Assessment of cardiovascular risk--PROCAM and new algorithms]

The absolute risk for myocardial infarction is best determined using algorithms or point schemes which have been developed on the basis of large prospective, epidemiological studies. In Germany, for instance, the PROCAM score developed in Münster is leading in this field. The role of new risk factors lies in the optimum risk prediction in patients with intermediary or high risk for myocardial infarction. It is not necessary to determine additional risk factors in low-risk-patients. At present there is however no conclusive epidemiological evidence so that the decision for investigating newer risk factors lies with the physician in charge. So far the target level for LDL cholesterol in patients with intermediate risk was set at 130 mg/dL, whereas in high risk patients LDL cholesterol should not be more than 100 mg/dL. The results of newer studies suggest that it might be appropriate to define a new category of patients with very high risk in whom the LDL cholesterol value should not exceed 70 mg/dL. This inference must however be substantiated by further studies before it becomes part of the general recommendations for treatment.     

Pulmonary cryptococcosis in late pregnancy and review of published literature

Naturally occurring maternal immunosuppression increases the risk of infection by a variety of pathogens during pregnancy and the postpartum period. Pulmonary cryptococcosis during pregnancy is relatively rare. Here, we report on two cases of pulmonary cryptococcosis during pregnancy and puerperium. Both cases were successfully treated using oral fluconazole after parturition to avoid fetal toxicity. For the two patients, the placentas were checked and found to be pathologically normal, and the cryptococcal serum antigen in both infants was negative. Pulmonary cryptococcosis should be considered during differential diagnosis as a possible cause of abnormal chest shadow in pregnant patients.     

Essential hypertension and pregnancy.

Approximately 1% of pregnancies are complicated by essential hypertension. During pregnancy the blood pressure often stabilizes or improves. In patients with sustained hypertension, prospective controlled studies have demonstrated enhanced fetal survival when the blood pressure was controlled with antihypertensive medication. Such medication must be chosen carefully to avoid fetal and mateerial toxicity, and diuretics and salt restriction during pregnancy should be avoided. Among patients with essential hypertension the problem accelerates late in pregnancy in 2% to 11%; the acceleration may be predicted by determination of maternal mean arterial pressures and intravascular volumes early in pregnancy. The treatment of accelerated hypertension is identical to that of severe pre-eclampsia. Fetal loss is considerable but can be lessened by careful fetal and maternal monitoring and early controlled delivery. The risks of pregnancy in most patients with essential hypertension are small, and essential hypertension is not a uniform contraindication to pregnancy.     

Graves Hyperthyroidism and Pregnancy: A Clinical Update

ObjectiveTo provide a clinical update on Graves’ hyperthyroidism and pregnancy with a focus on treatment with antithyroid drugs.MethodsWe searched the English-language literature for studies published between 1929 and 2009 related to management of hyperthyroidism in pregnancy. In this review, we discuss differential diagnosis of hyperthyroidism, management, importance of early diagnosis, and importance of achieving proper control to avoid maternal and fetal complications.ResultsDiagnosing hyperthyroidism during pregnancy can be challenging because many of the signs and symptoms are similar to normal physiologic changes that occur in pregnancy. Patients with Graves disease require prompt treatment with antithyroid drugs and should undergo frequent monitoring for signs of fetal and maternal hyperthyroidism and hypothyroidism. Rates of maternal and perinatal complications are directly related to control of hyperthyroidism in the mother. Thyroid receptor antibodies should be assessed in all women with hyperthyroidism to help predict and reduce the risk of fetal or neonatal hyperthyroidism or hypothyroidism. The maternal thyroxine level should be kept in the upper third of the reference range or just above normal, using the lowest possible antithyroid drug dosage. Hyperthyroidism may recurin the postpartum period as Graves disease or postpartum thyroiditis; thus, it is prudent to evaluate thyroid function 6 weeks after delivery. Preconception counseling, a multidisciplinary approach to care, and patient education regarding potential maternal and fetal complications that can occur with different types of treatment are important.ConclusionPreconception counseling and a multifaceted approach to care by the endocrinologist and the obstetric team are imperative for a successful pregnancy in women with Graves hyperthyroidism. (Endocr Pract. 2010;16:118-129)     

Is there an embryopathy associated with first-trimester exposure to angiotensin-converting enzyme inhibitors and angiotensin receptor antagonists? A critical review of the evidence

