In This Issue of Archives of Internal Medicine FREE

Meiland et al examined the association between Escherichia coli bacteriuria and a decline in renal function or the development of end-stage renal failure in a population of generally healthy adult women during 12 to 14 years of follow-up. No differences in the mean ± SD creatinine clearances were found between the 2 groups; for women with vs without bacteriuria, it was 87 ± 21 (1.5 ± 0.4 mL/s) and 85 ± 18 mL/min (1.4 ± 0.3 mL/s), respectively. In the nested case-control analysis, the prevalence of E coli bacteriuria was 14% among both cases (renal failure) and controls. The odds ratio for the development of renal failure in the presence of E coli bacteriuria at baseline, corrected for age, was 1.1 (95% confidence interval, 0.4-2.8; P = .86).

See page 253

The influence of patient adherence with warfarin on anticoagulation control has not been well quantified. Kimmel et al used Medication Event Monitoring System medication bottle caps in a prospective cohort study of 136 patients treated with warfarin to quantify the association between adherence and anticoagulation control. Participants with greater than 20% missed bottle openings (1-2 missed days each week) or greater than 10% extra pill bottle openings (about 1 extra dose per week) had a statistically significant increase in the odds of underanticoagulation or overanticoagulation, respectively. This level of poor adherence was observed among 40% of patients. These results indicate that patients have substantial difficulties maintaining adequate adherence with warfarin and that this poor adherence can have a significant effect on anticoagulation control.

See page 229

This multicenter survey involving 323 resident and staff physicians examined the relationship between internal medicine consultants and referring physicians. Specialty-dependent differences were found in consult preferences of physicians. These differences varied from the extremes of orthopedic surgeons desiring a comprehensive comanagement approach with the consultant to general internists and family medicine physicians desiring to retain control over order writing and to have a more focused consultant approach.

See page 271

This study involves a large gastrointestinal (GI) bleeding cohort of almost 2000 patients, in whom the authors have shown that a value of 0 or 1 on an easy-to-use tool (the modified Blatchford risk score [mBRS]) has a high negative predictive value for rebleeding and death, as well as for high-risk stigmata on endoscopy. Unlike other GI bleeding risk scores (eg, Rockall score), one does not need endoscopy to complete the mBRS score, which comprises simple factors such as hemoglobin, vital signs, comorbidities, and presence of melena. It can therefore potentially be used on the frontlines by nonendoscopists, especially after-hours, to consider discharge pending urgent outpatient endoscopy in approximately 10% of patients. It also performs much better than the clinical (nonendoscopic) component of the conventional Rockall score for this purpose. Specifically, an mBRS of 1 or lower was associated with lower rebleeding (5% vs 19%; P<.001) and mortality (0.5% vs 5.8%; P = .003) and was significant in a multivariate analysis (correcting for medical and endoscopic therapies) for both rebleeding (odds ratio [OR], 0.2; 95% confidence interval [CI], 0.1-0.5) and mortality (OR, 0.1; 95% CI, 0.02-0.9). High-risk stigmata were also less frequent when the mBRS was 1 or lower (17% vs 33%; OR, 0.4; 95% CI, 0.3-0.6). Even the patients with a low mBRS who had apparently been documented to have high-risk stigmata had a low rebleeding rate (3%) and, consequently, a lower apparent benefit from endoscopic therapy.

See page 265

Glazer et al examined compliance with antithrombotic guidelines among patients with newly detected atrial fibrillation and identified patient characteristics associated with warfarin use. In a population-based study of 572 patients with incident atrial fibrillation, 73% received antithrombotic therapy, yet 40% of the patients with atrial fibrillation at high risk for stroke did not receive warfarin. Atrial fibrillation classification, rather than stroke risk factors, was strongly associated with warfarin use.

See page 246