Data quality in trauma transfusion studies and the impact of missing data on predicting massive transfusion

Maximizing data quality may be especially difficult in trauma-related clinical research. Strategies are needed to improve data quality and assess the impact of data quality on clinical predictive models. This study had two objectives. The first was to compare missing data between two multi-center trauma transfusion studies: a retrospective study (RS) using medical chart data with minimal data quality review and the PRospective Observational Multi-center Major Trauma Transfusion (PROMMTT) study with standardized quality assurance. The second objective was to assess the impact of missing data on clinical prediction algorithms by evaluating blood transfusion prediction models using PROMMTT data. RS (2005-06) and PROMMTT (2009-10) investigated trauma patients receiving ≥ 1 unit of red blood cells (RBC) from ten Level I trauma centers. Missing data were compared for 33 variables collected in both studies using mixed effects logistic regression (including random intercepts for study site). Massive transfusion (MT) patients received ≥ 10 RBC units within 24h of admission. Correct classification percentages for three MT prediction models were evaluated using complete case analysis and multiple imputation based on the multivariate normal distribution. A sensitivity analysis for missing data was conducted to estimate the upper and lower bounds of correct classification using assumptions about missing data under best and worst case scenarios. Most variables (17/33=52%) had <1% missing data in RS and PROMMTT. Of the remaining variables, 50% demonstrated less missingness in PROMMTT, 25% had less missingness in RS, and 25% were similar between studies. Missing percentages for MT prediction variables in PROMMTT ranged from 2.2% (heart rate) to 45% (respiratory rate). For variables missing >1%, study site was associated with missingness (all p≤0.021). Survival time predicted missingness for 50% of RS and 60% of PROMMTT variables. MT models complete case proportions ranged from 41% to 88%. Complete case analysis and multiple imputation demonstrated similar correct classification results. Sensitivity analysis upper-lower bound ranges for the three MT models were 59-63%, 36-46%, and 46-58%. Prospective collection of ten-fold more variables with data quality assurance reduced overall missing data. Study site and patient survival were associated with missingness, suggesting that data were not missing completely at random, and complete case analysis may lead to biased results. Evaluating clinical prediction model accuracy may be misleading in the presence of missing data, especially with many predictor variables. The proposed sensitivity analysis estimating correct classification under upper (best case scenario)/lower (worst case scenario) bounds may be more informative than multiple imputation, which provided results similar to complete case analysis.^

RESPIRATORY COMPENSATION MECHANISMS IN THE UPPER AND LOWER RESPIRATORY TRACTS OF AN ANIMAL MODEL AFTER SUBCHRONIC INHALATION EXPOSURE TO FORMALDEHYDE: IMPLICATIONS FOR TOXICOLOGY RISK ASSESSMENT (DEPOSITION, DOSE RECEIVED, RESPONSE)

The potential for significant human populations to experience long-term inhalation of formaldehyde and reports of symptomatology due to this exposure has led to a considerable interest in the toxicologic assessment of risk from subchronic formaldehyde exposures using animal models. Since formaldehyde inhalation depresses certain respiratory parameters in addition to its other forms of toxicity, there is a potential for the alteration of the actual dose received by the exposed individual (and the resulting toxicity) due to this respiratory effect. The respiratory responses to formaldehyde inhalation and the subsequent pattern of deposition were therefore investigated in animals that had received subchronic exposure to the compound, and the potential for changes in the formaldehyde dose received due to long-term inhalation evaluated. Male Sprague-Dawley rats were exposed to either 0, 0.5, 3, or 15 ppm formaldehyde for 6 hours/day, 5 days/week for up to 6 months. The patterns of respiratory response, deposition and the compensation mechanisms involved were then determined in a series of formaldehyde test challenges to both the upper and to the lower respiratory tracts in separate groups of subchronically exposed animals and age-specific controls (four concentration groups, two time points). In both the control and pre-exposed animals, there was a characteristic recovery of respiratory parameters initially depressed by formaldehyde inhalation to at or approaching pre-exposure levels within 10 minutes of the initiation of exposure. Also, formaldehyde deposition was found to remain very high in the upper and lower tracts after long-term exposure. Therefore, there was probably little subsequent effect on the dose received by the exposed individual that was attributable to the repeated exposures. There was a diminished initial minute volume response in test challenges of both the upper and lower tracts of animals that had received at least 16 weeks of exposure to 15 ppm, with compensatory increases in tidal volume in the upper tract and respiratory rate in the lower tract. However, this dose-related effect was probably not relevant to human risk estimation because this formaldehyde dose is in excess of that experienced by human populations. ^

Carcinoma of the lower uterine segment: A newly described association with Lynch Syndrome

Purpose. We performed a case-comparison study to describe the characteristics of LUS tumors and their association with risk factors for endometrial cancer. ^ Patients and Methods. From January 1996 through October 2007, 3,892 women were identified with a diagnosis of primary endometrial carcinoma or primary cervical adenocarcinoma. Pathology records from the 1,009 women who had a hysterectomy were reviewed. Subjects were included in the LUS group only if the tumor was clearly originating from the area between the lower corpus and upper cervix in the hysterectomy specimen. The LUS group was compared to all patients with endometrial corpus carcinoma who underwent hysterectomy at our institution in a 12-month period randomly selected from the study period. Risk factors for endometrial carcinoma such as body mass index (BMI) and Lynch Syndrome were assessed. Expression of estrogen receptor (ER), vimentin, carcinoembryonic antigen (CEA), p16, and human papilloma virus DNA (HPV DNA) was assessed; this panel is known to be effective in distinguishing adenocarcinomas of endometrial versus endocervical origin. Fisher's Exact, Chi-square, Mann-Whitney, and Student's t-tests were utilized for statistical analysis. ^ Results. Thirty-five of 1,009 women had endometrial carcinoma of the LUS (3.5%; 95% CI: 2–4%). Compared to patients with corpus tumors, LUS patients were younger (54.2 vs. 62.9 years, P = .001), had higher stage (P < .001), and more invasive tumors (P = .001). Preoperative diagnosis of the LUS tumors more frequently included the possibility of endocervical adenocarcinoma ( P < .001), leading to preoperative radiation therapy in 4 patients. Median BMI was similar in the LUS and corpus groups. Seventy-three percent of the available LUS tumors had a similar immunohistochemical expression pattern to conventional endometrioid adenocarcinoma. Because of the young median age for the LUS group, we performed immunohistochemistry for Lynch syndrome-associated DNA mismatch repair proteins MLH1, MSH2, MSH6, and PMS2. Microsatellite instability testing (MSI) and MLH1 promoter hypermethylation were performed when indicated. Thirty-six percent of the LUS tumors were MSI-high. Ten of thirty-five (29%) women with LUS tumors were either confirmed to have Lynch Syndrome or were strongly suspected to have Lynch Syndrome based on tissue-based molecular assays (95% CI, 16 to 45%). ^ Conclusions. Endometrial carcinoma arising in the LUS is a clinical and pathologic entity which can be diagnostically confused with cervical adenocarcinoma. In general, LUS tumors can be correctly identified as being endometrial carcinoma using the immunohistochemical panel noted above. The prevalence of Lynch Syndrome in patients with LUS tumors is much greater than that of the general endometrial cancer population (1.8%) or in endometrial cancer patients younger than 50 years of age (8–9%). Based on our results, the possibility of Lynch Syndrome should be considered in women with LUS tumors. ^