic acid, and plasma levels of alpha-tocopherol (vitamin E), Se, and Se-dependent glutathione peroxidase (Se-GSHPx). Nine home parenteral-nutrition (HPN) patients received added Se in intravenous solutions and were compared with 10 normal control subjects. The excretion of pentane (pmol.kg-1.min-1, means +/- SEM) in control subjects (6.34 +/- 0.96) was significantly lower than in HPN patients (15.02 +/- 1.12, p less than 0.001). alpha-Tocopherol (mumol/L), Se (mumol/L), and Se-GSHPx (U) values were, respectively, 18.13 +/- 1.70, 1.70 +/- 0.05, and 5.34 +/- 0.27 in control subjects and 10.21 +/- 1.66, 1.35 +/- 0.14, and 7.01 +/- 0.31 in HPN patients. All differences were statistically significant. Significant negative correlations were observed between plasma alpha-tocopherol levels and HPN duration and between pentane output and plasma alpha-tocopherol levels (r = -0.58, p less than 0.01). In HPN patients with reduced plasma alpha-tocopherol levels associated with increased pentane output, there is, inferentially, increased lipid peroxidation despite normal plasma Se and Se-GSHPx levels.     

Evidence of choline depletion and reduced betaine and dimethylglycine with increased homocysteine in plasma of children with cystic fibrosis

Cystic fibrosis (CF) is associated with many clinical complications including steatosis for which the relation to defective CF transmembrane conductance regulator protein is unclear. Choline deficiency results in hepatic steatosis. Choline is the precursor of betaine, which donates methyl groups for remethylation of homocysteine to methionine and dimethylglycine. Previously, we have shown phospholipid malabsorption and increased plasma homocysteine in children with CF. In these studies we used normal phase HPLC with tandem mass spectrometry to determine plasma choline, betaine, and dimethylglycine in children with CF (n = 34) and healthy control children without CF (n = 15). Plasma choline, betaine, and dimethylglycine were significantly lower in children with CF (means +/- SEM, 6.48 +/- 0.35, 23.8 +/- 1.49, 1.49 +/- 0.13 mumol/L, respectively) than in children without CF (8.98 +/- 0.46, 37.3 +/- 1.84, 3.01 +/- 0.17 mumol/L, respectively). Plasma choline (r = 0.373, P = 0.007) and betaine (r = 0.399, P = 0.005) were positively related to methionine, and choline was inversely related to homocysteine (r = -0.316, P = 0.03). Choline, betaine, and dimethylglycine were all significantly and positively related to the plasma S-adenosylmethionine:S-adenosylhomocysteine (SAM:SAH) ratio (r = 0.294, r = 0.377, r = 0.442, respectively; P < 0.05). The plasma choline:betaine and betaine:dimethylglycine ratios did not differ between the children with CF and the control children, suggesting no increase in betaine synthesis, or betaine-dependent remethylation of homocysteine. These studies suggest that choline depletion may contribute to increased homocysteine in children with CF. Choline depletion and altered thiol metabolism may contribute to the clinical complications associated with CF.     

Trace element nutrition status and dietary intake of children with phenylketonuria

The trace element status (copper, iron, zinc, manganese, chromium, and selenium) of 20 dietetically treated phenylketonuric (PKU) children was assessed. Significantly higher intakes of copper (p = 0.002) and iron (p = 0.005) were noted in PKU children compared with their siblings. No significant differences were found for zinc, manganese, or chromium. Intake of selenium was significantly lower (p = 0.0001) in PKU children (8.4 +/- 3.9 micrograms/d) than in siblings (41.6 +/- 9.4 micrograms/d). Plasma and urine selenium and erythrocyte glutathione peroxidase activity (GSHpx) were significantly lower (p = 0.001) in PKU children (0.38 +/- 0.11 mumol/L, 58.0 +/- 34.5 nmol/d, and 14.2 +/- 5.5 U/g Hb, respectively) than in siblings (0.82 +/- 0.15 mumol/L, 165.2 +/- 49.4 nmol/d, and 22.7 +/- 5.2 U/g Hb, respectively). No differences were found in plasma and urine concentrations of other elements. Intake of selenium was significantly correlated with erythrocyte GSHpx (r = 0.87, p = 0.0001) and plasma selenium (r = 0.71, p = 0.0001) for the combined groups. The need and possible procedures, including dietary manipulation, for increasing selenium intake in PKU subjects are discussed.     

