Peter McNair

Prevention of early postmenopausal bone loss: controlled 2-year study in 315 normal females.

Abstract. With the aim of preventing postmenopausal bone loss, a placebo‐controlled double‐blind trial of 2 years duration was performed. We randomized 315 healthy volunteers in their early natural menopause to seven treatment and three placebo groups: 17β‐oestra‐diol, oestriol and sequential norethisteron (hormones); bendroflumethiazide 5 mg/day (thiazide); hormones and thiazide; sodium fluoride 20 mg/day; vitamin D3 2000 IU/day (D3); fluoride and D3; and lα (OH) vitamin D3 0–25 μg/day (lαD3). All participants were given daily calcium supplement of 500 mg. Every 3 months we measured the bone mineral content (BMC) of both forearms by photon absorptiometry and chemical quantities in blood and 48 h urinary collections. The study was completed by 264 (84%).

Renal hypomagnesaemia in human diabetes mellitus: its relation to glucose homeostasis.

Abstract. Interrelations between magnesium and glucose metabolism were studied in 215 insulin‐treated diabetic out‐patients aged 7–70 years. All had normal serum creatinine concentrations (below 115 μmol/l) and none had other diseases or received drugs known to interfere with mineral metabolism. A definite hypomagnesaemia (< normal mean —2 SD) and hypermagnesiuria (> normal mean + 2 SD) occurred in 38.6% and 55% of the patients. In the presence of hypermagnesiuria the serum magnesium concentration was inversely correlated to the urinary magnesium excretion rate (R=–0.23, P<0.02). Serum magnesium correlated inversely with both fasting blood glucose (R=—0.32, P<0.001) and the urinary glucose excretion rate (R=–0.22, P<0.005). The urinary magnesium excretion rate correlated directly with the same variables (R= 0.27, P<0.001 and R= 0.58, P<0.001, respectively). These data indicate that the net tubular reabsorption of magnesium is decreased in diabetic patients in presence of hyperglycaemia, leading to hypermagnesiuria and hypomagnesaemia.

Thiazide for the postponement of postmenopausal bone loss

The effect of thiazide on bone mineral loss in normal postmenopausal women was examined during a 3 yr placebo-controlled clinical trial. Sixty-three healthy women in their early menopause were randomized to treatment with bendroflumethiazide 5 mg/day or placebo for 2 yr, while both groups received placebo for the third year of the trial. Calcium supplement 0.5 g/day was given throughout the 36 mo to all participants. Bone mineral content (BMC) determined by 125I-photon absorptiometry of the forearms decreased 2% per year in the placebo group (p less than 0.001). In the thiazide group no fall in BMC was seen during the first 6 mo. whereafter BMC declined with the same rats as in the placebo group. At the end of the 3 yr trial BMC averaged 94.1% in the placebo group and 95.2% in the thiazide group (p greater than 0.05). Despite a daily supplement of 0.5 g calcium, thiazide induced a persistent fall in the urinary calcium excretion of 25% (p less than 0.001), whereas the calcium supplement in the placebo group caused a significant increase in mean urine calcium of 10%-20% (p less than .001). At stop of thiazide medication a rebound effect caused a marked rise in urine calcium. One month after withdrawal of the calcium supplement the urinary calcium excretion had returned to the initial level in both groups. It is concluded that despite a sustained urine calcium lowering action the effect of thiazide upon postmenopausal bone loss is shortlived.

