The utility of routine laboratory testing in hypoglycaemic emergency department patients.

BACKGROUND: After treating a hypoglycaemic patient, how in-depth a work-up is required to prevent their next hypoglycaemic episode? OBJECTIVE: To determine the utility of routine laboratory testing in the management of hypoglycaemia. METHODS: A cross-sectional study at two urban teaching hospitals from July 2006 to July 2007. The study included adult patients (> or = 18 years) with hypoglycaemia (fingerstick glucose < or = 60 mg/dl (3.33 mmol/l) in the emergency department or altered mental status resolved by glucose or glucagon). Predictor variables were age, gender, medical history, physical examination, hypoglycaemic agent (insulin vs oral hypoglycaemic). Outcome variables were electrolyte abnormality (serum sodium outside the range of 135-145 mmol/l; serum potassium outside the range of 3.5-5.0 mEq/dl), leucocytosis (white blood cell count >15,000/high power field) or urinary tract infection. Continuous data are presented as mean (SD). Categorical data are presented as percentages with 95% confidence intervals. Student's t and Fisher's exact tests were used to compare data when appropriate (alpha = 0.05, two-tailed). RESULTS: 291 patients were studied with a mean age of 64 years (SD 16) (range 22-95) 54 women. 200 patients (69%, 95% CI 63% to 74%) had at least one laboratory abnormality. These included newly diagnosed renal failure (23%), pre-existing renal failure (32%), hypokalaemia (8%), hyperkalaemia (11%), leucocytosis (4.2%) and pyuria (19%). No significant difference was found between the rates of abnormal laboratory results in patients on insulin versus those on oral hypoglycaemics. Higher admission rates (p = 0.001) were also observed in patients with abnormal (70%) compared with normal (53%) laboratory results. CONCLUSION: The high rate of laboratory abnormalities in hypoglycaemic patients justifies routine testing.

Serum ionized magnesium and calcium levels in adult patients with seizures.

OBJECTIVE: Prior studies have been equivocal about whether or not serum levels of the divalent ions calcium and magnesium are altered during different types of seizures. Magnesium is a potential modulator of seizure activity because of its ability to antagonize the excitatory calcium influx through the N-methyl-D-aspartate (NMDA) receptor. We hypothesize that serum ionized levels of calcium (Ca(2+)) and magnesium (Mg(2+)) would be altered significantly during certain types of seizures. MATERIAL AND METHODS: A convenience sample of seizure patients presenting to an emergency department (ED) were enrolled in this prospective study. Novel ion-selective electrodes were used to measure Ca(2+) and Mg(2+). Data were reported as mean values+/-standard deviations. Group comparisons were analyzed by ANOVA with post-hoc testing using the Bonferroni, or the Fisher exact test, where appropriate, alpha = 0.05 (two-tailed). RESULTS: Forty-nine patients with seizure and 32 healthy racially matched controls were included in the study. Seizure patients had a significantly (p<0.001) lower mean Mg(2+), but not total serum Mg and a significantly (p<0.001) higher Ca(2+)/Mg(2+) ratio than that in controls. CONCLUSIONS: We were able to show significantly lower Mg(2+) and higher ionized Ca(2+)/Mg(2+) ratios in seizure patients compared with a racially matched control group.

The effect of pregnancy on the response to blood loss in a rat model.