Drugs that interfere with the renin-angiotensin system, such as angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs), are widely used to manage hypertension and heart failure. Adequate functioning of the RAS is essential for normal fetal kidney development. The potential for ACEIs and ARBs to impair fetal and neonatal renal function if taken after the first trimester of pregnancy has been well documented. Although these drugs were not found to be teratogenic in animals, until recently little was known about the teratogenic effects of ACEIs and ARBs in humans when exposure was limited to the first trimester of pregnancy. New evidence from epidemiologic studies indicates that there may be an elevated teratogenic risk when these drugs are taken during the first trimester of pregnancy. However, this elevated risk does not appear to be specific to ACEIs and ARBs, but is instead related to maternal factors and diseases that typically coexist with hypertension in pregnancy, such as diabetes, advanced maternal age, and obesity. Women who become pregnant while being treated with an ACEI or ARB should be advised to avoid exposure to these drugs during the second and third trimesters of pregnancy by switching to a different class of antihypertensive drugs between weeks 8 and 10 after conception. Birth Defects Research (Part A) 94:576-598, 2012. ? 2012 Wiley Periodicals, Inc.     

Fenofibrate Increases Serum Creatinine in a Patient with Familial Nephropathy Associated to Hyperuricemia

Background: Kidney function progressively deteriorates in patients with familial juvenile hyperuricemiac nephropathy (FJHN, OMIN 162000) and chronic renal disease is commonly associated to dyslipidemia. We report for the first time abrupt renal insufficiency in a patient with FJHN and hypertrygliceridemia following fenofibrate administration.Case report: A 53-year-old man was diagnosed clinically with FJHN at age 24 years which was subsequently confirmed by genotypic analysis of the UMOD gene at age 40 years. His mother and two brothers suffered the disease. At that time, renal size and function were normal, as was his blood pressure and serum lipids. At age 34 years, serum urate was 8.5 mg/dL and creatinine 1.7 mg/dL (GFR, 58 mL/min/1.73 m²). He was treated with allopurinol, losartan, and lovastatin. Serum TG levels ranged between 150 and 250 mg/dL. At age 52 years, serum urate was 4.1 mg/dL, creatinine 3.2 mg/dL, LDLc 99 mg/dL (atorvastatin 40 mg/d), and TG 275 mg/dL. Fenofibrate (160 mg/d) was added. One month later, serum creatinine increased to 4.2 mg/dL and TG decreased to 125 mg/dL. He did not complain of muscle pain, weakness, or changes in urinary frequency or color and rabdomyolysis was discarded. Fenofibrate was withheld and three months later serum creatinine decreased to baseline levels (3.2 mg/dL) and TG increased to 197 mg/dL.Conclusion: To our knowledge, this is the first patient with FJHN in whom fenofibrate administration was associated to a further impairment in renal function not attributable to rabdomyolysis.     

High-risk pregnancy and hypoluteoidism in the bitch

High-risk pregnancies are those in which the prevalence of maternal, fetal and/or perinatal morbidity or mortality is likely to be higher than that of the general obstetrical population. Some maternal characteristics associated with risk to maternal, fetal and/or perinatal health are readily identifiable prior to conception, such as advanced maternal age, brachycephalic breed, or a previous history of pregnancy loss. Others, such as gestational diabetes or a singleton litter, are recognized after conception. Early recognition of the problem (i.e. the risk), anticipation of the potential sequelae, and development of an aggressive management scheme are essential for a successful outcome of a high-risk pregnancy. A previous history of pregnancy loss is a high-risk factor for recurrence during subsequent pregnancies. Infection is a common cause. In some instances, recurrent pregnancy loss is associated with low serum concentrations of progesterone. Although the mechanism(s) by which this occurs is not fully understood, the situation has been called hypoluteoidism. Whatever the cause of the risks to pregnancy, the goals of managing high-risk pregnancies are to optimize maternal, fetal and perinatal health, so as to maintain maternal health throughout pregnancy and lactation and maximize the number of healthy pups surviving to weaning age.     

