The effect of hormone replacement therapy on serum homocysteine levels in perimenopausal women: a randomized controlled trial

Serum homocysteine levels may be lowered by hormone replacement therapy, but randomized controlled trial data are scarce. We performed a single center randomized placebo-controlled trial to assess the 6 months effect of hormone replacement therapy compared with placebo on fasting serum homocysteine levels in 121 perimenopausal women free of cardiovascular disease, and recruited from the general population. The trial was double-blind with respect to a sequential combined regimen of oral 17 beta-estradiol and desogestrel (17 beta E(2)-D) and the placebo group and open with respect to a combination of conjugated equine estrogens and norgestrel (CEE-N). At baseline and after 6 months, fasting serum homocysteine levels were measured. Differences in 6 months serum homocysteine levels from baseline between treatment and placebo groups were calculated, and expressed as a percentage of the 6 months placebo level. After 6 months, the difference in serum homocysteine levels between women receiving 17 beta E(2)-D and placebo was -6.3% (95% CI, -12.4%; 0.0%, P=0.06). The difference between women receiving CEE-N and placebo was -10.1% (95% CI, -16.7%; -2.9%, P<0.01). The difference between the combined group of both types of hormone replacement therapy users and placebo was -7.8% (95% CI, -13.2%; -2.0%, P=0.01). No significant difference was observed between the two active regimens. Our results indicate that hormone replacement therapy decreases homocysteine levels in perimenopausal women.

Does iron-deficient erythropoiesis in pregnancy influence fetal iron supply?

BACKGROUND

It was investigated whether the iron status in newborns is negatively influenced by iron-deficient erythropoiesis of the mother during pregnancy.

METHODS

The iron status is characterized by hemoglobin, erythrocyte zinc protoporphyrin and serum ferritin values. Iron-deficient erythropoiesis of the mother is characterized by erythrocyte zinc protoporphyrin values. Measurement of erythrocyte zinc protoporphyrin was performed in 103 non-anemic females within 24 hours after delivery. The iron status of their newborns was measured in cord blood.

RESULTS

Erythrocyte zinc protoporphyrin concentrations were normal in 62% and elevated in 38% of the females, hence 39 women had iron-deficient erythropoiesis. There were no significant differences in mean values of the iron status parameters between neonates born to females with iron-deficient erythropoiesis and neonates born to females not having iron-deficient erythropoiesis. No correlation was found between maternal erythrocyte zinc protoporphyrin values and any of the neonatal parameters.

CONCLUSION

It is concluded that fetal iron supply is not negatively influenced by iron-deficient erythropoiesis in the mother.

Erythrocyte zinc protoporphyrin testing in pregnancy

BACKGROUND

To compare the usefulness of serum ferritin, hemoglobin and erythrocyte zinc protoporphyrin estimations as indicators of the need of iron supplementation in pregnancy.

METHODS

In 97 healthy pregnant women serum ferritin, hemoglobin and erythrocyte zinc protoporphyrin concentrations were measured at each trimester period. Iron supplementation was prescribed at hemoglobin values below 113 g/L.

RESULTS

Neither in the first nor in the second trimester differences in serum ferritin and erythrocyte zinc protoporphyrin values have been found between women who got iron therapy prescribed and women who did not get iron therapy, whereas already in the first trimester hemoglobin values were lower in the group that got iron therapy. These results indicate that hemodilution rather than iron deficiency determined which women got iron therapy. In the third trimester 14 out of 80 women without iron supplementation had depleted iron stores and a significant erythrocyte zinc protoporphyrin increase, indicating evolving iron deficiency anemia. However, only five of them showed hemoglobin levels below 113 g/L and got iron therapy. Low serum ferritin levels were found in 23% of the women at booking and in 92% in the third trimester. Serum ferritin levels neither predicted the development of iron deficient erythropoiesis nor of anemia.

CONCLUSIONS

Iron deficiency can neither be diagnosed nor predicted by serum ferritin or hemoglobin measurements at booking or later on in pregnancy. Erythrocyte zinc protoporphyrin measurements can be of help in determining which women have iron deficient erythropoiesis and may benefit by iron therapy.

