Emily B. Cochran

Prevalence of, and risk factors for, upper gastrointestinal tract bleeding in critically ill pediatric patients

ObjectiveTo determine the occurrence of, and risk factors for, the development of upper gastrointestinal (GI) tract bleeding in critically ill pediatric patients. DesignProspective, descriptive, comparative study. SettingICU in a tertiary care pediatric hospital affiliated with a university. PatientsAll patients <19 yrs of age who were admitted to the ICU for a 4-month period (n = 429) were eligible for inclusion. A total of 221 patients were excluded for reasons listed below. Thus, 208 patients were studied. InterventionNone. Measurements and Main ResultsPatients were evaluated for overt upper GI bleeding as indicated by coffee ground material or bright red blood in gastric aspirates or black, tarry stools. Excluded were patients who were transferred out of the ICU within 24 hrs of admission, were receiving medications that would alter their risk for upper GI bleeding, or had a GI tract surgical procedure. Patients were categorized by diagnoses and analyzed for relative risk for upper GI bleeding. Of the 208 patients included, 25% had evidence of upper GI bleeding. There was no association between upper GI bleeding and age, weight, race, or sex. Diagnoses independently associated with an increased risk for upper GI bleeding were: circulatory shock, an operative procedure 3 hrs in duration, and trauma. No clinically important sequelae were directly attributable to upper GI bleeding in this group of patients; however, intervention with antacids and histamine-2 receptor (H-2) antagonists likely decreased the progression of GI bleeding. ConclusionsOvert evidence of upper GI bleeding is not uncommon in critically ill pediatric patients. Certain diagnoses or risk factors may predispose these patients to develop upper GI bleeding.

Digoxin-like immunoreactive substance in pregnancy

Our aims were to determine the potential usefulness of digoxin-like immunoreactive substances in the prediction of preeclampsia, to study the relationship between fetal production of these substances and maternal serum levels, and to evaluate the association between digoxin-like immunoreactive substances and plasma volume findings in preeclamptic pregnancies. Serum digoxin-like immunoreactive substance concentrations were measured in normotensive and preeclamptic pregnant women and in umbilical artery and vein blood samples. None of the patients in the first trimester (n = 53) and 11% of those in the second (n = 56) had detectable levels of this substance. However, 91% of the patients in the third trimester (n = 161) had positive results. The concentrations of digoxin-like immunoreactive substances in the preeclamptic group (n = 78) were significantly (p less than 0.005) lower than those of third-trimester (n = 83) normotensive patients (0.22 +/- 0.12 versus 0.32 +/- 0.15 ng/ml). However, there were no significant differences between the two groups regarding digoxin-like immunoreactive substance concentrations when matched for gestational age (41 patients in each group). Digoxin-like immunoreactive substance concentrations in umbilical vessels were significantly higher (p less than 0.001) than the corresponding maternal levels. Umbilical vessel digoxin-like immunoreactive substance levels demonstrated good correlation with fetal gestational age and birth weight in both normotensive and preeclamptic pregnancies. On the other hand, there was a poor (r = 0.02; p = 0.91) correlation between plasma volume findings and digoxin-like immunoreactive substance concentration. We conclude that the digoxin-like immunoreactive substance level may be of very little value in the prediction of preeclampsia. The presence of digoxin-like immunoreactive substance at greater concentrations in the umbilical cord blood samples suggests the possibility of the fetus as the source of this substance. Digoxin-like immunoreactive substances may not play a major role in plasma volume expansion during pregnancy.

The influence of gestational age and preeclampsia on the presence and magnitude of serum endogenous digoxin-like immunoreactive substance(s)☆☆☆

Abstract Digoxin-like immunoreactive substance(s) has been measured in serum during pregnancy. Because of its presence in pregnancy, investigators have suggested that digoxin-like immunoreactive substance may play an etiologic role in the development of preeclampsia. The objectives of this study were to evaluate the relationship between maternal digoxin-like immunoreactive substance and gestational age and compare digoxin-like immunoreactive substance concentrations in patients with and without preeclampsia who were in the third trimester. Two hundred twenty patients were studied during either the first (n = 53), second (n = 56), or third (n = 111) trimester of pregnancy. Digoxin-like immunoreactive substance was undetectable in the serum of patients during the first trimester; however, 11% of second-trimester and 96% of third-trimester patients had measurable levels of serum digoxin-like immunoreactive substance (p

Effect of the Continuous Administration of Fat Emulsion on the Infiltration of Intravenous Lines in Infants Receiving Peripheral Parenteral Nutrition Solutions

Animal data and anecdotal human experience suggest that vascular damage induced by the infusion of dextrose/amino acid solutions may be ameliorated by the concomitant administration of fat emulsion. We prospectively evaluated the effect of the continuous infusion of peripheral nutrition solutions with and without fat emulsion on the incidence of, probability of, and time to infiltration of peripheral venous lines in infants (median age: 1.0 month; range 1 day-11.9 months). Ninety-seven peripheral venous lines were studied in 53 infants who received 10% dextrose (n = 34), 10% dextrose/2% amino acids (n = 30), or 10% dextrose/2% amino acids/fat emulsion (n = 33). Solutions were administered by positive pressure infusion devices through Teflon catheters with similar gauge. Dextrose, amino acid, electrolyte, and mineral content was standardized for the dextrose/amino acid and dextrose/amino acid/fat emulsion groups. The three groups were similar with respect to age, race, gender, weight, administration of intravenous medications, and catheter site (p greater than 0.05). Patients receiving dextrose/amino acid or dextrose/amino acid/fat emulsion had greater rates of solution administration than those receiving dextrose alone (p less than 0.002). Infiltration occurred in 71% of dextrose, 66% of dextrose/amino acid, and 67% of dextrose/amino acid/fat emulsion solutions (p greater than 0.05). The probability of infiltration was greater for infants receiving dextrose/amino acid than for those receiving either dextrose or dextrose/amino acid/fat emulsion (p = 0.01). The mean +/- SEM length of time the intravenous sites were viable was significantly shorter for the dextrose/amino acid solutions (26.3 +/- 3.3 hr) compared to the dextrose (54.9 +/- 7.8 hr) and dextrose/amino acid/fat emulsion (43.6 +/- 4.2 hr) groups. No site complications were associated with the infiltration of any solution. We conclude that the incidence of infiltration among the three solution groups studied is not different. However, the time to infiltration is prolonged and the probability of infiltration is decreased following the infusion of either dextrose alone or dextrose/amino acid/fat emulsion solutions when compared to the administration of dextrose/amino acid solutions without concomitant fat emulsion infusion.