High-risk pregnancy and hypoluteoidism in the bitch

High-risk pregnancies are those in which the prevalence of maternal, fetal and/or perinatal morbidity or mortality is likely to be higher than that of the general obstetrical population. Some maternal characteristics associated with risk to maternal, fetal and/or perinatal health are readily identifiable prior to conception, such as advanced maternal age, brachycephalic breed, or a previous history of pregnancy loss. Others, such as gestational diabetes or a singleton litter, are recognized after conception. Early recognition of the problem (i.e. the risk), anticipation of the potential sequelae, and development of an aggressive management scheme are essential for a successful outcome of a high-risk pregnancy. A previous history of pregnancy loss is a high-risk factor for recurrence during subsequent pregnancies. Infection is a common cause. In some instances, recurrent pregnancy loss is associated with low serum concentrations of progesterone. Although the mechanism(s) by which this occurs is not fully understood, the situation has been called hypoluteoidism. Whatever the cause of the risks to pregnancy, the goals of managing high-risk pregnancies are to optimize maternal, fetal and perinatal health, so as to maintain maternal health throughout pregnancy and lactation and maximize the number of healthy pups surviving to weaning age.     

Pregnancy management in the bitch

Pregnancy management to optimize maternal and neonatal health begins with breeding management and the selection of normal, healthy brood stock in ideal body condition. After breeding, a commercial diet appropriate for reproduction and lactation should be fed. Typically these contain 29-32% protein of animal source, at least 18% fat, 20-30% carbohydrate, and essential vitamins, minerals and fatty acids. Pregnancy is confirmed approximately 25 d after breeding. A "maternity ward" and whelping box should be provided. Steady increases in caloric intake and body weight are expected as pregnancy progresses. Weight loss should not occur. Throughout pregnancy, changes in the bitch's attitude, activity, appetite, body weight, and physical findings should be monitored by the owner. If appetite and body weight do not continue to increase, or if any signs of illness develop, maternal health should be assessed with a complete physical examination and a CBC, biochemical profile, and free-catch urinalysis. Fetal health should be assessed with ultrasonography. Maternal or fetal abnormalities will put the pregnancy at risk. Impending parturition and the progress of labor and delivery can be monitored by assessing rectal temperature, serum concentrations of progesterone, and/or uterine and fetal monitors. This article reviews the physiology of canine pregnancy and parturition, and typical schemes used to manage normal canine pregnancy to optimize maternal and puppy health.     

Thalassaemia,iron, and pregnancy.

Haematological values of 35 pregnant women with beta-thalassaemia trait were followed during pregnancy. The discriminant function, calculated from haematological indices, was of no value in diagnosing beta-thalassaemia trait during pregnancy. Initially patients were given iron supplements only if the serum iron and total iron binding capacity levels indicated iron deficiency, but bone marrow biopsies performed in the first 22 patients at 32 weeks indicated deficient iron stores. These patients were therefore given iron irrespective of their serum iron level. All subsequent patients with beta-thalassaemia were also put on iron routinely at booking. Retrospectively the patients were divided into two groups. Patients in group 1 (18 patients) had received iron for less than 12 weeks, and their haemoglobin levels fell significantly during pregnancy (P less than 0-001). Haemoglobin levels in 16 patients who had received iron for more than 12 weeks (group 2), however, did not fall significantly during pregnancy (P less than 0-6). It is suggested (contrary to common practice) that patients with beta-thalassaemia trait should be given iron supplements during pregnancy. Serum folate and vitamin B12 levels did not change significantly in these patients and there was no increase in the incidence of maternal or fetal complications.     

Pregnancy management in the bitch

Pregnancy management to optimize maternal and neonatal health begins with breeding management and the selection of normal, healthy brood stock in ideal body condition. After breeding, a commercial diet appropriate for reproduction and lactation should be fed. Typically these contain 29–32% protein of animal source, at least 18% fat, 20–30% carbohydrate, and essential vitamins, minerals and fatty acids. Pregnancy is confirmed approximately 25 d after breeding. A “maternity ward” and whelping box should be provided. Steady increases in caloric intake and body weight are expected as pregnancy progresses. Weight loss should not occur. Throughout pregnancy, changes in the bitch’s attitude, activity, appetite, body weight, and physical findings should be monitored by the owner. If appetite and body weight do not continue to increase, or if any signs of illness develop, maternal health should be assessed with a complete physical examination and a CBC, biochemical profile, and free-catch urinalysis. Fetal health should be assessed with ultrasonography. Maternal or fetal abnormalities will put the pregnancy at risk. Impending parturition and the progress of labor and delivery can be monitored by assessing rectal temperature, serum concentrations of progesterone, and/or uterine and fetal monitors. This article reviews the physiology of canine pregnancy and parturition, and typical schemes used to manage normal canine pregnancy to optimize maternal and puppy health.     