Total homocysteine and cognitive decline in a community-based sample of elderly subjects: the Rotterdam Study

Homocysteine has been associated with an increased risk of cardiovascular disease. Cardiovascular diseases have been related to cognitive decline. The authors investigated the association of homocysteine with concurrent cognitive impairment and subsequent cognitive decline in a random sample of 702 community-dwelling respondents aged 55 years or over to the prospective Rotterdam Study in 1990-1994. Multiple logistic regression was used to calculate odds ratios and 95 percent confidence intervals for the association between total homocysteine levels and cognitive impairment (Mini-Mental State Examination (MMSE) score <26) and cognitive decline (drop in MMSE score of >1 point/year). Mean duration of follow-up was 2.7 years. After adjustment for age, sex, and education, there was no relation between total homocysteine and cognitive impairment (highest vs. lowest tertile: odds ratio (OR) = 1.30, 95% confidence interval (CI): 0.50, 3.38) or cognitive decline (middle vs. lowest tertile: OR = 1.14, 95% CI: 0.67, 1.93; highest vs. lowest tertile: OR = 0.91, 95% CI: 0.52, 1.58). Subjects who were lost to follow-up due to death or nonresponse had slightly higher age-adjusted homocysteine levels and lower MMSE scores at baseline. Sensitivity analyses showed that selective loss to follow-up was not a likely explanation for the absence of an association in the participants. Although a relation between homocysteine and reduced cognitive function is biologically plausible, this study suggests no such association in a community-based sample of the elderly.

Serum ferritin and risk of myocardial infarction in the elderly: the Rotterdam Study

BACKGROUND

Elevated body iron stores have been suggested to be a risk factor for ischemic heart disease.

OBJECTIVE

We examined whether elevated serum ferritin concentrations, other indicators of iron status, and dietary iron affected the incidence of myocardial infarction (MI) in an elderly population.

DESIGN

A nested, case-control study of 60 patients who had their first MI and 112 age- and sex-matched control subjects embedded in the population-based cohort of the Rotterdam Study.

RESULTS

The age- and sex-adjusted risk of MI for subjects with serum ferritin concentrations > or = 200 microg/L was 1.82 (95% CI: 0.90, 3.69; P = 0.096). The odds ratio (OR) was 1.26 (95% CI: 0.98, 1.64; P = 0.078) for the highest tertile of serum ferritin and was only slightly altered in a multivariate model. Risk of MI associated with the highest tertile of ferritin was most evident in current or former smokers (OR: 1.68; 95% CI: 1.17, 2.47; P for trend = 0.008) and in subjects with hypercholesterolemia (OR: 1.43; 95% CI: 0.99, 2.11; P for trend = 0.056) or diabetes (OR: 2.41; 95% CI: 1.12, 7.67; P for trend = 0.027). No association with risk of MI was observed for tertiles of serum iron, serum transferrin, or total dietary iron. For dietary heme iron, risk of MI was significantly increased in a multivariate model in which dietary energy, fat, saturated fat, and cholesterol were adjusted for (OR: 4.01; 95% CI: 1.17, 15.87; P for trend = 0.031).

CONCLUSION

In the presence of other risk factors, serum ferritin may adversely affect ischemic heart disease risk in the elderly.

Optimization of apolipoprotein(a) genotyping with pulsed field gel electrophoresis

BACKGROUND

Increased lipoprotein(a) is a risk factor for atherosclerosis, and its concentration in serum is inversely correlated with the size of the apoliprotein(a) [apo(a)] component. The size of the apo(a) gene is determined mainly by the Kringle IV size polymorphism. We have optimized and characterized pulsed field gel electrophoresis (PFGE) for apo(a) genotyping.

METHODS

Established PFGE protocols were adjusted. The changes included the following: (a) increased DNA yields by the use of all leukocytes for isolation from either 3 mL of fresh EDTA whole blood or 250 microL of frozen buffy coats; (b) increased efficiency of Kpn1 digestion by the inclusion of a digestion buffer wash; (c) reduction of assay time by the use of capillary blotting; (d) increased sensitivity by the use of four digoxigenin-labeled apo(a) probes; and (e) identification using a single film by the inclusion of a digoxigenin-labeled lambda marker probe in addition to apo(a) probes in the hybridization mix.

RESULTS

In older Caucasians, 93% (buffy coats, n=468) were heterozygous for apo(a) gene size. An inverse correlation between serum lipoprotein(a) and the sum of Kringle IV alleles was found (y = -23x + 1553; r = -0.442; n = 468). Gel-to-gel variation was minimal (3%). Imprecision (SD) was one Kringle IV repeat (control sample containing eight fragments of 72-233 kb; n=34 electrophoretic runs).

