Jerome A. Boscia

Epidemiology of bacteriuria in an elderly ambulatory population

This study of bacteriuria in elderly (mean age 85 years, range 68 to 103) Jewish subjects of mostly middle and upper class attempted to determine disease prevalence, define the turnover in infected subjects, and assess the relation between functional status and infection. The prevalence of bacteriuria (midstream clean-catch method) was assessed in 373 women and 150 men. It was higher in women (18.2 percent) than in men (6.0 percent) (p less than 0.001) and was more common in functionally impaired nursing home residents (23.5 percent) than in apartment house dwellers (12.1 percent) (p less than 0.01). In longitudinal studies, 260 subjects (184 women and 76 men) had three urine culture surveys at six-month intervals. The cumulative percent infected on at least one survey was high (women 30.4 percent, men 10.5 percent). However, persistence of the same organism on all three surveys was surprisingly infrequent (women 6.0 percent, men 1.3 percent), and the turnover of infected and noninfected subjects was considerable. Persistence of bacteriuria on all three surveys was significantly more common in nursing home residents (13.9 percent) than in apartment house dwellers (3.1 percent) (p less than 0.01). Thus, bacteriuria is common in the elderly and appears related to functional status. However, the turnover of infected and noninfected subjects was high, and surprisingly, persistence was not found in most. The transient nature of bacteriuria in most provides support against the treatment of asymptomatic bacteriuria in the elderly.

Does Asymptomatic Bacteriuria Predict Mortality and Does Antimicrobial Treatment Reduce Mortality in Elderly Ambulatory Women