Low iron diet and parenteral cadmium exposure in pregnant rats: the effects on trace elements and fetal viability

The effects of latent iron deficiency combined with parenteral subchronic or acute cadmium exposure during pregnancy on maternal and fetal tissue distribution of cadmium, iron and zinc, and on fetal viability were evaluated. Timed-pregnant Sprague-Dawley rats were fed on semisynthetic test diets with either high iron (240 mg kg) or low iron (10 mg kg), and concomitantly exposed to 0, 3 or 5 mg cadmium (as anhydrous CdCl₂) per kilogram body weight. Animals were exposed to cadmium from gestation day 1 through 19 by subcutaneously implanted mini pumps (Subchronic exposure) or on gestation day 15 by a single subcutaneous injection (Acute exposure). All rats were killed on gestation day 19. Blood samples, selected organs and fetuses were removed and prepared for element analyses by atomic absorption spectrometry. Low iron diet caused decreases in maternal body weight, maternal and fetal liver weights, placental weights and tissue iron concentrations. By cadmium exposure, both subchronic and acute, tissue cadmium concentrations were increased and the increase was dose-related, maternal liver and kidney zinc concentrations were increased, and fetal zinc concentration was decreased. Cadmium concentration in maternal liver was additionally increased by low iron diet. Acute cadmium exposure caused lower maternal body and organ weights, high fetal mortality, and decreased fetal weights of survivors. In conclusion, parenteral cadmium exposure during pregnancy causes perturbations in essential elements in maternal and fetal compartments. Acute cadmium exposure in the last trimester of gestation poses a risk for fetal viability especially when combined with low iron in maternal diet.     

The apparent effect of iron supplementation on serum selenium levels in teenage pregnancy

Numerous studies have suggested a significant role of selenium in the prevention of gynecological carcinoma. These were epidemiological and prospective in humans and therapeutic in laboratory animals. However, no studies have been reported regarding the normal serum selenium levels during pregnancy. The maternal total blood volume increases 30-50% during the second and third trimesters, resulting in lower measured serum levels for those metabolites, which are not increased significantly during pregnancy. A longitudinal study of the serum selenium levels in teenage pregnancy during the last two trimesters and 3 mo postpartum showed progressive elevation from 49 +/- 7 microg/dL after the 32nd week of pregnancy to 114 +/- 7 microg/dL at term, which was statistically significant (p < or = 0.001). Prenatal supplementation with 18 mg of iron per day prevented this elevation. The results of this study suggest that serum selenium levels in women normally double during pregnancy and this doubling is prevented by the minimal daily supplementation of 18 mg of iron, which may be due to increased absorption of selenium into the erythrocytes and incorporation into the glutathione peroxidase enzyme.     

The Relation between Serum Phosphorus Levels and Clinical Outcomes after Acute Myocardial Infarction

Background

Elevated serum phosphorus levels have been linked with cardiovascular disease and mortality with conflicting results, especially in the presence of normal renal function.

Methods

We studied the association between serum phosphorus levels and clinical outcomes in 1663 patients with acute myocardial infarction (AMI). Patients were categorized into 4 groups based on serum phosphorus levels (<2.50, 2.51–3.5, 3.51–4.50 and >4.50 mg/dL). Cox proportional-hazards models were used to examine the association between serum phosphorus and clinical outcomes after adjustment for potential confounders.

Results

The mean follow up was 45 months. The lowest mortality occurred in patients with serum phosphorus between 2.5–3.5 mg/dL, with a multivariable-adjusted hazard ratio of 1.24 (95% CI 0.85–1.80), 1.35 (95% CI 1.05–1.74), and 1.75 (95% CI 1.27–2.40) in patients with serum phosphorus of <2.50, 3.51–4.50 and >4.50 mg/dL, respectively. Higher phosphorus levels were also associated with increased risk of heart failure, but not the risk of myocardial infarction or stroke. The effect of elevated phosphorus was more pronounced in patients with chronic kidney disease (CKD). The hazard ratio for mortality in patients with serum phosphorus >4.5 mg/dL compared to patients with serum phosphorus 2.50–3.50 mg/dL was 2.34 (95% CI 1.55–3.54) with CKD and 1.53 (95% CI 0.87–2.69) without CKD.

