American Association for Clinical Chemistry
Better health through laboratory medicine
November 2009 Clinical Laboratory News: Diagnostic Profiles

CLN Banner Logo

November 2009: Volume 35, Number 11

Treatment for Mild Gestational Diabetes Beneficial

New research indicates that treatment of mild gestational diabetes reduces the risk of common birth and maternal complications (NEJM 2009;361:1339–1348). The study, which was conducted by 15 medical centers that are part of the Maternal Fetal Medicine Units Network of NIH, provided the first conclusive evidence that even mild gestational diabetes should be treated. While virtually all pregnant women are screened for the condition and treatment efficacy for severe cases has been demonstrated, there had been not been sufficient evidence to encourage treatment in milder cases.

The study involved 958 women who were between the 24th and 31st week of pregnancy and had abnormal oral glucose tolerance test results but fasting glucose levels <95 mg/dL. A total of 458 were assigned to treatment, which included a minimum of 4 blood glucose checks per day, counseling on diet and exercise, and insulin as necessary (only 7% of patients). There was no significant difference between treatment and control subjects in the primary composite outcome of stillbirth or perinatal death and neonatal complications, including hyperbilirubinemia and hypoglycemia. However, there were significant differences in several secondary outcomes, including mean birth weight (3,302 g in treatment subjects versus 3,408 g in controls) and neonatal fat mass (427 g in treatment subjects versus 464 g in controls). In addition, the rates of shoulder dystocia, cesarean delivery, preeclampsia, and gestational hypertension were significantly lower in the treatment arm.

Study Suggests New Role for CRP

New research suggests that C-reactive protein may play a protective role in some patients with acute respiratory distress syndrome (ARDS) (Chest 2009;136:471–480). If confirmed in other patient populations and mechanistic studies, the findings could lead to a reappraisal of conventional views of the role of CRP in disease.

The prospective observational study sought to characterize plasma CRP levels in patients with early ARDS. CRP levels have been demonstrated to have prognostic and/or diagnostic value in many diseases such as sepsis, pneumonia, coronary artery disease, stroke, and rheumatic disease, among others. In most instances, higher CRP levels have correlated with adverse outcomes. However, the relationship between CRP levels and ARDS and acute lung injury has not been explored extensively.

The study involved 177 patients admitted to intensive care units at Massachusetts General Hospital in Boston who had risk factors for ARDS and no exclusion criteria. CRP measurements were taken within 48 hours of ARDS onset. Outcomes included 60-day mortality, 28-day daily organ dysfunction scores, and number of ventilator-free days. The researchers found that CRP levels were significantly higher in ARDS survivors versus non-survivors (median, 176.5 mg/L versus 133.5 mg/L), and that patients with higher CRP levels had less organ dysfunction over time and more ventilator-free days than those with lower levels. An increasing CRP level was associated with a significantly higher probability of survival at 60 days, a difference that persisted after adjustment in a multivariate analysis. The greatest difference in mortality occurred between patients in the two highest and two lowest deciles of CRP levels.

According to the investigators, the findings seem to contradict long-held views about the role of CRP as an inflammatory marker predictive of risk. Studies have demonstrated a link between elevated CRP levels and adverse outcomes in coronary artery disease, sepsis, and septic shock, but some, particularly those involving sepsis and septic shock, have not found CRP levels to be associated with altered outcomes, and others have suggested that failure of CRP to decrease over time is predictive of worse outcomes.
Neutrophils recruited to the lung through chemoattractant molecules accumulate in the lungs in patients with ARDS and are thought to be crucial in lung injury. The authors speculate that lower levels of CRP may stimulate neutrophil chemotaxis but higher levels inhibit it, along with other neutrophil functions.

Intervention Timing Not Associated with Outcomes in ACS

New research indicates that among patients with non-ST elevation acute coronary syndrome (NSTE-ACS), an immediate intervention strategy compared with waiting until the next working day does not result in a difference in myocardial infarction as defined by peak troponin I (cTnI) level (JAMA 2009;302:947–954).

Investigators in the Angioplasty to Blunt the Rise of Troponin in Acute Coronary Syndromes Randomized for an Immediate or Delayed Intervention (ABOARD) sought to evaluate the optimal timing of coronary angiography and intervention in high-risk patients. Little information has been published in this regard, although invasive strategies have generally been accepted as the best treatment option. ABOARD involved 352 patients at 13 institutions, all of whom had NSTE-ACS and a thrombolysis in MI (TIMI) score of 3 or greater. Patients were randomized to undergo invasive treatments immediately or on the next working day, which was defined as 8 to 60 hours after enrollment. Patients also received antithrombotic and anticoagulant medications including abciximab, aspirin, and clopidogrel. β-blockers, statins and angiotensin-converting enzyme inhibitors also were strongly recommended as concomitant therapies. The primary end point was the peak cTn I value during hospitalization. The secondary endpoint was a composite of death, MI, or urgent revascularization at 1-month follow-up.

