Colleen M. Brensinger

Genetic warfarin dosing: tables versus algorithms.

OBJECTIVES The aim of this study was to compare the accuracy of genetic tables and formal pharmacogenetic algorithms for warfarin dosing. BACKGROUND Pharmacogenetic algorithms based on regression equations can predict warfarin dose, but they require detailed mathematical calculations. A simpler alternative, recently added to the warfarin label by the U.S. Food and Drug Administration, is to use genotype-stratified tables to estimate warfarin dose. This table may potentially increase the use of pharmacogenetic warfarin dosing in clinical practice; however, its accuracy has not been quantified. METHODS A retrospective cohort study of 1,378 patients from 3 anticoagulation centers was conducted. Inclusion criteria were stable therapeutic warfarin dose and complete genetic and clinical data. Five dose prediction methods were compared: 2 methods using only clinical information (empiric 5 mg/day dosing and a formal clinical algorithm), 2 genetic tables (the new warfarin label table and a table based on mean dose stratified by genotype), and 1 formal pharmacogenetic algorithm, using both clinical and genetic information. For each method, the proportion of patients whose predicted doses were within 20% of their actual therapeutic doses was determined. Dosing methods were compared using McNemars chi-square test. RESULTS Warfarin dose prediction was significantly more accurate (all p < 0.001) with the pharmacogenetic algorithm (52%) than with all other methods: empiric dosing (37%; odds ratio [OR]: 2.2), clinical algorithm (39%; OR: 2.2), warfarin label (43%; OR: 1.8), and genotype mean dose table (44%; OR: 1.9). CONCLUSIONS Although genetic tables predicted warfarin dose better than empiric dosing, formal pharmacogenetic algorithms were the most accurate.

Violent injuries among women in an urban area

BACKGROUND Although the rate of death from injuries due to violent acts is much higher among black women than among white women in the United States, little is known about the nature and correlates of violent injuries among black women living in urban areas. METHODS In this case-control study conducted at three emergency departments in one inner-city community (in west Philadelphia), we studied 405 adolescent girls and women who had been intentionally injured and 520 adolescent girls and women (control subjects) who had health problems not related to violent injury. Data were collected by conducting standardized interviews with use of questionnaires and by screening urine for illicit drugs. Individual logistic-regression models were constructed to identify factors associated with violent injuries inflicted by partners and those inflicted by persons other than the partners of the victims. RESULTS The male partners of the injured women were much more likely than the male partners of control subjects to use cocaine (odds ratio, 4.4; 95 percent confidence interval, 2.3 to 8.4) and to have been arrested in the past (odds ratio, 3.1; 95 percent confidence interval, 1.8 to 5.2). Fifty-three percent of violent injuries to the women had been perpetrated by persons other than their partners. Womens use of illicit drugs and alcohol abuse were factors associated with both violence on the part of partners and violence on the part of other persons. Neighborhood characteristics, including low median income, a high rate of change of residence, and poor education, were independently associated with the risk of violent injuries among women. CONCLUSIONS Women in this urban, low-income community face violence from both partners and other persons. Substance abuse, particularly cocaine use, is a significant correlate of violent injuries. Standard Census data may help identify neighborhoods where women are at high risk for such violence and that would benefit from community-level interventions.

A cognitive neuroscience based computerized battery for efficient measurement of individual differences: Standardization and initial construct validation

There is increased need for efficient computerized methods to collect reliable data on a range of cognitive domains that can be linked to specific brain systems. Such need arises in functional neuroimaging studies, where individual differences in cognitive performance are variables of interest or serve as confounds. In genetic studies of complex behavior, which require particularly large samples, such trait measures can serve as endophenotypes. Traditional neuropsychological tests, based on clinical pathological correlations, are protracted, require extensive training in administration and scoring, and leave lengthy paper trails (double-entry for analysis). We present a computerized battery that takes an average of 1h and provides measures of accuracy and speed on 9 neurocognitive domains. They are cognitive neuroscience-based in that they have been linked experimentally to specific brain systems with functional neuroimaging studies. We describe the process of translating tasks used in functional neuroimaging to tests for assessing individual differences. Data are presented on each test with samples ranging from 139 (81 female) to 536 (311 female) of carefully screened healthy individuals ranging in age from 18 to 84. Item consistency was established with acceptable to high Cronbach alpha coefficients. Inter-item correlations were moderate to high within domain and low to nil across domains, indicating construct validity. Initial criterion validity was demonstrated by sensitivity to sex differences and the effects of age, education and parental education. These results encourage the use of this battery in studies needing an efficient assessment of major neurocognitive domains such as multi-site genetic studies and clinical trials.

