A computerized reminder system to increase the use of preventive care for hospitalized patients.

BACKGROUND: Although they are effective in outpatient settings, computerized reminders have not been proved to increase preventive care in inpatient settings. METHODS: We conducted a randomized, controlled trial to determine the effects of computerized reminders on the rates at which four preventive therapies were ordered for inpatients. During an 18-month study period, a computerized system processed on-line information for all 6371 patients admitted to a general-medicine service (for a total of 10,065 hospitalizations), generating preventive care reminders as appropriate. Physicians who were in the intervention group viewed these reminders when they were using a computerized order-entry system for inpatients. RESULTS: The reminder system identified 3416 patients (53.6 percent) as eligible for preventive measures that had not been ordered by the admitting physician. For patients with at least one indication, computerized reminders resulted in higher adjusted ordering rates for pneumococcal vaccination (35.8 percent of the patients in the intervention group vs. 0.8 percent of those in the control group, P<0.001), influenza vaccination (51.4 percent vs. 1.0 percent, P< 0.001), prophylactic heparin (32.2 percent vs. 18.9 percent, P<0.001), and prophylactic aspirin at discharge (36.4 percent vs. 27.6 percent, P<0.001). CONCLUSIONS: A majority of hospitalized patients in this study were eligible for preventive measures, and computerized reminders significantly increased the rate of delivery of such therapies.

Severe pneumonia. When and why to hospitalize.

Relatively simple objective criteria are now available to predict which patients are at risk for bad outcomes from community-acquired pneumonia. In general, these include older patients and those with certain coexisting illnesses (especially neoplastic disease) or findings of altered mental status, hypotension, severe tachycardia, tachypnea, fever, acidemia, azotemia, hypoxemia, hyperglycemia, anemia, or hyponatremia. The major causes of severe pneumonia are S pneumoniae, H influenzae, and L pneumophila. Less common causes include mixed aerobic and anaerobic mouth flora, as well as M pneumoniae, C pneumoniae, gram-negative bacilli, and S aureus. Specific diagnosis is hampered by a lack of reliable diagnostic tests, but Gram's stain of expectorated sputum and cultures of sputum and blood may occasionally be helpful. Many empirical treatment regimens have been recommended, including those of the American Thoracic Society and the Infectious Diseases Society of America, which are reviewed here. It is hoped that better diagnostic tools will permit future targeting of microbes with narrow-spectrum therapy to diminish the risk of selection of resistant strains with empirical regimens.

Parenteral followed by oral ofloxacin for nosocomial pneumonia and community-acquired pneumonia requiring hospitalization.

We report a multicentric, open trial of intravenous followed by oral ofloxacin, 400 mg every 12 h, as therapy for 100 cases of nosocomial pneumonia and community-acquired pneumonia requiring hospitalization. The typical subject was 57 yr old, and underlying diseases, such as chronic obstructive pulmonary diseases (COPD), diabetes mellitus, and congestive heart failure, were common. For 10 subjects previous therapy had failed. There were 118 pathogens isolated in blood or sputum; S. pneumoniae was the most common (42), followed by H. influenzae (13), Klebsiella spp. (11), and S. aureus (10). Ofloxacin was administered for an average of 5.7 days intravenously followed by 6.9 days orally. Response to therapy was judged to be cure in 71 subjects, improvement in 24, and failure in 5. Among the more seriously ill subjects, ofloxacin therapy was successful for four of five immunocompromised subjects, for 12 of 12 subjects with nosocomial pneumonia, three of whom were on the ventilator, and for nine of 10 subjects with community-acquired pneumonia and bacteremia, including seven of eight cases due to S. pneumoniae. Univariate risk factor analysis revealed underlying COPD and/or tachypnea upon admission to be associated with failure of ofloxacin therapy, with bacteremia suggestive of failure. Conversely, ofloxacin was equally effective in cases in whom previous therapy failed and in cases of nosocomial pneumonia, multilobar pneumonia, and/or pneumonia due to S. pneumoniae. Results for P. aeruginosa were inconclusive. Intravenous followed by oral ofloxacin was highly effective in many difficult cases of pneumonia.

Evaluation of the L-CLONE Legionella pneumophila Serogroup 1 Urine Antigen Latex Test.

The purpose of this study was to evaluate the L-CLONE Legionella pneumophila Serogroup 1 Urine Antigen Latex Test (Access Medical Systems, Inc., Branford, Conn.) for detection of Legionella antigen in urine. A total of 481 frozen urine samples previously tested by an in-house solid-phase radioimmunoassay (RIA) was thawed and retested by using L-CLONE. Included in this sample were 140 RIA-positive samples from culture-positive or serologically confirmed cases of legionellosis and 341 RIA-negative samples from patients with non-Legionella respiratory disease or bacteriuria. The original RIA test result was accepted as the true value. L-CLONE correctly identified 76 of 140 (54%) known positive samples. False-negative results could not be attributed to a low Legionella antigen concentration or to a Legionella antigen subgroup. L-CLONE correctly identified 252 of 341 (74%) known negative samples. False-positive results were experienced in all groups of negative samples, regardless of the patients' underlying diseases. A total of 141 fresh urine samples was tested; all were Legionella antigen negative by RIA. L-CLONE provided 86% specificity. The sensitivity of the L-CLONE in testing fresh urine samples could not be evaluated because of the lack of Legionella antigen RIA-positive samples.