Development of bone mineral loss in insulin‐treated diabetes: a 1 1/2 years follow‐up study in sixty patients

Abstract. The change in bone mass during 1 1/2 years was determined in a longitudinal study of sixty adult insulin‐treated diabetic out‐patients. During the study period the mean bone mass decreased by 1.30 ± 0.28 (SEM) % (P < 0.001), to a mean value of 91.0 ± 1.7% of normal (P < 0.001). The rate of bone loss was significantly higher in patients with 1–6 years of diabetes (n= 29, bone loss: 1.96 ± 0.32%/1 1/2 years) than in patients with longer duration of the disease (n= 31, bone loss: 0.61 ± 0.44%/l 1/2 years, P < 0.02). The endogenous insulin secretion estimated with the glucagonstimulated serum C‐peptide concentration decreased during the observation period in 58.6% of the patients with 1–6 years of diabetes compared to 16.1% among patients with 7–11 years of diabetes (P < 0.002). The rate of bone mineral loss was almost 3 times higher in the twenty‐two patients in whom endogenous insulin secretion had deteriorated (2.12 ± 0.30%/l 1/2 years) than in the thirty‐eight patients without as well as with unchanged or increased insulin secretion (0.78 ± 0.39%/l 1/2 years, P < 0.01). In twenty patients with an increased insulin dose during the study period the mean bone mineral loss was 2.05 ± 0.36%/l 1/2 years compared to a mean bone mineral loss of 0.91 ± 0.38%/l 1/2 years in the forty patients with unchanged or decreased insulin dosage (P < 0.05).

Increased parathyroid hormone as a consequence of changed complex binding of plasma calcium in morbid obesity

To evaluate whether changed plasma calcium binding might lead to a secondary increase of parathyroid hormone in morbid obesity, fasting measurements of serum ionized, ultrafiltrable and total calcium, calcium binding substances, and parathyroid hormone were undertaken in age- and sex-matched groups of obese (n = 44) and normal weight subjects (n = 52). The 24-hour urinary calcium excretion and clearance of creatine were also measured. Calcium binding to proteins was changed. Serum total proteins and protein-bound calcium did not differ, but serum albumin was decreased in obesity. Consequently, obese subjects did not reveal the normal dependency of protein-bound calcium upon albumin. Calcium binding to other substances was also changed. Serum phosphate and bicarbonate were decreased, while the concentrations of citrate, lactate, acetoacetate, 3-hydroxybutyrate, free fatty acids, and urate were all increased, leaving the total concentration of plasma complex-bound calcium unchanged. Nevertheless, these reciprocal changes increase the concentrations of less readily reabsorbable anions in the renal ultrafiltrate. The changed pattern of calcium binding in serum of the obese subjects may serve to explain our findings of increased urinary calcium excretion, lowering of serum ionized calcium and increased parathyroid hormone levels, changes being significantly correlated with degree of overweight.

Secondary hyperparathyroidism of morbid obesity regresses during weight reduction

In order to test the relation between obesity and the secondary hyperparathyroidism found in markedly overweight subjects, 24 morbidly obese patients were studied before and after a weight loss of 35.9 kg obtained by a nutritionally adequate, intermittent very-low-calorie diet. Overweight was reduced from 98 +/- 34% to 44 +/- 19%. Serum total calcium did not change, but serum ionized calcium (Ca2+) increased from 1.22 +/- 0.04 mmol/L to 1.25 +/- 0.04 mmol/L (P less than .001). A corresponding fall was observed in serum parathyroid hormone (s-PTH), which decreased from 47.2 +/- 21.7 pmol/L to 35.2 +/- 19.4 pmol/L (P = .01). The change of s-PTH was positively associated with the reduction of body weight (r = .50, P less than .05) and with the reduction of overweight (r = .55, P less than .01). Regarding calcium binding substances, serum albumin remained low. The initially lowered serum phosphate and bicarbonate both rose (P less than .001). Plasma lactate and plasma free fatty acids (FFAs) decreased (P less than .001). The study supports our hypothesis that the change profile of calcium complexing anions in obesity interferes with the tubular reabsorption of calcium, which in turn lowers serum Ca2+, thus promoting hyperparathyroidism. Along with weight loss, concentrations of calcium complexing anions returns towards normal values and the secondary hyperparathyroidism regresses.