Liver transplantation for hepatocellular carcinoma validation of present selection criteria in predicting outcome

Appropriate patient selection is crucial in ensuring acceptable outcomes from orthotopic liver transplantation (OLT) for hepatocellular carcinoma (HCC). The United Network for Organ Sharing (UNOS) has elected to prioritize HCC patients for OLT based on criteria of tumor burden. However, it is unclear whether these criteria correlate with outcome, or with the pathobiological features associated with tumor recurrence. Therefore, we analyzed 109 consecutive patients undergoing OLT for HCC at our center, to determine the utility of present selection criteria in predicting outcome. Pathologic tumor staging of the explanted liver was based on the American Tumor Study Group modified tumor node metastases (pTNM) classification system. Multifocality was defined as >4 tumor nodules on explant. Survival analysis was performed using Kaplan‐Meier and Cox proportional hazards regression methods. At a median follow‐up of 18.9 months, the overall mortality was 19% with 15 patients (14%) dying of recurrent HCC. Kaplan‐Meier 1, 3 and 5‐year survival rates were 89.5%, 68%, and 65%, respectively. Recurrence‐free rates of 1, 3, and 5 years were 89%, 75%, and 65%, respectively. On univariate analysis, the factors found to be significantly associated with recurrence of HCC were explant features of macrovascular invasion, tumor size (per centimeter increase), pTNM stage (per 1‐stage increase), and pre‐transplant serum alphafetoprotein (AFP) >300 ng/mL. In defining a threshold level, we found that explant tumor diameter ≥3 cm, and those tumors classified as at least pT3 on pathological examination, were significantly associated with recurrence (P = .01 and .03, respectively). Tumor size on explant was found to be strongly correlated with multifocality (P = .017) and vascular invasion (P = .02). Patients exceeding pathological UNOS criteria were 3.1 times more likely to have recurrence of HCC (P = .03). In conclusion, we found that tumor size appears to be a surrogate marker for negative pathobiological predictors of outcome, i.e., vascular invasion and multifocality. Present UNOS selection criteria for HCC based on tumor burden appear to provide adequate discriminatory power in predicting outcome of OLT. (Liver Transpl 2004;10:911–918.)

Biopsy features are associated with primary symptoms in interstitial cystitis: results from the interstitial cystitis database study ☆

The purpose of this study was to investigate associations between bladder biopsy features and urinary symptoms for patients enrolled in the Interstitial Cystitis Database (ICDB) Study. Bladder biopsies were obtained during baseline screening in the ICDB Study and were evaluated for histopathologic features. Multivariable models for nighttime voiding frequency, urinary urgency, and pain were developed, incorporating biopsy features from the most diseased area of the bladder as predictors, adjusting for significant clinical factors, and clinical center variation. Among 204 interstitial cystitis (IC) patients providing biopsy specimens, cystoscopic pathology findings were not statistically associated (P >0.1) with primary IC symptoms, although the presence of Hunners ulcer (n = 12) was suggestive of increased urinary frequency. Within a multivariable predictive model for nighttime voiding frequency, adjusting for age and minimum volume per void, 4 pathology features were noted: (1) mast cell count in lamina propria on tryptase stain; (2) complete loss of urothelium; (3) granulation tissue in lamina propria; and (4) vascular density in lamina propria on factor VIII (F8) stain were statistically significant (P <0.01). Similarly, in a multivariable model for urinary urgency, minimum volume, and percentage of submucosal granulation tissue remained statistically significant (P <0.01). Finally, the percentage of mucosa denuded of urothelium and the percentage of submucosal hemorrhage remained highly associated (P <0.01) with pain in a multivariable predictive model. The fact that the presence or severity of glomerulations was not selected for any of these predictive models suggests that cystoscopic findings of glomerulations are not predictive of IC symptoms. Furthermore, these results suggest an important role for certain pathologic features in the predictive modeling of IC symptoms.

