Design and statistical analysis of oral medicine studies: common pitfalls.

A growing number of articles are emerging in the medical and statistics literature that describe epidemiologic and statistical flaws of research studies. Many examples of these deficiencies are encountered in the oral, craniofacial, and dental literature. However, only a handful of methodologic articles have been published in the oral literature warning investigators of potential errors that may arise early in the study and that can irreparably bias the final results. In this study, we briefly review some of the most common pitfalls that our team of epidemiologists and statisticians has identified during the review of submitted or published manuscripts and research grant applications. We use practical examples from the oral medicine and dental literature to illustrate potential shortcomings in the design and analysis of research studies, and how these deficiencies may affect the results and their interpretation. A good study design is essential, because errors in the analysis can be corrected if the design was sound, but flaws in study design can lead to data that are not salvageable. We recommend consultation with an epidemiologist or a statistician during the planning phase of a research study to optimize study efficiency, minimize potential sources of bias, and document the analytic plan.

Hyperlipidemia in glycogen storage disease type III: effect of age and metabolic control.

While the presence of hyperlipidaemia in glycogen storage disease (GSD) type Ia and Ib is generally accepted, few investigators have adequately assessed lipid profiles of GSD III in children, in whom the presence of hyperlipidaemia may be most prominent. We analysed the lipid profiles in 44 GSD III patients from 6 months to 30 years of age. Hypertriglyceridaemia and hypercholesterolaemia were common in children younger than 3 years of age. Hypertriglyceridaemia correlated negatively with age, and may reflect increased severity of hypoglycaemia in this younger population. The presence of hyperlipidaemia during childhood in these patients identifies another GSD population that could be at risk for early cardiovascular disease (CVD). Consequently, the outcome of clinical trials investigating the vascular effect of hyperlipidaemia in GSD applies to types other than GSD I.

Dichloroacetate therapy attenuates the blood lactate response to submaximal exercise in patients with defects in mitochondrial energy metabolism.

We determined acute and chronic effects of dichloroacetate (DCA) on maximal (MAX) and submaximal (SUB) exercise responses in patients with abnormal mitochondrial energetics. Subjects (n = 9) completed a MAX treadmill bout 1 h after ingesting 25 mg/kg DCA or placebo (PL). A 15-min SUB bout was completed the next day while receiving the same treatment. After a 1-d washout, MAX and SUB were repeated while receiving the alternate treatment (acute). Gas exchange and heart rate were measured throughout all tests. Blood lactate (Bla) was measured 0, 3, and 10 min after MAX, and 5, 10, and 15 min during SUB. MAX and SUB were repeated after 3 months of daily DCA or PL. After a 2-wk washout, a final MAX and SUB were completed after 3 months of alternate treatment (chronic). Average Bla during SUB was lower (P < 0.05) during both acute (1.99 +/- 1.10 vs. 2.49 +/- 1.52 mmol/liter) and chronic (1.71 +/- 1.37 vs. 2.39 +/- 1.32 mmol/liter) DCA vs. PL despite similar exercise intensities between conditions ( approximately 75 and 70% maximal exercise capacity during acute and chronic treatment). Thus, although DCA does not alter MAX responses, acute and chronic DCA attenuate the Bla response to moderate exercise in patients with abnormal mitochondrial energetics.

Distinct phenotypes distinguish the molecular classes of Angelman syndrome.