CONCLUSIONS

The practicality and sensitivity of the apo(a) genotyping technique by PFGE were improved, and accuracy and reproducibility were preserved. The optimized procedure is promising for apo(a) genotyping on frozen buffy coats from large epidemiological studies.

Serum caeruloplasmin as a coronary risk factor in the elderly: the Rotterdam Study

Serum Cu and caeruloplasmin levels have been suggested to be independent risk factors for CHD operating through oxidative modification of LDL. However, given its function as an acute-phase protein, the question has been raised whether an elevated caeruloplasmin level is not merely an indicator of inflammation. In the current study, we investigated whether serum caeruloplasmin was associated with subsequent myocardial infarction, taking into account indices of inflammation. The study population consisted of 210 cases of first myocardial infarction and controls, frequency-matched on age (5-year categories) and sex, selected from the population-based cohort of the Rotterdam Study. Serum caeruloplasmin levels were significantly elevated in cases of myocardial infarction compared with controls (510 (SD 110) v. 470 (SD 100) mg/l; P = 0.007). Risk of myocardial infarction for the highest compared with the lowest quartile of caeruloplasmin was 2.46 (95% CI 1.04, 6.00; Ptrend = 0.043) after adjustment for age, sex, BMI, pack-years smoked, serum cholesterol, systolic blood pressure, and income. The relative risk was most evident in current smokers. Adjustment for C-reactive protein and leucocyte count reduced the excess risk by 33%. This suggests that a substantial part of the observed association between serum caeruloplasmin and CHD may be attributed to inflammation processes rather than to the pro-oxidant activity of caeruloplasmin.

Influence of iron deficiency anaemia on haemoglobin A2 levels: possible consequences for beta-thalassaemia screening

Iron deficiency modulates the synthesis of HbA2, resulting in reduced HbA2 levels in patients with iron deficiency anaemia. The diagnosis heterozygous beta-thalassaemia is based on a raised HbA2 level. Patients with beta-thalassaemia and concomitant iron deficiency can show normal HbA2 levels. It is of clinical importance to know the quantitative effect of iron-deficient erythropoiesis on the levels of HbA2 in order to be able to determine which iron-deficient patients with normal HbA2 levels have to be retested after iron therapy in thalassaemia screening programmes. In this study, HbA2 levels in 150 patients with iron-deficiency anaemia and 71 healthy controls have been measured. A linear correlation is found in the patient group between HbA2 and Hb, HbA2 and MCV, and HbA2 and erythrocyte zinc protoporphyrin (ZPP). In future studies, the correlation between HbA2 and erythrocyte parameters in patients with heterozygous beta-thalassaemia and concomitant iron deficiency has to be examined. We recommend that ZPP be measured in these studies too, as ZPP levels may be a better indicator of concomitant iron deficiency than Hb or MCV in thalassaemic patients.

Homocysteine and short-term risk of myocardial infarction and stroke in the elderly: the Rotterdam Study

BACKGROUND

Elevated homocysteine level increases vascular disease risk. Most data are based on subjects younger than 60 years; data for the elderly are more limited. We examined the relationship of homocysteine level to incident myocardial infarction and stroke among older subjects in a nested case-control study.

METHODS

Subjects were participants in the Rotterdam Study, a cohort study among 7983 subjects residing in the Ommoord district of Rotterdam, the Netherlands. Baseline examinations were performed from March 1, 1990, to July 31, 1993. The analysis is restricted to myocardial infarction and stroke that occurred before December 31, 1994. One hundred four patients with a myocardial infarction and 120 with a stroke were identified with complete data. Control subjects consisted of a sample of 533 subjects drawn from the study base, free of myocardial infarction and stroke. Nonfasting total homocysteine levels were measured.

RESULTS

Results were adjusted for age and sex. The risk of stroke and myocardial infarction increased directly with total homocysteine. The linear coefficient suggested a risk increase by 6% to 7% for every 1-micromol/L increase in total homocysteine. The risk by quintiles of total homocysteine level was significantly increased only in the group with levels above 18.6 micromol/L (upper quintile): odds ratios were 2.43 (95% confidence interval, 1.11-5.35) for myocardial infarction and 2.53 (95% confidence interval, 1.19-5.35) for stroke. Associations were more pronounced among those with hypertension.