Hospitalization and mortality rates from peptic ulcer disease and GI bleeding in the 1990s: relationship to sales of nonsteroidal anti-inflammatory drugs and acid suppression medications

OBJECTIVES:Nonsteroidal anti-inflammatory drugs (NSAIDs) can cause peptic ulcer disease and upper GI bleeding. Acid suppression medications effectively treat NSAID-induced ulcers. However, it is unknown what effect the availability of proton pump inhibitors and over-the-counter preparations of NSAIDs and histamine type 2 receptor antagonists have had on population rates of hospitalization and mortality from GI toxicity. This study examines trends in hospitalization and mortality rates from GI toxicity during the 1990s.METHODS:We performed an analysis of secular trends of hospitalization and mortality rates from peptic ulcer disease, upper GI bleeding, and any GI bleeding using data from the National Hospital Discharge Survey, comparing them with sales of NSAIDs, aspirin, and acid suppression medications from 1992 to 1999.RESULTS:From 1992 to 1999, annual rates of hospitalization and mortality per 100,000 population for peptic ulcer disease declined from 205 to 165 and 7.7 to 6.0, respectively; calendar year was negatively correlated with both peptic ulcer disease hospitalization rates (ρ=−0.88, p = 0.007) and mortality rates (ρ=−0.71, p = 0.058). In contrast, these correlations did not reach statistical significance for upper or any GI bleeding (p > 0.1 for all comparisons). Sales of acid suppression medications were negatively correlated with peptic ulcer disease hospitalization rates (ρ=−0.76, p = 0.037) and mortality rates (ρ=−0.83, p = 0.015). Sales of NSAIDs were not positively correlated with hospitalization or mortality rates from peptic ulcer disease or GI bleeding (p > 0.2 for all comparisons).CONCLUSIONS:Despite changing patterns of use of NSAIDs and acid suppression medications during the 1990s, mortality rates from GI bleeding and peptic ulcer disease have been relatively stable, with an apparent decline in hospitalization rates and mortality from peptic ulcer disease. Changing rates of peptic ulcer disease morbidity and mortality were temporally related to increasing sales of antiulcerants but not to change in sales of NSAIDs.

Differences in facial expressions of four universal emotions

The facial action coding system (FACS) was used to examine recognition rates in 105 healthy young men and women who viewed 128 facial expressions of posed and evoked happy, sad, angry and fearful emotions in color photographs balanced for gender and ethnicity of poser. Categorical analyses determined the specificity of individual action units for each emotion. Relationships between recognition rates for different emotions and action units were evaluated using a logistic regression model. Each emotion could be identified by a group of action units, characteristic to the emotion and distinct from other emotions. Characteristic happy expressions comprised raised inner eyebrows, tightened lower eyelid, raised cheeks, upper lip raised and lip corners turned upward. Recognition of happy faces was associated with cheek raise, lid tightening and outer brow raise. Characteristic sad expressions comprised furrowed eyebrow, opened mouth with upper lip being raised, lip corners stretched and turned down, and chin pulled up. Only brow lower and chin raise were associated with sad recognition. Characteristic anger expressions comprised lowered eyebrows, eyes wide open with tightened lower lid, lips exposing teeth and stretched lip corners. Recognition of angry faces was associated with lowered eyebrows, upper lid raise and lower lip depression. Characteristic fear expressions comprised eyes wide open, furrowed and raised eyebrows and stretched mouth. Recognition of fearful faces was most highly associated with upper lip raise and nostril dilation, although both occurred infrequently, and with inner brow raise and widened eyes. Comparisons are made with previous studies that used different facial stimuli.