BACKGROUND: Angelman syndrome (AS) is a severe neurobehavioural disorder caused by defects in the maternally derived imprinted domain located on 15q11-q13. Most patients acquire AS by one of five mechanisms: (1) a large interstitial deletion of 15q11-q13; (2) paternal uniparental disomy (UPD) of chromosome 15; (3) an imprinting defect (ID); (4) a mutation in the E3 ubiquitin protein ligase gene (UBE3A); or (5) unidentified mechanism(s). All classical patients from these classes exhibit four cardinal features, including severe developmental delay and/or mental retardation, profound speech impairment, a movement and balance disorder, and AS specific behaviour typified by an easily excitable personality with an inappropriately happy affect. In addition, patients can display other characteristics, including microcephaly, hypopigmentation, and seizures. METHODS: We restricted the present study to 104 patients (93 families) with a classical AS phenotype. All of our patients were evaluated for 22 clinical variables including growth parameters, acquisition of motor skills, and history of seizures. In addition, molecular and cytogenetic analyses were used to assign a molecular class (I-V) to each patient for genotype-phenotype correlations. RESULTS: In our patient repository, 22% of our families had normal DNA methylation analyses along 15q11-q13. Of these, 44% of sporadic patients had mutations within UBE3A, the largest percentage found to date. Our data indicate that the five molecular classes can be divided into four phenotypic groups: deletions, UPD and ID patients, UBE3A mutation patients, and subjects with unknown aetiology. Deletion patients are the most severely affected, while UPD and ID patients are the least. Differences in body mass index, head circumference, and seizure activity are the most pronounced among the classes. CONCLUSIONS: Clinically, we were unable to distinguish between UPD and ID patients, suggesting that 15q11-q13 contains the only significant maternally expressed imprinted genes on chromosome 15.

Quantitative assessment of brainstem development in Joubert syndrome and Dandy-Walker syndrome.

Key features of Joubert syndrome include developmental delay, hypotonia, hyperpnea and apnea, oculomotor apraxia, and the presence of the molar tooth sign on axial imaging through the brainstem isthmus--the junction of the pons and mesencephalon. Interestingly, 1 in 10 patients with Joubert syndrome has abnormal cerebrospinal fluid collections misdiagnosed as Dandy-Walker variants. Because of important differences in patient management, genetic counseling, and prognosis between these conditions, we undertook a study to determine if the brainstem isthmus is normal in Dandy-Walker syndrome. Using standard landmarks, we evaluated development of the isthmus in normal subjects and in subjects with Joubert syndrome and Dandy-Walker syndrome. Four of five brainstem measures increased with age in normal subjects. In subjects with Joubert syndrome, the depth and length of the interpeduncular fossa were increased, and the width of the isthmus was decreased. In subjects with Dandy-Walker syndrome, the width of the brainstem isthmus was normal, and the molar tooth sign was absent. Although the pons can be hypoplastic in Dandy-Walker syndrome, we conclude that the pontomesencephalic junction is normal. Thus, the molar tooth sign can effectively distinguish between Joubert and Dandy-Walker syndromes. Genetic heterogeneity or epigenetic factors may account for abnormal cerebrospinal fluid collections in some cases of Joubert syndrome.

Platelet transfusions in the neonatal intensive care unit:factors predicting which patients will require multiple transfusions.

BACKGROUND: Previous studies suggest that recombinant thrombopoietin (rTPO) will increase platelet production in thrombocytopenic neonates. However, the target populations of neonates most likely to benefit should be defined. Studies suggest that rTPO will not elevate the platelet count until 5 days after the start of treatment. Therefore, the neonates who might benefit from rTPO are those who will require multiple platelet transfusions for more than 5 days. This study was designed to find means of prospectively identifying these patients. STUDY DESIGN AND METHODS: A historic cohort study of all patients in the neonatal intensive care unit (NICU) at the University of Florida who received platelet transfusions from January 1, 1997, through December 31, 1998, was conducted. RESULTS: Of the 1389 patients admitted to the NICU during the study period, 131 (9.4%) received platelet transfusions. Seventeen were treated with extracorporeal membrane oxygenation and were excluded from further analysis. Of the remaining 114 patients, 55 (48%) received one transfusion and 59 (52%) received more than one transfusion (21 had >4). None of the demographic factors examined predicted multiple platelet transfusions. However, two clinical conditions did; liver disease and renal insufficiency. Neonates who received one platelet transfusion had a relative risk of death 10.4 times that in neonates who received none (p = 0.0001). Neonates who received >4 platelet transfusions had a risk of death 29.9 times that in those who received no transfusions (p = 0.0001). CONCLUSION: NICU patients with liver disease or renal insufficiency who receive one platelet transfusion are likely to receive additional transfusions. Therefore, these patients constitute a possible study population for a Phase I/II rTPO trial.