Home parenteral nutrition: does affiliation with a national support and educational organization improve patient outcomes?

BACKGROUND: For patients receiving home parenteral nutrition (HPN), catheter-related bloodstream infection (CRBSI) and reactive depression may significantly impact quality-of-life. This study evaluated the influence of patient affiliation with a national organization promoting HPN education and peer support on these outcome variables. METHODS: Using a case-control design, we compared 2 groups of affiliated patients with nonaffiliated controls, who were matched for diagnosis, HPN duration, sex, and age. Group 1 data were obtained from patients in large HPN medical practice programs. Group 2 data were obtained from patients in small medical practices with a small number of HPN patients. All participants were evaluated by structured interviews every 6 months over 18 months. RESULTS: In both data collection groups, affiliated patients (A) had significantly higher (mean +/- SD) quality-of-life scores compared with nonaffiliated patients (NA): (Gr 1: A, 19.8 +/- 4.7 versus NA, 17.6 +/- 5.6, p = .05; Gr 2: A, 20.4 +/- 5.2 versus NA, 17.3 +/- 4.8, p = .05). Affiliated patients also had lower depression scores (Gr 1: A, 10.9 +/- 10.4 versus NA, 20.4 +/- 13.6, p = .01; Gr 2: A, 12.5 +/- 9.6 versus NA, 18.5 +/- 10.8, p = .03) and a lower incidence of catheter-related infections (Gr 1: A, 0.10 +/- 0.3 versus NA, 0.60 +/- 0.55, p = .01; Gr 2: A, 0.27 +/- 0.55 versus NA, 0.71 +/- 0.64, p = .02) than nonaffiliated patients. CONCLUSIONS: Affiliation with an organization that provides ongoing HPN education and peer support was associated with significantly better HPN outcomes. Alternative explanations are discussed in relation to limitations of the case-control design.     

Metabolic bone disease in patients receiving home parenteral nutrition: a Canadian study and review

BACKGROUND: Metabolic bone disease (MBD) is a significant complication in patients receiving long-term home parenteral nutrition (HPN). Pamidronate has been poorly studied in this population. We examine the prevalence and risk factors for MBD and examine changes in bone mineral density (BMD) after pamidronate administration. METHODS: First, a chart review of patients receiving HPN for >1 year was performed, and Pearson correlations were used to assess associations between MBD (defined as t score<-1) and risk factors. Second, the effect of IV pamidronate on BMD was studied prospectively in 11 HPN patients. Results were compared using a t-test. RESULTS: Charts were reviewed in 25 patients (15 F, 10 M): age, 56.9+/-3.1 years; body mass index (BMI), 21.2+/-0.57 kg/m2; months receiving HPN, 113.2+/-0.09; and days per week receiving HPN, 5.08+/-0.39. MBD was present in 33% of patients for the spine and hip and in 50% for the femoral neck; 24% had previous fractures. There was a significant negative correlation between the duration of HPN and BMD (r=-0.40) for all measurements. From those patients, 11 received IV pamidronate for a mean of 22.2+/-5.4 months. At baseline, their mean HPN treatment duration was 10.6+/-6.3 years. Overall, BMD results showed a trend toward improvement in the mean t score of the spine and hip postpamidronate therapy (pre, -3.1+/-0.75; post, -2.9+/-0.69; p=.07). After excluding 2 patients receiving corticosteroids, the mean t score of the spine showed significant improvement (prepamidronate -3.4+/-0.57 vs post-pamidronate -3.1+/-0.65, p=.036). CONCLUSIONS: In our HPN population, 76% had MBD and 24% had previous fractures. The results suggest that these patients may benefit from pamidronate. More studies are needed to assess the efficacy of pamidronate.     