A PROSPECTIVE STUDY OF INTERSTITIAL CYSTITIS: RESULTS OF LONGITUDINAL FOLLOWUP OF THE INTERSTITIAL CYSTITIS DATA BASE COHORT

PURPOSE We present baseline characteristics and longitudinal profiles of symptoms in the Interstitial Cystitis Data Base study, a prospective cohort study of patients with interstitial cystitis. MATERIALS AND METHODS A total of 637 eligible patients were entered into the study and followed for symptoms of pain, urgency and urinary frequency. Median followup was 31 months. RESULTS More than 90% of patients were white women with a median age of 43 years. Using the overall pain-urgency-frequency score 7% of participants presented with mild, 44% with moderate and 49% with severe symptoms. Severe urgency in 41% of cases and severe 24-hour frequency in 41% were more common than severe pain in 29%. Of the patients 51% reported nighttime frequency of 2 or more voids. Median duration of interstitial cystitis symptoms was 8 years and 68% of participants were previously diagnosed with the condition. The 36% of patients who withdrew from study or were lost to followup were more likely to have had more severe symptoms at baseline. Patterns of change with time suggest initial symptom improvement due to regression to the mean, and an intervention effect associated with the increased followup and care of cohort participants. Although all symptoms fluctuated, there was no evidence of significant long-term change in overall disease severity. CONCLUSIONS Our observations support the clinical observation that interstitial cystitis is a chronic disease and no current treatments have a significant impact on symptoms with time. These results provide a foundation for the design and performance of future clinical trials in interstitial cystitis using these end points in a similar patient population.

Renal function after orthotopic liver transplantation is predicted by duration of pretransplantation creatinine elevation

In patients with recent onset renal insufficiency, the decision to perform combined kidney/liver transplantation (CKLT) vs. orthotopic liver transplantation alone (OLTa) can be difficult. We hypothesized that duration of renal dysfunction may correlate with creatinine elevation after liver transplantation. We retrospectively identified 69 liver transplantation patients with pretransplantation creatinine ≥1.5 mg/dL (53 OLTa, 13 CKLT). Variables analyzed were presence of hepatorenal syndrome, creatinine, Model for End‐Stage Liver Disease score, albumin, age, race, gender, cause of liver disease, diabetes mellitus, hypertension, and history of ascites, spontaneous bacterial peritonitis, variceal bleeding, hepatic encephalopathy, renal replacement therapy (RRT), and transjugular intrahepatic portosystemic shunting. Duration of pretransplantation renal dysfunction was predictive of 6‐ and 12‐month creatinine post‐OLTa. Area under the receiver operating characteristic (ROC) curve for prediction of 12‐month renal insufficiency by renal dysfunction duration was 0.71; optimal duration cutoff was 3.6 weeks. We applied a multivariable model, derived from OLTa patients, to CKLT subjects with definite or possible hepatorenal syndrome. Predicted 12‐month creatinine without renal transplantation was >2.0 mg/dL for each patient. CKLT patients as opposed to OLTa patients had longer duration of renal dysfunction (median, 18.1 vs. 2.7 weeks, P < 0.001), higher creatinine (median 4.0 versus 1.7 mg/dL, P < 0.001), and higher rate of pretransplantation RRT (62% vs. 7%, P < 0.001). Adjusting for baseline characteristics, CKLT patients had lower creatinine than OLTa patients at 6 months (P =0.15) and 12 months (P =0.01) after transplantation. In conclusion, duration, but not cause, of renal dysfunction predicts renal outcome in OLTa recipients. Prospective studies may use duration of renal dysfunction to help identify CKLT candidates. (Liver Transpl 2005;11:1048–1055.)

Dosing algorithms to predict warfarin maintenance dose in Caucasians and African Americans.

The objective of this study was to determine whether warfarin dosing algorithms developed for Caucasians and African Americans on the basis of clinical, environmental, and genetic factors will perform better than an empirical starting dose of 5 mg/day. From April 2002 through December 2005, 259 subjects (Caucasians and African Americans) who started using warfarin were prospectively followed until they reached maintenance dose. The Caucasian algorithm included 11 variables (R2 = 0.43). This model (which predicted 51% of the doses to within 1 mg of the observed dose) performed better than 5 mg/day (which predicted 29% of the doses to within 5 ± 1 mg). The African‐American algorithm included 10 variables (R2 = 0.28). This model predicted 37% of the doses to within 1 mg of the observed dose, representing a small improvement compared with 5 mg/day (which predicted 34% of the doses to within 1 mg of 5 mg/day). These results were similar to the results we obtained from testing other published algorithms. The dosing algorithms explained <45% of the observed variability in Caucasians, and the algorithms performed only marginally better for African Americans when compared with giving 5 mg empirically.