Abstract
We have read with interest the article by Al-Ozairi et al. 1 in the March issue of the journal emphasizing the value and safety of prednisolone in the management of hyperemesis gravidarum. We recently cared for a woman in whom prednisolone failed to control vomiting because of drug intolerance but parenteral dexamethasone succeeded, even as an outpatient. Furthermore, her ketosis-prone type 1 diabetes mellitus was not aggravated by her dexamethasone therapy.
The 23-year-old primigravid patient with type 1 diabetes and controlled Grave's disease for nine years, managed with insulin aspart thrice daily, insulin glargine nocte and neomercazole, developed progressively severe vomiting from five weeks gestation. Despite a range of treatments including B vitamins, ondansetron, promethazine, prochlorperazine and intermittent trials of prednisolone, she lost 14% of her prepregnant body weight, reaching a nadir of 51 kg by 13 weeks. Four hospital admissions for rehydration, intravenous hydrocortisone and insulin infusion for diabetic ketoacidosis occurred over this period, due to a combination of hyperemesis and patient compliance issues with self-discharges. At 13 weeks as an ambulatory patient by her choice, a further attempt of oral prednisolone 25 mg twice daily was unsuccessful, leading to a trial of intramuscular dexamethasone 4 mg daily for four days followed by oral administration for a further week. This dramatically arrested her vomiting and allowed a tapering low dose of prednisolone (initially 10 mg daily) to be re-introduced, ceasing at 16 weeks with only residual occasional nausea. Her glycaemic control stabilized with HbA1c being 7.7% at 10 weeks and 7.4% at 20 weeks. Body weight progressively increased, albeit suboptimally, almost reaching her prepregnant weight by 29 weeks. She remained biochemically euthyroid throughout the pregnancy. A healthy female infant (3210 g) was delivered vaginally at 36 weeks after premature rupture of membranes.
Although there was a hesitancy to use dexamethasone, given its placental transfer with incomplete inactivation 2 and unknown teratogenic risk, it was demonstrated here to be very useful and apparently safe but it would be prudent to restrict its use to after eight weeks gestation when lip fusion has been completed. We speculate that its superior effect here may be due to its initial parenteral use, greater potency, prolonged action and possibly better blood–brain barrier penetration, given its likely medullary site of action demonstrated experimentally in cancer chemotherapy-induced emesis.