Bone mineral content in patients on home parenteral nutrition

Bone mineral content (BMC) was monitored in 15 patients with short bowel syndrome receiving home parenteral nutrition (HPN). Thirteen patients had Crohn's disease and 2 ulcerative colitis (mean age 36 years, range 23-69 years). During the study the patients received HPN for a mean period of 62 months, range 20-106 months. At the time of inclusion the patients had a significantly reduced BMC of lumbar spine and femoral neck compared to normals (Z-scores = -3.35 +/- 3.49, p < 0.05 and Z-score = -2.23 +/- 2.11, p < 0.05). During HPN the Z-score of lumbar spine BMC decreased in 8 patients and increased slightly or was unchanged in 7 patients. The mean Z-score of BMC of lumbar spine declined by 1.46 +/- 2.48 (p < 0.05) and the Z-score of femoral neck BMC declined by 0.831 +/- 1.14 (p < 0.05). This corresponds to a yearly decrease of lumbar spine BMC of 4%. There was no correlation between the decline in BMC during the study and the period of length the patients were on HPN. We conclude that patients on HPN have a low bone mineral density and that the bone loss continues during prolonged HPN.     

Serum free carnitine and total triglycerid levels in children with iron deficiency anemia

Iron deficiency anemia and hyperlipidemia are common public health problems in Turkey. The connection between iron and lipid metabolisms has not been clarified yet. The aim of the study was to determine the effect of iron deficiency on carnitine and lipid metabolism. Study group was consisted of 70 children (mean age 14.7 +/- 1.3 months) suffering from iron deficiency anemia and 20 healthy children (mean age 13.7 +/- 1.2 months) attended to outpatient clinics of Cumhuriyet University, Sivas were enrolled the study as the control group. Assessments of serum free carnitine concentrations, total triglyceride, total cholesterol and VLDL levels were made in both groups. The mean serum free carnitine concentration was significantly lower than the control group (18.9 +/- 0.43 nmol/ml and 45.9 +/- 1.47 nmol/ml respectively, t = 17.5 p < 0.01). Results of our study also indicated higher serum total triglyceride, total cholesterol and VLDL levels in iron deficient patients than the healthy controls. Regression analyses indicated a negative correlation between serum free carnitine and total triglyceride levels in iron deficient patients. This study confirms that iron deficiency anemia may be linked to the endogenous carnitine synthesis in pediatric age group, and thus hyperlipidemia appears to be a risk factor for premature cardiovascular diseases.     

Verbal and visual memory improve after choline supplementation in long-term total parenteral nutrition: a pilot study

BACKGROUND: Previous investigations have demonstrated that choline deficiency, manifested in low plasma-free choline concentration and hepatic injury, may develop in patients who require long-term total parenteral nutrition (TPN). Preliminary studies have suggested lecithin or choline supplementation might lead to improved visual memory in the elderly and reverse abnormal neuropsychological development in children. We sought to determine if choline-supplemented TPN would lead to improvement in neuropsychological test scores in a group of adult, choline-deficient outpatients receiving TPN. METHODS: Eleven subjects (8 males, 3 females) who received nightly TPN for more than 80% of their nutritional needs for at least 12 weeks before entry in the study were enrolled. Exclusion criteria included active drug abuse, mental retardation, cerebral vascular accident, head trauma, hemodialysis or peritoneal dialysis, (prothrombin time [PT] >2x control), or acquired immune deficiency syndrome (AIDS). Patients were randomly assigned to receive their usual TPN regimen (n = 6, aged 34.0 +/- 12.6 years) over a 12-hour nightly infusion or their usual TPN regimen plus choline chloride (2 g) (n = 5, aged 37.3 +/- 7.3 years). The following neuropsychological tests were administered at baseline and after 24 weeks of choline supplementation (or placebo): Weschler Adult Intelligence Scale-Revised (WAIS-R, intellectual functioning), Weschler Memory Scale-Revised (WMS-R, two subtests, verbal and visual memory), Rey-Osterrieth Complex Figure Test (visuospatial functioning and perceptual organization), Controlled Oral Word Association Test (verbal fluency), Grooved Pegboard (manual dexterity and motor speed), California Verbal Learning Test (CVLT, rote verbal learning ability), and Trail Making Parts A & B (visual scanning, psychomotor speed and set shifting). Scores were reported in terms of standard scores including z scores and percentile ranks. Mean absolute changes in raw scores were compared between groups using the Wilcoxon rank sum test, where p values < .05 constituted statistical significance. RESULTS: Significant improvements were found in the delayed visual recall of the WMS-R (7.0 +/- 2.7 vs -.33 +/- 5.7, p = .028), and borderline improvements in the List B subset of the CVLT (1.0 +/- 0.8 vs -2.0 +/- 2.4, p = .06) and the Trails A test (-3.8 +/- 8.1 vs 3.7 +/- 4.5 seconds, p = .067). No other statistically significant changes were seen. CONCLUSIONS: This pilot study indicates both verbal and visual memory may be impaired in patients who require long-term TPN and both may be improved with choline supplementation.     

