| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters |
1 Medizinische Klinik und Poliklinik I Allgemeine Innere Medizin and
2
Department of Clinical Biochemistry University of Bonn D-53105 Bonn, Germany
aAddress correspondence to this author at: Medizinische Klinik und Poliklinik I, Allgemeine Innere Medizin, Universitaet Bonn, Sigmund-Freud-Strasse 25, D-53105 Bonn, Germany. Fax 49-228-287-4323; e-mail Woitas{at}uni-bonn.de.
To the Editor:
In their recent publication, Ferré et al. (1) investigated homocysteine concentrations in 76 patients with liver cirrhosis. The majority of these patients suffered from alcoholic liver disease (63%). The authors described significantly increased homocysteine in nonabstaining alcoholic patients with cirrhosis. Furthermore, the homocysteine concentrations were inversely correlated with the severity of the liver disease (Child–Pugh score) and with the folate concentrations in these patients. The authors proposed that the homocysteine concentration was influenced by alcohol intake and by the degree of liver impairment. The authors conclude that nonnutritional factors may affect homocysteine concentrations.
Hyperhomocysteinemia also occurs in patients with decreased renal function, but the investigators provided no information about the renal function of these patients. Renal function is considerably reduced in patients with compensated cirrhosis (2). An inverse relationship between the glomerular filtration rate (GFR) and plasma homocysteine has been noted, and declining renal function is associated with higher plasma homocysteine concentrations (3).
We studied 23 patients with liver cirrhosis (median age, 54 years; range, 20–68 years; 16 males, 7 females). The etiology of cirrhosis was alcoholic liver disease in 18 (78%), and ascites was present in 21 (91%). The median Child–Pugh score was 9 (range, 7–14), and the median bilirubin concentration was 36 µmol/L (range, 6.8–416 µmol/L). GFR was determined in all by steady-state inulin clearance and adjusted to a standard body surface of 1.73 m2. The median GFR was 22 mL · min-1 · 1.73 m-2 (range, 13.5–61 mL · min-1 · 1.73 m-2).
The median homocysteine measured by a microplate enzyme immunoassay (Bio-Rad Laboratories Diagnostic Group, manufactured by Axis Shield AS) was 13.3 µmol/L (range, 3.4–27.2 µmol/L), matching the results of Ferré et al. (1). Median serum creatinine was 106 µmol/L (range, 44–168 µmol/L), and median cystatin C was 1.4 mg/L (range, 0.5–2.8 mg/L).
Homocysteine was significantly correlated with serum bilirubin [r = 0.473 (95% confidence interval, 0.076–0.741); z-statistic, 2.3; P = 0.022], serum creatinine [r = 0.514 (95% confidence interval, 0.129–0.764); z-statistic, 2.5; P = 0.01], and serum cystatin C [r = 0.676 (95% confidence interval, 0.366–0.851); z-statistic, 3.7; P = 0.0002], respectively. As expected, we found an inverse correlation of homocysteine with the GFR measured as inulin clearance [r = -0.512 (95% confidence interval, -0.773 to -0.102); z-statistic, -2.4; P = 0.016]. Homocysteine was not significantly correlated with the Child–Pugh score [r = 0.356 (95% confidence interval, -0.066 to 0.670); z-statistic, 1.7; P = 0.1].
Proposed mechanisms of hyperhomocysteinemia include reduced renal elimination of homocysteine and impaired vitamin-dependent liver metabolism (4). Urinary homocysteine excretion in humans is minimal. However, loss of putative renal homocysteine extraction in chronic renal failure has been hypothesized because significant homocysteine uptake has been demonstrated in the healthy rat kidney. In fasting humans with normal renal function, however, no significant net renal uptake of homocysteine occurs (5).
To date, the pathways of methionine-homocysteine metabolism are not fully understood, and further research is needed. Baseline data on renal function are mandatory to allow interpretation of hyperhomocysteinemia, which correlates with both renal and liver impairment.
Acknowledgments
We thank Claudia Blasius, Department of Clinical Biochemistry, University of Bonn for excellent technical assistance.
References
-cystathionase?. Scand J Gastroenterol 2000;35:866-872.[Medline]
[Order article via Infotrieve]
Centre de Recerca Biomèdica Institut de Recerca en Ciències de la Salut Hospital Universitari de Sant Joan C/. Sant Joan s/n 43201-Reus, Catalunya, Spain
bAddress correspondence to this author at: Centre de Recerca Biomèdica, Hospital Universitari de Sant Joan, C/. Sant Joan s/n, 43201-Reus, Catalunya, Spain. Fax 34-977-312569; e-mail jcamps{at}grupsgs.com.
To the Editor:
We read with interest the letter by Woitas et al. commenting our report (1). We agree with these authors in that renal function must be considered in the interpretation of hyperhomocysteinemia. We reported (2) that renal failure and tubular damage are main predictors of hyperhomocysteinemia in patients with the nephrotic syndrome. However, we do not think that renal impairment is a major determinant of plasma homocysteine alterations in the cirrhotic patients described in our study (1).
We measured plasma creatinine concentrations in 59 of our patients. Most of them (n = 50) were classified as Child–Pugh A and had well-preserved renal function with a mean (± SD) plasma creatinine of 80 (± 4) µmol/L and a relatively high mean plasma homocysteine concentration of 12.7 (± 3.2) µmol/L. Seven patients were classified as Child–Pugh B and presented with a similar plasma creatinine (82 ± 9 µmol/L) and, interestingly, lower homocysteine (9.3 ± 0.8 µmol/L). Only two patients were classified as Child–Pugh C, having impaired renal function (plasma creatinine, 159 ± 94 µmol/L) and a relatively low plasma homocysteine (9.1 ± 3.1 µmol/L).
Although our results may appear to be different from those of Woitas et al., we do not think they are. We think that our two cirrhotic populations are not comparable because the degree of severity of the disease was very different: most of their patients were decompensated (21 of 23 had ascites), and their Child–Pugh score was considerably higher than that of our patients (median score, 9 vs 6).
In our opinion, the two studies are complementary and may reflect a quite complex picture in which increases in plasma homocysteine concentrations (secondary to alcohol intake) are observed at the earliest stages of the disease, followed by normalization at intermediate stages (secondary perhaps to the loss of functional liver mass) and further increases (related to renal impairment) in the more severe groups of patients.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
