Prevalence of admission hyperglycaemia across clinical subtypes of acute stroke

Abstract

Admission hyperglycaemia is known to be an independent risk factor for poor outcome following stroke but the prevalence of hyperglycaemia across the clinical subtypes of acute stroke is unknown. Any novel stroke therapy designed to modulate plasma glucose in the acute phase should be applicable to most patients across all stroke subtypes. It has been suggested that admission hyperglycaemia is only present in more severe strokes, indicating a predetermined poor prognosis, and therefore that intervention to modulate plasma glucose in the acute phase would be unlikely to alter subsequent outcome. Over a 9-month period consecutive stroke patients were assessed upon hospital admission to evaluate clinical stroke subtype (total anterior circulation syndrome [TACS], partial anterior circulation syndrome [PACS], lacunar syndrome [LACS], or posterior circulation syndrome [POCS]) and admission plasma glucose concentration. Admission plasma glucose values were classified as above 6·0 mmol/L or above 6·9 mmol/L. 303 patients were assessed. There were no statistically significant differences between stroke subtypes in terms of age, sex, or prevalence of diabetes mellitus. Patients with the TACS subtype were more likely to have primary intracerebral haemorrhage (p<0·001, chi-squared test), have admission hyperglycaemia of greater than 6·0 mmol/L (p=0·002, chi-squared test), and have a higher mean admission plasma glucose (p=0·003, Mann Whitney U-test) than those with the non-TACS subtypes. Over 50% of patients within each subtype had hyperglycaemia of greater than 6·0 mmol/L on admission (figure). Admission hyperglycaemia is therefore common in acute stroke patients, exists across the range of clinical stroke subtypes, and although hyperglycaemia is significantly more common in those with more severe strokes, it is not restricted to these patients. If treatment to modulate plasma glucose is found to be effective, it is likely to be applicable to over two-thirds of acute stroke patients and across the range of clinical stroke subtypes. These findings provide The man complained of impotence and was receiving trazodone prescribed by a urologist. His other medications were quinapril, meclizine, and omeprazole. He did not smoke or drink alcohol. On examination, his blood pressure was 140/90 mm Hg. His respiratory, cardiovascular, and abdominal assessments were normal. He had bilaterally absent ankle jerks and bilateral decreased vibratory sensation in the feet up to the ankles. His initial series of laboratory investigations showed a marginally low haemoglobin and mean corpuscular volume. Blood urea nitrogen, serum creatinine, electrolytes, and liver function tests were normal, as well as urinalysis and thyroid profile. Because of persistent impotence, his urologist prescribed sildenafil. The patient reported improvement in sexual performance. At this time, we screened him to rule out sex-hormone abnormalities. His total testosterone was 150 ng/dL (normal range 260–990 ng/dL), free testosterone 0·80 ng/dL (normal range 1·60–3·30 ng/dL), and serum prolactin concentration was 80·4 ng/mL (normal <15 ng/mL). Serum follicle-stimulating hormone and luteinising hormone concentrations were within normal limits. A magnetic resonance scan of the pituitary fossa showed a 4 mm pituitary microadenoma. The patient was treated with bromocriptine. We came across another case of missed hyperprolactinaemia before sildenafil became available. A man aged 57 years had been diagnosed with diabetes mellitus 2 months previously. There were no complications of retinopathy, nephropathy, coronary artery disease, or peripheral vascular disease, but he complained of impotence. History showed he had been started on testosterone injections for low serum testosterone. Presently, he was using testosterone patches. Physical examination was unremarkable other than decreased pinprick sensation on the big toes. His medications included nifedipine, indapamide, probucol, and fluoxetine. We advised him to have appropriate laboratory tests done and prescribed 500 mg metformin twice daily because of high sugar concentrations. The laboratory results showed a serum prolactin concentration of 149·1 ng/mL (normal limit <15 ng/mL), with low testosterone, follicle-stimulating hormone, and luteinising hormone concentrations. However, a magnetic resonance scan showed no evidence of a pituitary tumour or other abnormality. The patient was started on bromocriptine 2·5 mg at night. His prolactin concentrations normalised and there was dramatic improvement in his sexual potency after starting bromocriptine. His testosterone concentrations, however, remained low and he required continued treatment with testosterone. These cases show the importance of systematic assessment for the cause of impotence before prescription of sildenafil, as suggested by Chan-Tack. Many patients with diabetes who also have sexual disorders are not fully investigated, since impotence is attributed to autonomic neuropathy. Diabetic neuropathy should lead to failure of erection only, with libido remaining intact. Loss of libido should lead to suspicion of a hormonal cause of impotence. History, physical examination of testicular volume and consistency, shape of penis, penile pulsations, and peripheral vascular pulsations, and detailed neurological assessment to rule out brain, spinal cord, or peripheral nerve lesions should be the first steps. Serum testosterone, gonadotropin, and prolactin concentrations should be measured to rule out renal, hepatic, and thyroid disorders.