Predictors of hepatitis B screening among patients with cancer

180

Background: Patients with hepatitis B virus (HBV) infection are at risk of reactivation after chemotherapy. It is unclear whether all patients or only those with certain risk factors should be screened for HBV before chemotherapy. The purpose of our study was to determine the clinical predictors of HBV screening in a single institution.

Methods: In this retrospective, cross-sectional study, we evaluated new patients who received chemotherapy between 1/1/04 and 9/30/07. We collected data on patients' demographics, types of cancer and chemotherapy, and HBV risk factors such as a previous ICD-9 code for hepatitis C or other liver conditions. We searched for HBV screening as defined by an HBsAg or anti-HBc test ordered prior to chemotherapy. In univariate analyses, we examined the association between each of the patient-related variables and HBV screening. We then determined predictors of HBV screening in a multivariable logistic regression model.

Results: We found 10,729 patients who had chemotherapy during the study period. Overall, 16.7% had HBV screening. All of the following predictors in the univariate analysis were significant for HBV screening at the p<0.01 level: age, gender, ethnicity, U.S. residence, cancer type, chemotherapy type, and having a HBV risk factor. In multivariable logistic regression examining predictors for screening using HBsAg, we found that Asian ethnicity (OR 1.75; 1.16-2.66), hematologic malignancies (OR 22.65; 19.3-26.31), and the receipt of rituxamab (OR 3.71; 3.03-4.48) were significant predictors at the p<0.01 level. Women were less likely to be screened than men (OR 0.68; 0.59-0.78). We repeated the analyses using the anti-HBc screening test, and the results were similar. Overall, 1.5% of the screened patients had a positive HBV test.

Conclusions: We found that ethnicity, cancer types, and chemotherapy drugs predict physician-driven HBV screening. Future research is needed to study the predictors of a positive HBV screening test and reactivation of HBV after chemotherapy.

[Table: see text]

The impact of pharmacist-managed oral anticoagulation therapy in older veterans

BACKGROUND

Older adults frequently have conditions requiring oral anticoagulation. Although clearly benefiting from oral anticoagulation, they are at increased risk for bleeding complications. Regular monitoring to optimize anticoagulation and to reduce the chance of major bleeding complications is required. The impact of oral anticoagulation monitoring by pharmacists in patients older than 75 years of age has not been described well.

OBJECTIVE

To compare warfarin therapy prescribed and monitored by physicians to a pharmacist-monitored anticoagulation service in a cohort of older veterans.

METHODS

Retrospective chart review utilizing the Houston VA Medical Center's pharmacy database. Among all outpatients aged 75 years or older filling warfarin prescriptions between 1 March 2003 to 1 March 2005, and who were either monitored in a pharmacist's clinic or not, 103 patients per group were randomly selected. Information on demographics, indication for and length of warfarin therapy, INR values, and thromboembolic and bleeding events were abstracted. Differences were analysed using chi-squared test, Fisher's Exact test, and unpaired Student t-test.

RESULTS

A total of 1521 patients (440 in the pharmacist-monitored group, 1081 in the traditionally monitored group) met our inclusion criteria. One hundred and three patients per group were randomly selected for chart review. Although no significant difference in percentage of therapeutic INR values (48.1% pharmacist group, 46.4% conventional group) or in the incidence of major bleeding events was found, thromboembolic events occurred significantly less frequently in the pharmacist-monitored group (2 events vs. 12 events, P = 0.01). Minor bleeding events were more frequent in the pharmacist-monitored group (50 vs. 17, P < 0.01). However, time to follow-up after a sub- or supra-therapeutic INR was significantly shorter in the pharmacist monitored group (22 days vs. 68 days, and 14 days vs. 32 days, respectively).

CONCLUSION

Pharmacist-monitored anticoagulation was associated with reduced thromboembolic events, an increase in minor bleeding events, and no difference in major bleeding events. Overall such monitoring by pharmacists should be recommended for older adults.

Antiretroviral use and other risks for HIV-associated neuropathies in an international cohort

OBJECTIVE

To explore the association between specific nucleoside reverse transcriptase inhibitors and sensory neuropathies (SNs) and define the modifying roles of hepatitis C (HCV), vitamin B12 deficiency, and impaired glucose tolerance.

METHODS

The authors conducted a prospective cohort study of 147 HIV-infected adults at two sites chosen to emphasize demographic differences. Standardized assessments included detailed antiretroviral histories, neurologic examinations, skin biopsies for epidermal nerve quantitation, and quantitative sensory testing.

RESULTS

There were significant differences between subjects at Johns Hopkins University (JHU) and Monash University (MU) in gender, race, HIV transmission route, and HCV seroprevalence. Symptomatic SN was present in 49% at JHU and 55% at MU (chi2 = 4.02, p = 0.134) and was significantly more common in those at least age 40 than younger patients (odds ratio [OR] = 2.87, 95% CI = 1.27, 6.49). After adjusting for site, age, and CD4 cell count, exposure to didanosine (ddI) or stavudine (d4T) was associated with an significantly increased likelihood of symptomatic SN (ddI: OR = 3.21, 95% CI: 1.56, 6.60; d4T: OR = 7.66, 95% CI: 2.89, 20.33). Plasma HIV RNA, lactate, and HCV were not associated with SN. Quantitative vibratory testing identified neuropathy with a positive predictive value of 76% and epidermal nerve fiber densities 59%.

CONCLUSIONS

Exposure to stavudine and didanosine was significantly associated with a heightened risk for symptomatic sensory neuropathy. Reduced vibration thresholds and epidermal nerve fiber densities had the highest diagnostic efficiency of the laboratory indicators of neuropathy examined, but were relatively insensitive in isolation.

Position-dependent protein mutant profile based on mean force field calculation

The application of the mean force field in protein mutant stability prediction is explored. Based on protein main chain characteristics, including polar fraction, accessibility and dihedral angles, the mean force field was constructed to evaluate the compatibility between an amino acid residue and its environment, from which a position-dependent protein mutant profile was constructed. At each position along a protein sequence, the native residue was replaced by the other 19 types of amino acid residues. The matches were evaluated by energies from mean force field calculation, from which a mutant profile along the protein sequence was derived. General characteristics of such a profile were analyzed. Mutant stabilities for two sets of mutants in two proteins were found to be reasonable compared with experimental data, which indicates that the present method can act as a guide in protein engineering and as an effective scoring matrix in protein sequence-structure alignment studies.