Prefilter and Postfilter Cysteine/Cystine and Copper Concentrations in Pediatric Parenteral Nutrition Solutions

Pediatric amino acid products contain lower concentrations of methionine and require the addition of L-cysteine HCl just before infusion. Reports of a potential interaction between cysteine and copper, a routine addition to parenteral nutrition solutions, have appeared in the literature. This study serially evaluated cysteine/cystine and copper concentrations pre- and postfilter (0.22 microns) in two parenteral nutrition formulations prescribed for normal fluid (NF) or fluid-restricted (FR) pediatric patients. Solutions were infused via a peristaltic pump in vitro for 24 hours. Pre- and postfilter samples were obtained immediately after the infusion began (time 0) and at 1, 2, 3, 4, 8, 12, 16, 20, and 24 hours during the infusion. At 24 hours pre- and postfilter, cysteine/cystine were quantitated at 87.3% and 85% of the initial concentrations, respectively, for the NF solution and 92% and 91.5% for the FR solution. Pre- and postfilter copper was quantitated at 87.5% and 92.5% of the initial concentrations, respectively, for the NF solution and 88.2% and 95.2% for the FR solution. The 24 hour area under the curve for cysteine/cystine was 88.4% for the NF solution and 96.4% for the FR solution. For copper, the area under the curve for the NF solution was 99% and 96.7% for the FR solution. There was no visual evidence of incompatibility or precipitation. We conclude that copper-containing parenteral nutrition solutions with L-cysteine HCl added immediately before infusion are relatively stable after compounding and infusion over 24 hours.

Comparison of a Noninstrumented Immunoassay for Carbamazepine to High Performance Liquid Chromatography and Fluorescence Polarization Immunoassay

Summary: The accuracy, precision, and potential clinical utility of a new whole blood, noninstrumented immuno‐chromatographic assay (AccuLevel) for carbamazepine (CBZ) was evaluated in a multicenter trial including 100 pediatric and 205 adult patients. The AccuLevel assay, a fluorescence polarization immunoassay (FPIA), and high‐performance liquid chromatography (HPLC) were used to determine CBZ concentration in samples from 111 female and 194 male patients aged 2–72 years (median 25 years). Mean ± SD plasma CBZ concentrations in all patients were 7.4 ± 3.0 μg/ml with the AccuLevel assay and 7.5 ± 2.9 (μg/ml with FPIA. In 204 patients, the mean concentration determined by the HPLC assay was 7.7 ± 3.0 μ‐g/ml, whereas concentrations determined by the AccuLevel and FPIA assays were 8.0 ± 3.1 and 8.1 ± 3.1μg/ml, respectively. Concentrations determined by the AccuLevel and FPIA assays were significantly higher than those quantified by HPLC (p < 0.05), but not different from each other. In addition, the AccuLevel assay was highly correlated with FPIA (r= 0.97) and HPLC (r= 0.98). Coefficients of variation for the AccuLevel assay at 8 |xg/ml ranged from 6.8 to 7.5% for the three institutions. We conclude that the AccuLevel assay is a simple, reliable method for determining CBZ concentration in a small volume of whole blood and is an acceptable alternative for assessment of CBZ therapy and individualization of CBZ dosage in the physicians office or emergency room.

PERIPHERAL VENOUS LINE INFILTRATION IN INFANTS RECEIVING 10% DEXTROSE, 10% DEXTROSE/AMINO ACIDS, OR 10% DEXTROSE/AMINO ACIDS/FAT EMULSION

97 peripheral venous lines (PVL) were prospectively studied to determine 1) if the infiltration incidence differs for 10% dextrose (D) vs 10% dextrose/amino acid solutions (DAA) and 2) if the continuous infusion of fat emulsion concurrently with dextrose/amino acid solutions (DAAF) affects the incidence or time to infiltration, The patients had a mean age ± SD of 2.1 ± 2.9 months (range:0.03-12). 34 patients received (D), 30 (DAA)7 and 33 (DAAF). Electrolyte and mineral content was standardized for the DAA and DAAF groups. All solutions were infused via teflon catheters with similar gauge. The three groups were similar with respect to age, race, gender, weight, and site (p>0.05). Patients receiving DAA and DAAF had higher infusion rates than the D group (p<0.001). 70% of D, 66% of DAA and 67% of DAAF infiltrated (p>0.05). The probability of infiltration was significantly different with DAA > than D or DAAF (p=0.01).No complications were noted following infiltration of any solution. We conclude: 1) there is a significant difference in the time to and probability of infiltration when DAA are infused, and 2) continuous infusion of fat emulsion with DAA significantly prolongs the survival of a PVL.