Teaching hypothesis-oriented thinking to medical students: the University of Florida's clinical investigation program.

Recent studies show alarming decreases in the proportions of physicians applying for federal resources and of graduating medical students who declare strong interest in pursuing careers as physician-scientists. To expose medical students in their formative years to hypothesis-driven experimental investigations in a clinical setting, the first-year curriculum at the University of Florida has involved students as both investigators and study subjects in patient-oriented research conducted in the General Clinical Research Center (GCRC). Each year a hypothesis-driven experiment is conceived by first-year medical students in the university's MD-PhD program. Later in the year, the protocol is implemented in the GCRC by the entire freshman class, whose members serve as volunteer study subjects or as investigators. The experimental data are analyzed by the MD-PhD students, who report their findings at national biomedical research meetings and submit a manuscript on their project to a peer-reviewed journal. The authors describe students' research projects over the first six years of this GCRC-based program. They also describe the responses of former students to a questionnaire about their perceptions of the value of the research program. Most respondents considered the GCRC research exercise to have been useful and relevant to their overall education, and many more declared a current interest in pursuing research careers compared with the number who had declared such interest as freshmen. The authors conclude that early integration of hands-on, patient-oriented research into the medical school curriculum is a positive educational experience for students, and may contribute to their ultimate pursuit of academic research careers.

Epidemiologic and outcome studies of patients who received platelet transfusions in the neonatal intensive care unit.

STUDY DESIGN: We conducted a historic cohort study of neonates who received platelet transfusions at the National Institute of Perinatology, Mexico City, from January 1997 to May 2000. We obtained descriptive and outcome data, and assessed demographic and laboratory means of predicting "good candidates" for a future recombinant thrombopoietin (rTpo) trial. RESULTS: A minority of the transfused patients (11.4%) received only one transfusion; the majority (88.6%) received multiple transfusions. Neonates who received one or more platelet transfusions were more likely to die (24.5% mortality) than neonates who received no platelet transfusions (3.7% mortality). Regression analyses indicated that the presence of liver disease was the best predictor of a "good candidate" for rTpo administration. CONCLUSION: The majority of neonates in our institution who receive platelet transfusions receive multiple, not single, transfusions. Receiving any platelet transfusion is a marker for high risk of death. Neonates with liver disease who receive platelet transfusions might be a reasonable group for a phase I rTpo trial.

Methylenetetrahydrofolate reductase mutation (677C-->T) negatively influences plasma homocysteine response to marginal folate intake in elderly women.

Individuals who are homozygous for the methylenetetrahydrofolate reductase (MTHFR) 677C --> T mutation have depressed serum folate (SF) and elevated plasma total homocysteine (tHcy) concentrations, which may affect folate requirements and increase the risk for coronary artery disease. A controlled metabolic study (14 weeks) using a depletion/repletion protocol was performed in women (aged 60 to 85 years, N = 33) to provide age-specific data on the effects of the MTHFR mutation on SF and tHcy status. Subjects consumed a moderately folate-deplete diet (118 microg/d) for 7 weeks, followed by 7 weeks of folate repletion with 200 or 415 microg/d provided as two different treatments. Following folate depletion, the mean SF concentration was lower for homozygous (P = .017) versus heterozygous subjects. Homozygotes for the 677C --> T mutation showed a higher (P = .015) percent increase in plasma tHcy (44%) than heterozygous (20%) or normal (15%) subjects. At week 7, the mean plasma tHcy concentration was higher in homozygous subjects (12.5 +/- 5.3 micromol/L, mean +/- SD) versus the heterozygous (10.8 +/- 3.8 micromol/L, P = .008) or normal (11.3 +/- 2.7 micromol/L, P = .001) genotype groups. Following folate repletion, plasma tHcy concentrations were not different between genotype groups, despite a higher (P < .016) SF concentration in subjects with the homozygous genotype. These data suggest that older women who are homozygous for the MTHFR 677C --> T mutation may be at risk for greater elevations in plasma tHcy in response to moderately low folate intake as compared with individuals with the normal or heterozygous genotypes.