CONCLUSIONS

The present study, based on a relatively short follow-up period, provides evidence that among elderly subjects an elevated homocysteine level is associated with an increased risk of cardiovascular disease.

Portable blood gas and electrolyte analyzer evaluated in a multiinstitutional study

A recently introduced blood gas/electrolyte analyzer (SenDx 100((R)), renamed ABL70) intended for point-of-care, near-patient, or stat laboratory use was evaluated simultaneously in four different institutions and compared with three different laboratory bench analyzers with respect to imprecision, inaccuracy (assessed by tonometry), and patient-sample analyses. The analyzer is equipped with a sensor cassette and a reagent cartridge for 50, 100, or 200 analyses and 100 or more traditional quality-control measurements. One analysis requires 170 microL of whole blood and takes <90 s. Statistically, the instrument performed somewhat better (lower CVs) for PO2 and potassium and somewhat worse for pH, PCO2, and ionized calcium than the respective comparison analyzers. However, the overall performance (in terms of CV and accuracy) was satisfactory in terms of clinical (e.g., CLIA '88) goals in all institutions. The mean difference and the CV of that difference in some 400 patient-sample comparisons were as follows: 0.010 (+/- 0.002%) for pH, -0.65 mmHg (+/- 4%) for PCO2, -0.49 mmHg (+/- 6%) for Po2, 0.44 mmol/L (+/- 1.2%) for sodium, -0.013 mmol/L (+/- 2.9%) for potassium, -0.016 mmol/L (+/- 2.6%) for ionized calcium, and -0.016 L/L (+/- 7. 1%) for the hematocrit. Its acceptable analytical performance and ease of operation make the SenDx 100 suitable for the analysis of blood gases and electrolytes.

Combined use of erythrocyte zinc protoporphyrin and mean corpuscular volume in differentiation of thalassemia from iron deficiency anemia

In a retrospective study the diagnostic value of erythrocyte zinc protoporphyrin (ZPP) measurement as a means of distinguishing iron deficiency anemia from thalassemia syndromes in patients with microcytosis was explored. ZPP values were increased in all patients with iron deficiency and in part of the patients with thalassemia. The combined measurement of erythrocyte mean corpuscular volume (MCV) and ZPP resulted in a correct classification of patients with iron deficiency and with thalassemia in more than 95%. The predictive value of this method is better than the results obtained by using formulae derived from red cell indices. In population screening programs for thalassemia syndromes, in which MCV determination is used as the initial test, the ZPP test is recommended as a second test, in order to discriminate between patients with microcytosis due to iron deficiency and patients with microcytosis due to thalassemia syndromes.

Zinc protoporphyrin as screening test in female blood donors

Erythrocyte zinc protoporphyrin (ZPP) was measured in 102 women blood donors to evaluate its usefulness in screening for evolving iron deficiency anemia, a reason for the deferral of donors. The results were compared with serum ferritin determinations. Five women were deferred before their first donation and eight women were deferred after one or two donations. Women with increased ZPP values all had low serum ferritin concentrations, indicating iron-deficient erythropoiesis that was caused by iron depletion. The positive predictive value of an increased ZPP in predicting deferral of the donor after one or two donations was 75%, whereas a serum ferritin concentration < or = 12 microg/L predicted deferral in 26% of the donors. The results indicate that the ZPP test can be recommended as a feasible and inexpensive predonation test to determine a subset of donors with iron-deficient erythropoiesis at risk of developing iron deficiency anemia.

Methionine synthase deficiency without megaloblastic anaemia

We report findings on a child presenting with neonatal homocystinuria, hypomethioninaemia and severe neurological symptoms, including developmental delay and seizures. Methylmalonic aciduria was not present. The activity of methionine synthase in fibroblasts was severely deficient and formation of methylcobalamin from 57Co labelled cyanocobalamin was very low. The patients cells complemented with those of a cblE patient but not with those of two cblG patients. No biochemical or clinical response to injections of hydroxycobalamin was found. Both off treatment and on betaine and methionine supplementation the patient, at age 8 years, has not developed megaloblastic anaemia. In addition, the patient is homozygous for the C677T polymorphism in the 5,10 methylenetetrahydrofolate reductase (MTHFR) gene and the concomitant existence of this mutation with the methionine synthase defect may prevent folate <<trapping>> and thus anaemia.