Hyperzincuria in insulin treated diabetes mellitus—its relation to glucose homeostasis and insulin administration

In order to elucidate some pathogenic factors of diabetic hyperzincuria we studied 60 adult insulin treated diabetic out-patients (40 males and 20 females), all with normal serum creatinine concentrations and absence of proteinuria during a 24-h period. Diabetic males and females both had significantly (p less than 0.01) increased zinc excretion rates (1.14 +/- 0.06 (S.E.M.) mumol/mmol creatinine and 1.37 +/- 0.10 mumol/mmol creatinine) compared with normal males and females (0.55 +/- 0.06 and 0.48 +/- 0.08, respectively). The urinary zinc excretion rate correlated positively with the degree of glycosuria (r = 0.36, p less than 0.01), but was not associated with the duration of the disease. However, serum zinc levels gave no evidence of a state of zinc depletion in these patients. It was calculated that zinc originating from a diabetic bone loss and the exogenous insulin administration accounted for only a small part of the hyperzincuria. Compensatory hyperabsorption and/or increased zinc content in the diabetic diet may therefore serve to explain the lack of zinc depletion in the presence of hyperzincuria.

Vitamin D metabolites in diabetic patients: Decreased serum concentration of 24,25-dihydroxyvitamin D

In order to elucidate if changes in vitamin D metabolism play a role for diabetic bone loss, the serum concentrations of the major vitamin D metabolites were studied in 26 adult male ambulatory insulin-treated diabetics, selected to have normal renal function and a duration of diabetes below 11 years. The patients were studied during usual metabolic control and exhibited wide ranges of hyperglycaemia and glycosuria. The serum concentrations of the major metabolites of vitamin D, 25-hydroxyvitamin D(2 + 3) (25OHD), 24,25-dihydroxyvitamin D(2 + 3) (24,25(OH)2D), and 1,25-dihydroxyvitamin D(2 + 3) (1,25(OH)2D), were measured in diabetics, and in age and sex matched controls. The diabetics had slightly decreased serum levels of 25OHD (42.0 nmol/l versus 55.5 nmol/l in normals, P less than 0.05), markedly decreased serum levels of 24,25(OH)2D (2.98 nmol/l versus 5.91 nmol/l, P less than 0.01), but serum levels of 1,25(OH)2D were virtually normal (64.2 pmol/l versus 68.3 pmol/l, ns). The close correlation between serum concentrations of 25OHD and 24,25(OH)2D observed in the normal subjects, was absent in the diabetics. There were no correlations between the serum levels of any of the vitamin D metabolites and the measured indices of glucose and calcium metabolism. It is concluded that insulin-dependent diabetic patients demonstrate definite alterations in serum levels of vitamin D metabolites, the significance of which remains unknown at present.

Paper: On the quality of neural net classifiers

This paper describes several concepts and metrics that may be used to assess various aspects of the quality of neural net classifiers. Each concept describes a property that may be taken into account by both designers and users of neural net classifiers when assessing their utility. Besides metrics for assessment of the correctness of classifiers we also introduce metrics that address certain aspects of the misclassifications. We show the applicability of the introduced quality concepts for selection among several neural net classifiers in the domain of thyroid disorders.

The effect of magnesium on maternal blood pressure in pregnancy-induced hypertension A randomized double-blind placebo-controlled trial

The effects of magnesium were compared with those of placebo in a randomized double‐blind controlled study of 58 patients with pregnancy‐induced hypertension, of whom 27 received magnesium and 31 placebo. Twenty patients in each group were nulliparas. The treatment comprised 48 h of either intravenous magnesium or placebo infusion followed by daily oral magnesium or placebo tablets until one day after delivery. Magnesium supplementation significantly reduced maternal mean arterial blood pressure (MAP). The gestational age at delivery was the same in both groups, whereas the relative fetal birth weight among nulliparas was reduced in the placebo group. Unbalanced analyses of variance suggested an influence of magnesium supplement on birth weight. The infants in the magnesium supplemented group spent fewer days in the neonatal intensive care unit. There were no perinatal deaths. Magnesium appeared to be beneficial in the management of pregnancy‐induced hypertension. The better outcome associated with magnesium supplementation may not have been due to reduction of MAP and further studies are needed to clarify whether magnesium influences birth weight.