STUDY OBJECTIVES: A commonly held belief is that the blunted hemodynamic response to hemorrhage observed in pregnant women is secondary to expanded blood volume. In addition to increased blood volume, pregnancy is also a vasodilated state. Vasodilatation may have deleterious effects on the response to hemorrhage by inhibiting central blood shunting after blood loss. How these conflicting variables of increased blood volume and vasodilatation integrate into a whole body model of maternal hemorrhagic shock has yet to be studied in a controlled experiment. We tested the null hypothesis that there would be no difference in the hemodynamic and metabolic responses to hemorrhage between pregnant (PRG) and non-pregnant (NPRG) rats. METHODS: Twenty-four adult female Sprague-Dawley rats (12 PRG and 12 NPRG) were anesthetized with Althesin via the intraperitoneal route. Femoral arteries were cannulated by cut-down. Twelve (six PRG and six NPRG) rats underwent controlled catheter hemorrhage of 25% of their total blood volume. Twelve rats (six PRG and six NPRG) served as non-hemorrhage controls. Mean arterial pressure (MAP) and base excess (BE) were measured pre-hemorrhage and then every 15 min post-hemorrhage for the next 90 min. Data were reported as mean+/-standard error of the mean (S.E.M.) over the 90-min post-hemorrhage observation period. Group comparisons were analyzed by ANOVA with repeated values post-hoc by Bonferroni. Statistical significance was defined by an alpha=0.05. RESULTS: PRG and NPRG rats were evenly matched for MAP (P=0.788) and BE (P=0.146) pre-hemorrhage. Post-hemorrhage there were no mortalities in either group. Post-hemorrhage both the PRG and NPRG groups experienced significant (P=0.011) drops in systolic and diastolic blood pressures as compared to their non-hemorrhage controls. Post-hemorrhage there was no significant (P=0.43) difference in MAP between the PRG (89+/-2 mmHg) and NPRG (80+/-2 mmHg) rats. BE also dropped significantly within both PRG (P=0.004) and NPRG (P=0.001) groups post-hemorrhage. No significant (P=0.672) difference was noted in BE between PRG and NPRG groups post-hemorrhage -6.1+/-0.3 mEq/l and -6.9+/-0.4 mEq/l, respectively. CONCLUSION: After a controlled hemorrhage of 25% of total blood volume we found no significant differences in MAP and BE between pregnant and non-pregnant rats. Pregnancy does not affect the response to hemorrhage.

Generalized seizure in a 30-year-old man with presumed intracranial hemorrhage: a case report.

A 30-year-old man presented to the emergency department of another institution with recurrent episodes of generalized tonic-clonic seizures. He was found to be hypocalcemic and was treated with an intravenous infusion of calcium followed by Dilantin. A computed tomography scan of the head was obtained, and the patient was transferred to our institution for neurosurgical evaluation and possible intervention of what was thought to be bilateral intracranial hemorrhages. After further evaluation at our institution, the diagnoses of hypoparathyroidism associated with hypocalcemic seizures and basal ganglia calcifications were established on both clinical and biochemical grounds. This case report discusses the clinical presentation, pathogenesis, diagnostic work-up, and management of hypoparathyroidism and associated seizures, highlighting the possible diagnostic and therapeutic pitfalls that are most pertinent to the emergency physician.

The effect of hypertension on the response to blood loss in a rodent model.

OBJECTIVE: Hypertensive patients having higher baseline peripheral resistance and sympathetic tone than normotensive patients may have aberrant responses to hemorrhage. In an attempt to further characterize this clinical observation, the authors compared the hemodynamic and metabolic responses to hemorrhage between spontaneously hypertensive rats (SHR) and normotensive rats (NTR). METHODS: Twenty adult rats (10 NTR and 10 SHR) were anesthetized with althesin via the intraperitoneal route. Femoral arteries were cannulated by cutdown. Twelve (6 SHR and 6 NTR) rats underwent controlled catheter hemorrhage of 25% of their total blood volumes. Eight rats (4 SHR and 4 NTR) served as nonhemorrhage controls. Mean arterial pressure (MAP) and base excess (BE) were measured prehemorrhage and then every 15 minutes for the next 120 minutes. Data were reported as mean +/- standard error of the mean (SEM). Group comparisons were analyzed by ANOVA with repeated values post-hoc by Bonferroni. Statistical significance was defined by an alpha = 0.05. RESULTS: Immediately after hemorrhage, the SHR group experienced a significantly (p < 0.001) greater drop in MAP of 70 +/- 4% in the SHR vs 40 +/- 6% in the NTR. Blood pressure in the NTR returned to control values 15 minutes after hemorrhage, but the SHR remained relatively hypotensive for the entire length of the experiment. Base excess in the SHR decreased significantly (p < 0.004) by 8.2 +/- 2 mmol/L from control values, as compared with no changes in BE for the NTR. CONCLUSIONS: The authors observed significant differences in the response to hemorrhage between hypertensive and normotensive rats. Hypertensive rats experienced a more profound hemorrhagic shock insult than normotensives for the same degree of blood loss.