Conclusion

We found a graded, independent association between serum phosphorus and all-cause mortality and heart failure in patients after AMI. The risk for mortality appears to increase with serum phosphorus levels within the normal range and is more prominent in the presence of CKD.     

High risk maternal factors related to fetal wastage in rural Bangladesh

An analysis of the relationship between fetal mortality (early fetal death and stillbirth), pregnancy order, maternal age, and previous fetal deaths in a rural Bangladesh population characterized by high fertility and mortality and the virtual absence of obstetric and other medical care indicates that early fetal wastage and stillbirth are higher among pregnancy orders 1 and 6, or higher than among orders 2 and 3, with the increased risk particularly apparent among those pregnancies following 2 or more previous fetal deaths. The data consist of the 21,144 pregnancies that occurred to the women in Matlab, Bangladesh, 1966-1969. By a multiple regression technique allowing for pregnancy order and previous fetal deaths, adjustments were made for age of the mother, and after allowances were made for previous fetal deaths, adjustments were made for pregnancy order. Results show the fewest fetal deaths in 2nd and 3rd pregnancies, and most at the highest parities. 10% of all pregnancy terminations 1966-1969 were registered as fetal deaths. Women in the higher pregnancy orders who have not experienced previous fetal deaths or only 1 fetal death have only a slight increase in the risk of fetal death compared to women in pregnancy orders 2 and 3. It is concluded that the virtual absence of medical care facilities is responsible for the large numbers of fetal deaths due to complications of gestation, delivery, and environmental influences. It also results in a higher maternal mortality of women with pregnancy complications related to fetal deaths. This absence of obstetric care and the high maternal mortality in this population may allow only women without reproductive impairments to reach the higher pregnancy orders.     

Fetal DNA in maternal serum: does it persist after pregnancy?

Fetal DNA and cells present in maternal blood have previously been used for non-invasive prenatal diagnosis. However, some fetal cells can persist in maternal blood after a previous pregnancy. Fetal rhesus status and sex determination have been performed by using amplification by real-time polymerase chain reaction (PCR) of fetal DNA sequences present in maternal circulation; no false-positive results related to persistent fetal DNA from a previous pregnancy have been reported. This idea has recently been challenged. An SRY real-time PCR assay was performed on the serum of 67 pregnant women carrying a female fetus but having previously given birth to at least one boy and on the serum of 30 healthy non-pregnant women with a past male pregnancy. In all cases, serum was negative for the SRY gene. These data suggest that fetal DNA from a previous pregnancy cannot be detected in maternal serum, even by using a highly sensitive technique. Therefore, non-invasive prenatal diagnosis by fetal sex determination for women at risk of producing children with X-linked disorders, and fetal RHD genotyping is reliable and secure as previously demonstrated.     

Total Parenteral Nutrition in Management of Hyperlipidemic Pancreatitis During Pregnancy

ObjectiveTo describe a case of severe gestational hyperlipidemic pancreatitis successfully managed with minimal-lipid-containing parenteral nutrition (PN) followed by a minimal-fat diet, which resulted in delivery of a healthy full-term neonate.MethodsWe present the case of a young woman with gestational hyperlipidemic pancreatitis whose management included the use of PN during pregnancy. In addition, we review the literature pertaining to the management of hyperlipidemic pancreatitis during pregnancy and discuss the role for PN.ResultsA 32-year-old gravida 2, para 1 woman at 27 weeks 3 days of gestation presented with 1 day of nausea, bilious emesis, and severe abdominal pain caused by pancreatitis attributable to hypertriglyceridemia. Her initial serum triglyceride concentration was 9,450 mg/dL. She received fluids intravenously and minimal-lipid PN until resolution of her symptoms. The serum triglyceride level remained less than 850 mg/dL during administration of PN. She subsequently tolerated a minimal-fat diet, while the serum triglyceride level was maintained at less than 1,400 mg/dL, until delivery of a full-term, healthy neonate.ConclusionIn severe gestational hyperlipidemic pancreatitis, PN offers a safe and flexible treatment option by providing pancreatic rest and controlling serum triglyceride concentrations while maintaining fetal and maternal nutritional support. (Endocr Pract. 2005;11:325-330)     