Asymptomatic bacteriuria, a common problem of the elderly, has been associated with increased mortality in the elderly [1-4], although not all studies have confirmed this finding [5-9]. To reconcile these conflicting results, we did a longitudinal study of urinary tract infection in ambulatory elderly women to evaluate the putative relation between asymptomatic bacteriuria and mortality. We considered resolution of this issue to be important because of the implications for clinical practice. If asymptomatic bacteriuria were shown to be an independent risk factor for mortality and if it could also be shown that eradication of the infection by antimicrobial therapy decreased the risk for death, then screening and antimicrobial treatment of elderly ambulatory women with asymptomatic bacteriuria might be warranted and the cost of identifying and treating such infections might be justified. Conversely, failure to confirm a relation would support the view that programs to screen for bacteriuria would not be justified if their goal was to enhance survival. This report summarizes the findings of our 9-year study to determine whether asymptomatic bacteriuria in elderly ambulatory women is a marker of increased mortality and, if so, whether it is because of an association with other determinants of mortality or because asymptomatic bacteriuria is itself an independent cause, the removal of which might improve longevity. The components of the study were a longitudinal study in elderly ambulatory women to compare mortality in those with and without asymptomatic bacteriuria and a double-blind, controlled clinical trial in which antimicrobial therapy was administered for asymptomatic bacteriuria to assess whether treatment decreases mortality. Methods Participant enrollment and the participating institutions have been described previously [10, 11]. Elderly ambulatory residents of the Philadelphia Geriatric Center and of 21 continuing care retirement communities in the greater Philadelphia metropolitan area who gave informed consent were enrolled in this long-term study of urinary tract infection in the elderly. Enrollment continued throughout the course of the study. Philadelphia Geriatric Center houses about 1000 residents who primarily are Jewish; incomes are higher than the maximum Social Security payment; and congregate living is provided either in an apartment house or in a nursing home. In contrast, the continuing care retirement communities are smaller (bed size range, 108 to 675); incomes are higher; residents are primarily not Jewish; and a higher proportion of residents are fully independent. All female residents were eligible to participate except those with indwelling catheters or those incapable of providing midstream clean-catch urine specimens for culture. Specimens were obtained on enrollment and every 6 months thereafter. The protocol was approved by the appropriate institutional review boards, and informed consent was obtained from the participants or their surrogates. Table 1 shows the study periods and chronology of important study events. Throughout the study, urine cultures were obtained at about 6-month intervals. An observational study to compare mortality of bacteriuric and nonbacteriuric volunteers regardless of treatment status was begun in January 1983 and ended in February 1992. Initially, residents with asymptomatic bacteriuria were identified and followed, but treatment was not given. However, on 10 October 1983, a controlled clinical trial was begun to evaluate whether antimicrobial therapy for asymptomatic bacteriuria decreased mortality; every bacteriuric study participant identified after this date was enrolled in the trial. Mortality among residents who were treated with antimicrobial agents for asymptomatic bacteriuria each time it was present was compared with the mortality of those who received no therapy for their episodes of bacteriuria. At enrollment, participants were assigned to the treatment group or to the control group based on the last digit of an identification number unrelated to the conduct of the study. Urine cultures were read by personnel blinded to the study group assignment. When asymptomatic bacteriuria was identified, participants with even numbers were given antimicrobial therapy according to a defined protocol (see below); those with odd numbers served as controls. From 10 October 1983 to 10 December 1987, controls were given no therapy. Thereafter, on the advice of external consultants, the protocol was changed so that participants not assigned to the active treatment group were given placebo pills in place of no treatment. The placebo pills were identical in appearance to each of the antimicrobial agents used. Thus, after 10 December 1987, volunteers with asymptomatic bacteriuria were given therapy in either the form of antimicrobics or placebo after a new consent was obtained; the volunteers and clinical personnel did not know study group assignments. Table 1. Study Design and Enrollment The methods for collecting first-morning urine and for processing the specimens in our research microbiology laboratory have been previously described [10, 11]. Participants were considered to have asymptomatic bacteriuria on a survey if two urine specimens were culture-positive (105 colony-forming units or more per mL of urine) for the same organism within 2 weeks. From 10 October 1983 through 10 December 1987, residents with asymptomatic bacteriuria who were assigned to receive antimicrobial treatment were given short-course therapy (single dose or 3 days) as follows: trimethoprim, 200 mg in one dose; trimethoprim-sulfamethoxazole, 1 double-strength tablet; cefaclor, 500 mg three times a day for 3 days; amoxicillin, 250 mg three times a day for 3 days; carbenicillin indanyl sodium, four times a day for 3 days; or macrodantin, 100 mg twice a day for 3 days, depending on susceptibility of the infecting organism and history of drug allergy. Participants were considered cured if test-of-cure cultures contained less than 104 colony-forming units/mL of the infecting organism on cultures obtained 5 to 10 days after antimicrobial treatment or placebo or on cultures obtained on the next survey in those receiving no therapy. When positive for the same organism, patients were retreated for 14 days with test-of-cure culture afterward. If the organism differed, reinfection was diagnosed and a single dose or 3-day therapy was used; treatment failures were treated as defined above. Test-of-cure cultures were obtained again after therapy and, if positive for the same organism, participants were treated for 14 days. No treatment was given after failure of a 14-day course or two reinfections after short courses. Controls received no therapy during this period. After 10 December 1987, culture-positive patients were assigned to antimicrobial treatment or placebo. Single-dose therapy was given with trimethoprim, 200 mg, or norfloxacin, 400 mg, depending on the susceptibility of the organism; the same drugs (trimethoprim, 100 mg twice daily, and norfloxacin, 400 mg twice daily) were used for 14 days of therapy in patients failing single-dose therapy. Single-dose therapy was used for reinfection. The placebo pills and regimen given to a placebo recipient matched the regimen administered to the participant in the active treatment group who was treated most recently (for example, if the active treatment participant received short-course trimethoprim followed by 14 days of trimethoprim, the next placebo participant received short-course trimethoprim placebo followed by 14 days of trimethoprim placebo). Symptomatic infections were managed by the participants personal physician or by physicians associated with the facility in which the patient lived. Reports were received on an annual or semiannual basis from the participating institutions that detailed changes in their census. All deaths were noted, and registry coordinators reviewed available documents to confirm each death. After 1 September 1987, detailed functional and mental status assessments were done when persons were newly enrolled into the study or were seen for an annual follow-up visit using techniques previously described [11, 12]. Self-care activities of daily living were assessed by a modification of the Multilevel Assessment Instrument [13], and mental status was assessed using a modified version of the Kahn and Goldfarb questionnaire [14]. A subjective measure of global health status (scale, 1 to 4) was based on responses to the question: How do you rate your health: excellent (score 1), good, fair, and bad or poor (score 4)? Diagnoses recorded in the persons medical record were extracted and provided a more objective measure of health status; they were coded according to the ICD-9-CM three-digit codes [15]. The Geriatric Depression Scale [16] was used to assess depressive symptoms, and walking ability was assessed on a scale of 1 (specialized help needed) to 3 (help not needed) [11]. Statistical Analysis Observational Study These analyses compared residents with asymptomatic bacteriuria with residents who did not have asymptomatic bacteriuria on any of the urine culture surveys done during the period of their participation. For the purposes of the survival analyses in the observational study, the results of urine cultures were considered a time-dependent covariate. Accordingly, participants were considered in the ever-positive group once asymptomatic bacteriuria was identified, and all subsequent time on study was contributed to the group with positive cultures regardless of urine culture results on subsequent surveys. Persons entering the study with a negative urine culture were considered in the never-positive group until asymptomatic bacteriuria was identified. Thus, a person who entered the study with negative cultures and who later became culture-positive would have contributed person-time to the follow-up of those in the categ