Pharmacokinetics and pharmacodynamic effects of nicotine nasal spray devices on cardiovascular and pulmonary function.

BACKGROUND: A nasal spray form of nicotine replacement therapy (Nicotrol NS, McNeil Consumer Products Co, Fort Washington, Pa) has been approved and, because of its rapid absorption across the nasal mucosa, may be more effective than nicotine gum or transdermal patches. We tested the hypothesis that the nicotine absorbed into the nasal mucosa would produce significant changes in hemodynamics and pulmonary function in 20 healthy, nonsmoking men and women. METHODS: In this double-blind, randomized study of Nicotrol NS versus placebo, we measured serum nicotine concentrations, blood pressure, heart rate, and indices of pulmonary function at timed intervals before and after nasal spray administration of 3 mg of nicotine. RESULTS: A peak serum nicotine concentration of 4.71 +/- 3.16 ng/mL occurred 10 minutes after drug administration. The maximum change in systolic blood pressure occurred 5 minutes after dosing and was significantly related to nicotine administration (7.1 +/- 9.4% for the nicotine group vs -1.6 +/- 7.3% for the placebo; P = 0.03). In contrast, neither diastolic blood pressure (P = 0.8) nor heart rate (P = 0.07) changed significantly after nicotine administration, when compared with placebo. Pulmonary function was not altered acutely by a single inhalation of nicotine. Pharmacokinetic modeling revealed a classic one-compartment model in which nicotine is absorbed into the systemic circulation by a zero-order process and eliminated by a first-order process. CONCLUSIONS: In this population of nonsmokers, hemodynamic effects of the nicotine nasal spray were observed shortly after administration and before the peak serum nicotine concentration.

Folate status of elderly women following moderate folate depletion responds only to a higher folate intake.

Dietary Reference Intakes (DRI) for folate for elderly women have been based primarily on data extrapolated from studies in younger women. This study was conducted to provide the first age-specific data in elderly women (60-85 y) from a controlled metabolic study on which to base folate intake recommendations. Subjects (n = 33) consumed a moderately folate-deplete (118 microg/d) diet for 7 wk, followed by repletion diets providing either 200 or 415 microg folate/d as diet plus folic acid (FA) or a combination of FA and orange juice (OJ) for 7 wk (n = 30). Comparisons among and within groups were made for serum folate (SF), RBC folate and plasma total homocysteine (tHcy) concentrations. SF concentrations decreased significantly (P < 0.001) during depletion (65 +/- 15%). Postrepletion, the adjusted SF concentration for subjects consuming 415 microg folate/d was significantly greater (P = 0.003) than for subjects consuming 200 microg folate/d. RBC folate concentrations decreased (P < 0.001) during depletion (21 +/- 10%) and further (P < 0.001) during repletion (5 +/- 14%). During depletion, plasma tHcy concentrations increased significantly (P < 0.001) and an inverse relationship between SF and plasma tHcy concentrations was observed in 94% of subjects (P < 0.001). Reversal of this inverse relationship was significant only for subjects consuming 415 microg folate/d (P < 0.001). Postrepletion, subjects consuming 200 microg folate/d had a significantly higher (P = 0.009) adjusted plasma tHcy concentration than subjects consuming 415 microg folate/d. These data in elderly women indicate that 415 microg/d folate, provided as a combination of diet, FA and OJ, or diet and FA, normalizes folate status more effectively than does 200 microg/d, thus providing age-specific data for future folate intake recommendations.