Mean time from randomization to sheath insertion in percutaneous coronary intervention was 70 minutes in the immediate intervention group, versus 21 in the delayed intervention arm. The researchers found that peak cTnI values did not differ between the two strategies. The secondary end point occurred in 13.7% of patients in the immediate intervention arm, and in 10.2% in the delayed intervention arm (p = 0.31), and the three components of the secondary end point evaluated individually did not differ significantly between the two groups. The study demonstrates that a strategy of immediate catheterization of NSTE-ACS patients is not superior to one of waiting until the next working day.

Timing of Levothyroxine Administration Affects Serum TSH Levels

Georgetown University researchers have documented that when patients take levothyroxine (LT4) in a nonfasting regimen, their serum thyrotropin (TSH) concentration is higher and more variable (J Clin Endocrinol Metab 2009;94:3905–3912). This finding has particular implications for patients in whom specific TSH targets are of great importance, such as pregnant women, the elderly, and those diagnosed with thyroid cancer, cardiac disease, or osteoporosis. Although studies have shown that optimal intestinal absorption of LT4 occurs when patients fast, and LT4 is known to have a narrow therapeutic index, analyses of the impact of variable LT4 regimens on serum TSH concentrations have had inconsistent results.

In this study, 65 patients were randomized to one of six sequences, each consisting of three 8-week LT4 regimens in a three-period crossover design. The regimens were: overnight fasting before breakfast; within 20 minutes of having eaten breakfast; and at least 2 hours after the last meal of the day. Serum TSH, free T4, and total T3 levels were measured at baseline and at the conclusion of each 8-week period—TSH using third-generation immunochemiluminometric assays with a sensitivity of 0.01 mIU/L, FT4 and T3 with chemiluminescent immunoassays.

The researchers found that when LT4 was taken in a fasting state, the mean TSH concentration was 1.06 ± 1.23 mIU/L. In contrast, TSH levels were significantly higher when LT4 was taken with breakfast (2.93 ± 3.29 mIU/L) or at bedtime (2.19 ± 2.66 mIU/L). As the authors note, these findings suggest that patients who have difficulty complying with a fasting regimen of LT4 could simply have their doses increased to achieve lower TSH levels. However, the study also revealed that when patients took LT4 with breakfast or at bedtime, their TSH concentrations were much more variable, with TSH extremes of 0 to 19 mIU/L observed. The authors conclude that in cases in which the goal of therapy is to maintain a specific serum TSH within a relatively narrow range and without significant oscillations, then patients should be advised to take LT4 in a fasting state. Alternatively, if a patient needs to take LT4 at bedtime, then serum TSH levels should be monitored closely for a while to determine whether that individual’s TSH concentrations vary by more than 1 mIU/L from the desired range.

Large Serum Sodium Level Fluctuations Associated with Functional Impairments in Premature Neonates

New research indicates that significant changes in serum sodium levels in preterm neonates are associated with increased risk of impaired functional outcomes after adjustment for gestational age and perinatal and neonatal hospitalization characteristics (Pediatrics 2009;124:e655–e661). The authors of the study cautioned that their findings simply may reflect the severity of illness and/or quality of care of these infants, although a causal relationship between serum sodium levels and development outcomes cannot be excluded. The results suggest that very premature infants should receive cautious fluid and electrolyte management.

Hyponatremia in infants has been associated with increased risk of cerebral palsy and hearing loss, while hypernatremia has been linked with cerebral edema and thrombosis, intracranial bleeding and severe hyperbilirubinemia. However, few studies have examined neurodevelopmental outcomes in very preterm babies who experience hypo- or hypernatremia in the first days of life.

The study involved 237 preterm infants born at <33 weeks who were hospitalized in the neonatal intensive care unit. A total of 3,927 sodium measurements were performed. The researchers assigned the babies to three tertiles based on changes in sodium levels, with small change defined as 0 to 7 mEq/L, large as 8 to13 mEq/L, and very large defined as >13 mEq/L. The infants subsequently were given clinical and development assessments when they were at the corrected age of 2, and were classified into two groups, either with impaired functional outcomes or not.

The investigators found that after adjustment for gestational age and perinatal and neonatal hospitalization characteristics, the odds ratio for risk of impaired functional outcome was 3.5 in babies with large serum sodium changes and 5.1 in those with very large changes. The authors speculate that significant variations in serum sodium levels may cause cerebral lesions and consequently impair functional outcomes.