A lack of association between bacteriuria and symptoms in the elderly

In a study of bacteriuria in elderly (mean age 85 years, range 69 to 101), mostly middle- and upper-class Jewish subjects, attempts were made to determine if bacteriuria without dysuria is otherwise asymptomatic. Seventy-two subjects (59 women and 13 men) without dysuria were questioned about other urinary symptoms (incontinence, frequency, urgency, suprapubic pain, flank pain, fever) and symptoms indicating a lack of well-being (anorexia, difficulty in falling asleep, difficulty in staying asleep, fatigue, malaise, weakness) when they were with and without bacteriuria. Twenty-two subjects had bacteriuria that resolved spontaneously; bacteriuria subsequently developed in 24 nonbacteriuric subjects; and 26 subjects had bacteriuria that resolved with antimicrobial therapy. Subjects occasionally reported urinary symptoms (especially incontinence) and commonly reported symptoms indicating a lack of well-being when they were with and/or without bacteriuria. However, no differences in symptoms were found when bacteriuric subjects were compared with themselves when they were nonbacteriuric. Thus, bacteriuria without dysuria in the elderly appears to be asymptomatic.

Epidemiology of Asymptomatic Bacteriuria in Elderly Women

We studied asymptomatic bacteriuria in elderly ambulatory women residents without indwelling catheters in self‐contained apartment houses at the Philadelphia Geriatric Center (PGC), in the nursing home at PGC, and in several life‐care communities (LCC). Subjects were studied every 6 months from January 1983 through January 1989, and since enrollment was continuous some participated in more surveys than others. PGC residents were middle class and lived either in a self care apartment house (CL) or nursing home (NH); LCC residents were middle or upper class. Antimicrobial therapy for asymptomatic bacteriuria was not given by the study team.

Pyuria and Asymptomatic Bacteriuria in Elderly Ambulatory Women

Excerpt Bacteriuria is common in elderly persons (1-4) and is usually asymptomatic (5). In the absence of symptoms, pyuria is the only readily available way to differentiate urinary infection with ...

Asymptomatic Bacteriuria in Elderly Persons: Treat or Do Not Treat?

Excerpt Bacteriuria occurs with much greater frequency in elderly persons than it does in younger persons. In young to middle-aged women and men, the prevalence is less than 5% and 0.1%, respective...