The incidence and clinical relevance of anti-HLA and/or MICA antibodies in patients with long-term survival renal allo-grafts

The purpose of this study was to analyze the incidence and clinical relevance of anti-HLA and/or MICA antibodies in renal allo-grafts transplant recipients with long-term renal survival (> 5 years). This retrospective study collected post-transplant serum samples from a total of 110 patients which were used to detect the incidence of anti-HLA and/or MICA antibodies as well as anti-HLA donor specific antibodies. Among these 110 patients, 72 patients had antibodies against HLA and/or MICA at the time of test, 61 had only anti-HLA antibodies, 31 had anti-MICA antibodies, and 38 were antibody negative. There was no difference in the number of patients developing antibodies against non-donor specific antibodies, donor specific antibodies, Class I donor specific antibodies, Class II donor specific antibodies or MICA antibodies between normal function group (serum creatinine level < 2.0 mg/dL) and dysfunction group (serum creatinine level > 2.0 mg/dL). For the serum creatinine level, estimated glomerular filtration rate and blood urea nitrogen level, patients with different antibodies were not statistically different to antibody-negative patients. Cox regression analysis showed that the type of transplantation and HLA mismatch number were significant negative risk factors for the development of anti-HLA (P < 0.05). Our results suggested that the anti-HLA antibody status has little impact on the renal graft function in the long-term survival allo-graft renal recipients.     

Influence of Lower Cutoff Values for 100-G Oral Glucose Tolerance Test and Glycemic Profile for Identification of Pregnant Women at Excessive Fetal Growth Risk

ObjectiveTo evaluate data from patients with normal oral glucose tolerance test (OGTT) results and a normal or impaired glycemic profile (GP) to determine whether lower cutoff values for the OGTT and GP (alone or combined) could identify pregnant women at risk for excessive fetal growth.MethodsWe classified 701 pregnant women with positive screening for gestational diabetes mellitus (GDM) into 2 categories-(i) normal 100-g OGTT and normal GP and (2) normal 100-g OGTT and impaired GP—to evaluate the influence of lower cutoff points in a 100-g OGTT and GP (alone or in combination) for identification of pregnant women at excessive fetal growth risk. The OGTT is considered impaired if 2 or more values are above the normal range, and the GP is impaired if the fasting glucose level or at least 1 postprandial glucose value is above the normal range. To establish the criteria for the OGTT (for fasting and 1,2, and 3 hours after an oral glucose load, respectively), we considered the mean (75 mg/dL, 120 mg/dL, 113 mg/dL, and 97 mg/dL), mean plus 1 SD (85 mg/dL, 151 mg/dL, 133 mg/dL, and 118 mg/dL), and mean plus 2 SD (95 mg/dL, 182 mg/dL, 153 mg/dL, and 139 mg/dL); and for the GP, we considered the mean and mean plus 1 SD (78 mg/dL and 92 mg/dL for fasting glucose levels and 90 mg/dL and 130 mg/dL for 1- or 2- hour postprandial glucose levels, respectively).ResultsSubsequently, the women were reclassified according to the new cutoff points for both tests (OGTT and GP). Consideration of values, in isolation or combination, yielded 6 new diagnostic criteria. Excessive fetal growth was the response variable for analysis of the new cutoff points. Odds ratios and their respective confidence intervals were estimated, as were the sensitivity and specificity related to diagnosis of excessive fetal growth for each criterion. The new cutoff points for the tests, when used independently rather than collectively, did not help to predict excessive fetal growth in the presence of mild hyperglycemia.ConclusionDecreasing the cutoff point for the 100g OGTT (for fasting and 1, 2, and 3 hours) to the mean (75 mg/dL, 120 mg/dL, 113 mg/dL, and 97 mg/dL) in association with the GP (mean or mean plus 1 SD—78 mg/dL and 92 mg/dL for the fasting state and 90 mg/dL and 130 mg/dL for 1- or 2-hour postprandial values—increased the sensitivity and specificity, and both criteria had statistically significant predictive power for detection of excessive fetal growth. (Endocr Pract. 2008;14:678-685)