TO THE EDITOR:
Prompted by Garfinkel and Zisapel's comment [1] on our paper, published in August 1995 [2], we present some consecutive data. We postulated that elevated plasma melatonin levels and an altered 24-hour profile of the hormone we found in patients with liver cirrhosis [2] could represent an altered output of the circadian clock [3], causing a central disturbance of circadian rhythmicity.
Garfinkel and Zisapel suggested that we measure the urinary metabolite of melatonin, 6-sulfatoxymelatonin, to determine whether increased production in the pineal gland or decreased catabolism by the liver is the main reason for elevated serum melatonin levels in patients with liver cirrhosis.
We therefore measured urinary 6-sulfatoxymelatonin levels in 21 hospitalized cirrhotic patients with normal renal function (14 men and 7 women; median age, 50 years [range, 29 to 80 years]; Child class A, 7 patients; class B, 11 patients; class C, 3 patients) and in 9 healthy persons (3 men and 7 women; median age, 49 years [range, 32 to 69 years]). Sixteen had alcoholic liver disease, 2 had hepatitis C, and 1 each had primary biliary cirrhosis and Wilson disease. Two patients had signs of clinically overt hepatic encephalopathy.
Eight-hour urine excretion (10:00 p.m. to 6:00 a.m.) was assayed for 6-sulfatoxymelatonin by enzyme-linked immunosorbent assay (Melatonin-Sulfat-ELISA, IBL Ges mbH, Hamburg, Germany). Routine laboratory testing was performed by using standard methods. Statistical analysis was done with the paired Student t-test or the Mann-Whitney rank-sum test.
In the 8-hour urine measurements, cirrhotic patients had a significantly decreased concentration (mean ± SE, 19.01 ± 2.76 ng/mL compared with 39.2 ± 5.41; P = 0.001) and total excretion (median, 8.28 mg [range, 0.85 to 28.1 mg] compared with 12.21 mg [range, 9.12 to 29.04 mg]; P < 0.05) of 6-sulfatoxymelatonin, compared with controls (Figure 1). Urine volumes were similar in the two groups. No correlation to Child class or liver function measures was seen.