CONCLUSION

We report the lack of megaloblastic anaemia in a patient with severe methionine synthase deficiency who is also homozygous for C677T in MTHFR, hypothesize that the MTHFR polymorphism protects the patient against anaemia and speculate that homozygosity for MTHFR C677T could cause the dissociation between haematological and neurological disease seen in some patients with vitamin B12 deficiency.

Modulation of lipoprotein(a) atherogenicity by high density lipoprotein cholesterol levels in middle-aged men with symptomatic coronary artery disease and normal to moderately elevated serum cholesterol. Regression Growth Evaluation Statin Study (REGRESS) Study Group

OBJECTIVES

This study sought to examine whether lipoprotein(a) levels predict coronary artery lumen changes in patients with symptomatic coronary artery disease (CAD) and normal to moderate hypercholesterolemia.

BACKGROUND

Recent conflicting reports have confirmed or refuted the association of lipoprotein(a) with clinical events or angiographically verified disease progression.

METHODS

The association between serum lipoprotein(a) and changes in coronary artery lumen was studied in 704 men entered into the Regression Growth Evaluation Statin Study (REGRESS), a double-blind, placebo-controlled, quantitative angiographic study that assessed the effect of 2 years of pravastatin treatment. The primary end points were changes in average mean segment diameter (MSD) and average minimal obstruction diameter (MOD). Pravastatin- and placebo-treated patients were classified as having progressing, regressing or stable CAD, and median lipoprotein(a) concentrations were compared. Bivariate and multivariate regression analyses were performed in the overall patient group and in high risk subgroups.

RESULTS

Pravastatin treatment did not affect serum apolipoprotein(a) levels. Median in-trial (sampled at 24 months) apolipoprotein(a) levels for regressing, stable and progressing CAD were, respectively, 130, 162 and 251 U/liter in placebo-treated patients and 143, 224 and 306 U/liter in pravastatin-treated patients. Predictors of MSD and MOD changes were baseline MSD and MOD, in-trial apolipoprotein(a), in-trial high density lipoprotein (HDL) cholesterol and baseline use of long-acting nitrates. The multivariate models predicted 14% of MSD changes and 12% of MOD changes; apolipoprotein(a) predicted only 2.6% and 4.8%, respectively. However, in patients with in-trial HDL cholesterol levels <0.7 mmol/liter, apolipoprotein(a) predicted up to 37% of the arteriographic changes.

CONCLUSIONS

Serum lipoprotein(a) levels predict coronary artery lumen changes in normal to moderately hypercholesterolemic white men with CAD; its atherogenicity is marked in the presence of concomitant hypoalphalipoproteinemia.

Folate-responsive homocystinuria and megaloblastic anaemia in a female patient with functional methionine synthase deficiency (cblE disease)

This first detailed report of a female patient with functional methionine synthase deficiency due to the cblE defect describes treatment with several vitamins and cofactors and clinical progress for 17 years. Before treatment, major findings were microcephaly, psychomotor retardation, episodic reduced consciousness, megaloblastic anaemia, increased plasma free homocystine (> 20 mumol/L), low plasma methionine (< 10 mumol/L) and increased excretion of formiminoglutamate. On high-dose folic acid, biochemical abnormalities such as formiminoglutamate excretion and homocystinuria nearly normalized, but clinical and haematological abnormalities remained. On replacement of folate with methylcobalamin, alertness, motor function, speech and the electroencephalogram improved, biochemical features were similar, but the mean corpuscular volume increased. The best control was observed on a combination of folate and methylcobalamin. At 17 years of age she remains severely mentally retarded. In cultured fibroblasts methionine synthesis was reduced (0.03 nmol/mg/per 16 h, controls 2.4-6.9); methionine synthase activity was normal under high reducing conditions but decreased on limiting the reducing agent, dithiothreitol, to 5 mmol/L (18% of total, controls 51-81%); formation of methylcobalamin was low (4.5% of total cobalamins, control 57.5%) and complementation studies indicated the cblE defect. Methionine formation showed only minor increases in cells grown in folate- or cobalamin-supplemented medium. Serine synthesis, which was low in normal medium, increased with cobalamin supplementation. These studies suggest further heterogeneity within cblE mutants, show the difficulty of establishing the enzyme defect in vitro, and indicate a role for folate in addition to cobalamin in treatment.