Rosiglitazone and Hepatic Failure

TO THE EDITOR: In this issue, Forman and colleagues (1) report on a patient who developed severe liver dysfunction while receiving rosiglitazone. Although we have not had the opportunity to review the manuscript of this case report in advance of publication, we would like to point out some important facts about the case. Our company was informed of this patients clinical status at the time the events occurred. We immediately requested, and were able to obtain, entire medical records from both the community hospital where the patient initially presented and from the university hospital where the patient was transferred. These records were extensively reviewed by us and other physicians at SmithKline Beecham Pharmaceuticals and were also sent out to three highly respected hepatologists who have particular expertise in drug-induced liver disease: Neil Kaplowitz (Los Angeles, California), James Lewis (Washington, D.C.), and Paul Watkins (Chapel Hill, North Carolina). These hepatologists independently concluded that this patients liver injury was probably the result of ischemia and not rosiglitazone. Among the many observations that support this conclusion, it should be noted that the patient had significant valvular heart disease, chronic atrial fibrillation, congestive heart failure, and a history of vascular disease previously requiring coronary artery bypass graft surgery and endarterectomy. On admission to the community hospital, he was noted to have a junctional rhythm, and the admitting physician commented that the patient was showing signs of peripheral hypoperfusion. The patient was admitted to the intensive care unit at this hospital, and an arterial line and Swann-Ganz catheter were inserted. The patients course in the intensive care unit at the initial hospital was complicated by hypotension (blood pressure, 76/59 mm Hg, with a cardiac output of 3.24 L/min) and hypoxia (Po 2, 48 mm Hg). In addition, the pattern and time course of biochemical abnormalities are characteristic of ischemic hepatitis, particularly the decrease in serum aspartate aminotransferase level from greater than 11 000 U/L to normal within 9 days. Such high and rapidly normalizing serum aminotransferase values are unusual for most cases of drug-induced liver disease and have not been characteristic of troglitazone (2). Indeed, our consultants each stated that they would also believe that the liver injury was probably due to ischemia even if this patient had been receiving troglitazone. In the rosiglitazone clinical trials program, the rate of liver chemistry abnormalities did not differ between patients receiving rosiglitazone and those receiving placebo. To date, rosiglitazone has been prescribed to more than 150 000 patients. Although we acknowledge the natural deficiencies of postmarketing reporting, the controlled clinical trials experience is thus far predictive of the rosiglitazone safety experience in the marketplace.

Evaluation of initiating a hepatitis B vaccination schedule with one vaccine and completing it with another

The purpose of this study was to determine if a hepatitis B vaccination schedule initiated with one recombinant DNA vaccine could be completed with another. Forty-eight adults on a hepatitis B vaccination schedule of 0, 1 and 6 months had received their first two doses with Merck Sharpe and Dohmes recombinant DNA (MSD rDNA) vaccine (Recombivax HB) at its adult dose of 10 micrograms. At month 6, the subjects were randomized to receive SmithKline Beechams recombinant DNA (SB rDNA) vaccine (Engerix-B) at its adult dose of 20 micrograms or MSD rDNA vaccine. Just prior to the third dose of SB rDNA or MSD rDNA vaccine, the geometric mean anti-HBs titres (GMT) were 161 and 168 mIU ml-1 for the two groups, respectively. The GMT at month 7 were 4077 and 2654 mIU ml-1 for those who had received SB rDNA or MSD rDNA vaccine, respectively. This study demonstrates that a hepatitis B vaccination schedule initiated with MSD rDNA vaccine can be completed with SB rDNA vaccine.

LACK OF ASSOCIATION BETWEEN MEDICATION USE AND THE PRESENCE OR ABSENCE OF BACTERIURIA IN ELDERLY WOMEN

This study was undertaken to determine if there is an association between medication use and the presence or absence of bacteriuria in elderly ambulatory women. Of 198 women who participated in three urine culture surveys (every 6 months) during the 18‐month study period, 66 (34.4%) had bacteriuria on at least one survey. Both univariate and multivariate analyses for the demographics, age, place of residence, and medication use (by drug class) revealed that only place of residence had a significant association with the presence or absence of bacteriuria. In this regard, bacteriuric subjects more commonly resided in the nursing home and less commonly lived in the apartment‐house complex compared with nonbacteriuric subjects (P < .05). Therefore, this study demonstrates that in elderly ambulatory women, medication use does not appear to be associated with the presence or absence of bacteriuria.

Disseminated Multiple Antibiotic-Resistant Gonococcal Infection: Needed Changes in Antimicrobial Therapy

Excerpt Neisseria Gonorrhoeaeusually causes a localized mucosal infection. However, in 1% to 3% of cases, the organism disseminates and produces various clinical syndromes (1). Two of the commoner ...