Reduced concentration of hepatic alpha-tocopherol in patients with alcoholic liver cirrhosis.

The concentration of alpha-tocopherol was measured in liver biopsy specimens obtained from 83 patients with alcoholic and non-alcoholic liver diseases. The mean hepatic vitamin E content (as alpha-tocopherol) was significantly lower in 23 patients with alcoholic cirrhosis (17.6 +/- 12.1 nmol/mg wet weight liver), compared with 12 patients with normal liver histology (39.2 +/- 29.7 nmol/mg, P less than 0.01). The mean serum concentration of alpha-tocopherol was lower in patients with alcoholic cirrhosis (13.9 +/- 7.0 mumol/l) than in individuals with alcoholic fatty liver (21.3 +/- 9.3 mumol/l, P less than 0.01) and patients with normal liver histology (23.4 +/- 11.6 mumol/l, P less than 0.01). A decreased ratio of serum alpha-tocopherol/total serum lipids was also observed in patients with alcoholic cirrhosis, compared with patients with normal liver histology (P less than 0.05). There was a significant correlation between concentrations of alpha-tocopherol in liver and serum (r = 0.43, P less than 0.001). Furthermore, serum alpha-tocopherol correlated with retinol (r = 0.53, P less than 0.001), selenium (r = 0.45, P less than 0.001), and albumin (r = 0.37, P less than 0.001) in serum. We suggest that the reduced content of hepatic alpha-tocopherol observed in some patients may play a role in ethanol-induced lipid peroxidation.     

The effect of choline supplementation on hepatic steatosis in the parenterally fed rat

Information regarding hepatic function during total parenteral nutrition in rats is often extrapolated to the clinical situation, but the steatosis observed in that species may simply reflect choline deficiency and be irrelevant to man. The effect of choline supplementation on hepatic lipid content and triglyceride secretion was examined in parenterally fed rats. Eighty to 90-day-old rats were randomized into three groups; group I received oral Purina Chow ad libitum, groups II and III received identical total parenteral nutrition regimens with the exception that group III received supplemental choline. After 7 days, peripheral triglyceride uptake was inhibited with Triton WR1339, the rate of secretion of 14C-labeled triglyceride measured after a bolus injection of 1-14C-palmitic acid, and total hepatic lipid content was measured. Total hepatic lipid content was elevated in group II (86.3 mg/g) and group III (83.3 mg/g), and both differed significantly from the control group I (35.2 mg/g, p less than 0.01), but the choline supplementation appeared to make no difference. Hepatic secretion of 14C-palmitic acid as 14C-triglyceride was reduced in group II (0.73%/ml plasma), and group III (0.72%/ml plasma) compared to group I (1.06%/ml plasma, p less than 0.05), and was unaffected by choline. The hepatic steatosis produced in the parenterally fed rat did not appear to be due to choline deficiency but to some other factors which may be important in man.