Thermolabile methylenetetrahydrofolate reductase in coronary artery disease

BACKGROUND

Hyperhomocysteinemia, an independent and graded risk factor for coronary artery disease (CAD), may result from both environmental and hereditary factors. Methylenetetrahydrofolate reductase (MTHFR) catalyzes the conversion of methylenetetrahydrofolate to methyltetrahydrofolate, the methyl donor in the remethylation of homocysteine to methionine. A 677C-->T mutation in the MTHFR gene has been associated with elevated homocysteine concentrations in homozygous (+/+) individuals.

METHODS AND RESULTS

We assessed the frequency of this common mutation in 735 CAD patients from the Regression Growth Evaluation Statin Study (REGRESS), a lipid-lowering coronary-regression trial, and in 1250 population-based control subjects. Furthermore, the association between the mutation and serum homocysteine concentrations was studied. The frequency of the homozygous (+/+) mutation was 9.5% among patients versus 8.5% among control subjects, resulting in an odds ratio of 1.21 (95% confidence interval [CI], 0.87 to 1.68), relative to the (-/-) genotype. Homocysteine concentrations were significantly elevated in both (+/+) and (+/-) individuals compared with (-/-) individuals (median homocysteine levels, 15.4, 13.4, and 12.6 micromol/L, for (+/+), (+/-), and (-/-) individuals, respectively). For a summary estimation of the risk of the (+/+) genotype for CAD, we performed a meta-analysis on 8 different case-control studies on thermolabile MTHFR in CAD. In the meta-analysis, the homozygous (+/+) genotype was present in 299 of 2476 patients (12.1%) and in 257 (10.4%) of 2481 control subjects, resulting in a significant odds ratio of 1.22 (95% CI, 1.01 to 1.47) relative to the (-/-) genotype.

CONCLUSIONS

Both the homozygous (+/+) and heterozygous (+/-) genotype result in elevated homocysteine concentrations. From our meta-analysis, we conclude that the homozygous (+/+) genotype is a modest but significant risk factor for CAD.

Significance of various parameters derived from biological variability of lipoprotein(a), homocysteine, cysteine, and total antioxidant status

Analytical and biological components of variability and various derived indices have been determined for lipoprotein(a) [Lp(a)], homocysteine (Hcy), cysteine (Cys), and total antioxidant status (TAOS) in ostensibly healthy adult Caucasians and in stable outpatients with an increased serum Lp(a). In healthy Caucasians, average intraindividual biological CVs (CVb) were 20.0% for Lp(a), 9.4% for Hcy, 5.9% for Cys, and 2.8% for TAOS, CVbs being similar in men and women. In the outpatient group, CVbs were comparable for Hcy, Cys, and TAOS, but significantly lower for Lp(a) (7.5% vs 20.0%; P <0.0001). Moreover, a significant inverse relation between both biological and analytical CVs (CVa) and serum Lp(a) concentrations was demonstrated. We conclude that average CVa and CVb values, and hence average derived indices, are adequate for Hcy, Cys, and TAOS, whereas individual values should be used for Lp(a).

Homocysteine, atherosclerosis and prevalent cardiovascular disease in the elderly: The Rotterdam Study

OBJECTIVES

Elevated homocysteine increases the risk of vascular disease, in particular amongst younger subjects (< 60 years). Very few studies have been performed amongst older subjects. We evaluated the relation of plasma total homocysteine (tHcy) to atherosclerosis and symptomatic cardiovascular disease amongst older men and women.

DESIGN

A cross-sectional study.

SETTING

General population.

SUBJECTS

A random sample of 630 men and women, participating in the Rotterdam Study, a prospective population-based cohort study amongst 7983 subjects aged 55 years and over residing in the Ommoord district of Rotterdam, the Netherlands.

MAIN OUTCOME MEASURES

Carotid atherosclerosis (carotid plaques and common carotid intima-media thickness) assessed by ultrasonography; lower extremity (peripheral) artery atherosclerosis measured by the ratio of the ankle to arm systolic blood pressure; prevalent cardiovascular disease assessed as a history of myocardial infarction or stroke.

RESULTS

Subjects, 55-74 years of age, with elevated tHcy levels (+/- 18.6 mumol L-1) had a thicker common carotid intima-media (difference 0.037 mm; 95% CI 0.001, 0.073), a lower ankle-arm index (-0.054; -0.104, -0.004), and an increased risk of cardiovascular disease (odds ratio 3.0; 1.5, 6.1), after adjusting for sex and age. There was no appreciable association of tHcy levels to atherosclerosis and cardiovascular disease in subjects aged 75 years and older.

