Safety aspects of pharmacological GH therapy in adults

The consequences of 'pharmacological' growth hormone administration have be en studied in a number of conditions, including those characterized by high rates of catabolism. The majority of studies have reported favourable effe cts on metabolism but recent reports indicate that GH treatment results in increased mortality in critically ill humans. The objective of the study wa s to assess the safety of large doses of rhGH therapy in human adults. Orig inal trials were identified by searching MEDLINE (1966-March 2000) and the Cochrane database (2000). References of all identified trials were also ins pected for more studies. All relevant trials in which GH had been administe red to non-GH-deficient (GHD) adult humans were selected from. Outcomes suc h as death, clinically significant change in function, change in length of hospital stay or need for treatment, and adverse effects were sought. Studi es were selected, quality-assessed and passed suitable for inclusion by two independent reviewers. Those studies that were placebo-controlled with sat isfactory randomization were considered for inclusion. Twenty-one reports w ere included in the review. A wide range of patient groups were studied by a variety of investigators, employing a range of doses and duration of GH t reatment. The study protocols differed markedly. The majority of studies we re small and were designed and/or powered to enable identification of speci fic effects on nutritional status, protein metabolism, level of function or quality of life. Only two studies were designed to assess safety issues an d mortality. In these, GH treatment was associated with a marked increase i n mortality in critically ill ICU patients, with a range of diagnoses. Mult i-organ failure and the effects of sepsis/infection accounted for most of t he excess mortality. In addition morbidity, in terms of length of ICU stay, was increased by GH administration. Other less marked effects were increas ed fluid retention and hyperglycaemia as a consequence of GH administration . Functional improvement following GH therapy was documented in some studie s. There have been few studies assessing the safety aspects of 'pharmacolog ical' GH treatment in adult humans. Two well-designed reports indicate that GH administration results in increased morbidity and mortality in a wide v ariety of critically ill subjects across a spectrum of age ranges. The mech anism(s) of the GH-associated mortality remain poorly understood. Based on current trial evidence, pharmacological GH treatment cannot be recommended for widespread use in critically ill subjects. Well-conducted and reported randomized trials are still needed to inform practice as to whether GH admi nistration will be safe in specific illness categories. (C) 2001 Harcourt P ublishers Ltd.