The natural history and appropriate treatment of giant liver haemangioma remain poorly defined. The diagnostic strategy and the results of a longitudinal study of 78 such lesions are described. Sixteen resected patients and 62 with asymptomatic giant haemangiomas were entered into a follow-up programme consisting of clinical assessment and abdominal ultrasonography every 6 months. Resection was considered only for symptomatic patients (14 cases) and rapidly growing lesions (two cases). The mean follow-up was 36 months for the resected patients and 55 months for the observed group. Surgery permanently relieved symptoms. No recurrence of haemangioma was observed. Some 32 of 36 unresected lesions followed up remained stable in size. Minor changes were observed in four of 36. None ruptured or became symptomatic. It is concluded that asymptomatic large haemangiomas can be managed safely by observation. However, the occasional occurrence of rapid growth might represent a further indication for resection and justifies strict imaging follow-up. The pattern of growth, rather than absolute size, of a lesion is suggested for selection of asymptomatic patients who might benefit from preventive surgical excision. When feasible, the authors prefer enucleation to remove giant haemangiomas.

1. The rate of conjugation of benzoic acid with glycine was measured in the homogenates of 110 specimens of human liver and in 67 specimens of human renal cortex. 2. The assay for the formation of benzoyl glycine consisted of measuring the formation of benzoyl glycine from (14C) benzoic acid and glycine in the presence of coenzyme A and ATP. 3. In human liver, the mean (+/- SD) and coefficient of variation for the formation rate of benzoyl glycine were 254 +/- 90.5 nmol min-1 per g liver and 36%, respectively. There was a weak, but significant, negative correlation (r = -0.339, p < 0.001) between the rate of formation of benzoyl glycine and the liver donor's age. 4. In the human kidney, the rate of benzoyl glycine formation was normally distributed. The mean (+/- SD) and coefficient of variation were 321 +/- 99.3 nmol min-1 per g kidney and 31%, respectively. 5. These in vitro results are consistent with the view that the in vivo rate of conjugation of carboxylic acid with glycine varies among subjects and is normally distributed.

No abstract