CONCLUSIONS

In subjects aged 55-74 years elevated tHcy is associated with an increased risk of atherosclerosis and cardiovascular disease. The lack of association in those aged > or = 75 years most probably reflect selective mortality.

Serum LP(a) levels in African aboriginal Pygmies and Bantus, compared with Caucasian and Asian population samples

Serum lipoprotein(a) (Lp(a)) and its correlates were studied in African Aboriginal Pygmies (n = 146) and Bantus (n = 208) from Cameroon. Geometric mean Lp(a) levels were 274 and 289 mg/l in Bantu males and females, respectively, and 220 and 299 mg/l in Pygmy males and females, the gender difference being significant in Pygmies (p = 0.024). In Pygmies 41% and 52% of the males and females, respectively, had Lp(a) levels above 300 mg/l, compared with 47% and 55% in Bantus. Overall, Lp(a) levels did not significantly differ between Pygmies and Bantus, and did not correlate with age, body mass index (BMI), systolic and diastolic blood pressure. Compared with healthy Asian and Caucasian population samples, age- and BMI-adjusted geometric Lp(a) means were 2.3- to 5.0-fold higher in Pygmy and Bantu males, and 2.9- to 3.6-fold higher in Pygmy and Bantu females (p < or = 0.05). Across the population samples studied ethnicity predicted 12% and 17% of serum Lp(a) variance in males and females, respectively.

Chromogranin A as serum marker for neuroendocrine neoplasia: comparison with neuron-specific enolase and the alpha-subunit of glycoprotein hormones

Chromogranin A (CgA) is gaining acceptance as a serum marker of neuroendocrine tumors. Its specificity in differentiating between neuroendocrine and nonneuroendocrine tumors, its sensitivity to detect small tumors, and its clinical value, compared with other neuroendocrine markers, have not clearly been defined, however. The objectives of this study were to evaluate the clinical usefulness of CgA as neuroendocrine serum marker. Serum levels of CgA, neuron-specific enolase (NSE), and the alpha-subunit of glycoprotein hormones (alpha-SU) were determined in 211 patients with neuroendocrine tumors and 180 control subjects with nonendocrine tumors. The concentrations of CgA, NSE, and alpha-SU were elevated in 50%, 43%, and 24% of patients with neuroendocrine tumors, respectively. Serum CgA was most frequently increased in subjects with gastrinomas (100%), pheochromocytomas (89%), carcinoid tumors (80%), nonfunctioning tumors of the endocrine pancreas (69%), and medullary thyroid carcinomas (50%). The highest levels were observed in subjects with carcinoid tumors. NSE was most frequently elevated in patients with small cell lung carcinoma (74%), and alpha-SU was most frequently elevated in patients with carcinoid tumors (39%). Most subjects with elevated alpha-SU levels also had elevated CgA concentrations. A significant positive relationship was demonstrated between the tumor load and serum CgA levels (P < 0.01, by chi 2 test). Elevated concentrations of CgA, NSE, and alpha-SU were present in, respectively, 7%, 35%, and 15% of control subjects. Markedly elevated serum levels of CgA, exceeding 300 micrograms/L, were observed in only 2% of control patients (n = 3) compared to 40% of patients with neuroendocrine tumors (n = 76). We conclude that CgA is the best general neuroendocrine serum marker available. It has the highest specificity for the detection of neuroendocrine tumors compared to the other neuroendocrine markers, NSE and alpha-SU. Elevated levels are strongly correlated with tumor volume; therefore, small tumors may go undetected. Although its specificity cannot compete with that of the specific hormonal secretion products of most neuroendocrine tumors, it can have useful clinical applications in subjects with neuroendocrine tumors for whom either no marker is available or the marker is inconvenient for routine clinical use.

A computer program for constructing multivariate reference models

From a statistical point of view the simultaneous interpretation of multiple variables should be performed with a multivariate reference model rather than with multiple univariate reference intervals. A computer program for constructing and testing multivariate reference models is described. The use of the computer program is illustrated with a data set of total serum calcium concentrations and serum albumin concentrations from 222 2nd year medical students. Using a single univariate reference interval for total serum calcium, 17 students were classified as having an abnormal calcemic status while using a bivariate reference model for total serum calcium and serum albumin, 13 of these 17 students had in fact normal total serum calcium concentrations, taking into account their serum albumin concentrations.