A controlled clinical trial of E5 murine monoclonal IgM antibody to endotoxin in the treatment of gram-negative sepsis. The XOMA Sepsis Study Group.

OBJECTIVE: To assess the efficacy of adjunctive monoclonal antibody antiendotoxin immunotherapy in patients with gram-negative sepsis. DESIGN: Double-blind, randomized, placebo-controlled trial. SETTING: Thirty-three university-affiliated centers, including Veterans Affairs, community, and municipal hospitals. PATIENTS: Hospitalized adults with signs of gram-negative infection and a systemic septic response. INTERVENTION: Patients were assigned to receive either 2 mg/kg of a murine monoclonal antibody directed against gram-negative endotoxin (E5) or placebo. A second infusion was administered 24 hours later. MAIN OUTCOME MEASURES: Mortality over the 30-day study period, resolution of organ failures, and safety. RESULTS: Four hundred eighty-six patients were enrolled. Three hundred sixteen had confirmed gram-negative sepsis (54% bacteremic, 46% nonbacteremic). The survival difference was not statistically significant for all patients. Among patients with gram-negative sepsis who were not in shock at study entry (n = 137), E5 treatment resulted in significantly greater survival (relative risk, 2.3; P = .01). Resolution of individual organ failures was more frequent among these patients, occurring in 19 (54%) of 35 patients in the E5 group vs eight (30%) of 27 in the placebo group (P = .05). Four reversible allergic reactions occurred among 247 patients (1.6%) receiving E5. No other toxicity was identified. CONCLUSIONS: Treatment with E5 antiendotoxin antibody appears safe. It reduces mortality and enhances the resolution of organ failure among patients with gram-negative sepsis who are not in shock when treated.

Accidental overdose of intravenous ofloxacin with benign outcome.

A patient accidentally received approximately 3 g of ofloxacin intravenously. She experienced only moderately severe central nervous system symptoms, which resolved within 9 h. The peak serum ofloxacin level was 39.3 micrograms/ml, approximately seven times the usual peak level. Ofloxacin may possess a considerable safety margin in humans.

Ofloxacin in community-acquired lower respiratory infections. A comparison with amoxicillin or erythromycin.

Ninety-one patients with community-acquired lower respiratory infections were treated orally in a comparative 10-day trial of ofloxacin versus amoxicillin or erythromycin. Approximately one-half of the patients had no major underlying disease and the other half had some form of chronic lung disease. Pneumonia was present in 31 percent of the patients and the remainder had purulent bronchitis. Bacterial pathogens were recovered from 60 percent of the patients, with Haemophilus influenzae (33 isolates) and Streptococcus pneumoniae (16 isolates) being the most common. Ofloxacin was found to be a safe, well-tolerated therapeutic agent, which was as effective clinically as amoxicillin or erythromycin and with an advantage of less frequent administration. Ofloxacin was more effective than amoxicillin (90 percent versus 75 percent; p = 0.05) in elimination of pathogenic bacteria from lower airway cultures. Caution should be exercised in the use of ofloxacin, at least in short-term treatment regimens, with anaerobic pulmonary infections; additional information is needed for S. pneumoniae given the relatively high minimal inhibitory concentrations for this species.

Histoplasma capsulatum polysaccharide antigen detection in diagnosis and management of disseminated histoplasmosis in patients with acquired immunodeficiency syndrome.

PURPOSE: Disseminated histoplasmosis is a serious and often rapidly progressive, opportunistic infection in patients with acquired immunodeficiency syndrome (AIDS), supporting the importance of rapid diagnostic tests. We investigated Histoplasma capsulatum polysaccharide antigen (HPA) detection, a promising new method for rapid diagnosis of histoplasmosis. PATIENTS AND METHODS: Sixty-one cases of disseminated histoplasmosis in patients with AIDS form the basis of this report. Control cases were patients with AIDS who had other opportunistic infections and whose cultures were negative for H. capsulatum. A slightly modified radioimmunoassay procedure was used to measure the levels of HPA in urine and blood specimens. RESULTS: High levels of HPA were detected in the urine of 59 of 61 (96.7%) and the blood of 37 of 47 (78.7%) patients with AIDS complicated by disseminated histoplasmosis. Treatment with amphotericin B reduced levels of HPA in the urine in 19 of 21 (90.5%) and the serum of all 10 patients tested. HPA levels increased in the urine in all eight and in the serum in all five patients with culture-proven relapse. CONCLUSION: In conclusion, HPA detection offers a rapid method for diagnosing disseminated histoplasmosis. Additional experience is required to establish the role of this test in monitoring the effects of treatment and in identifying relapse in patients with AIDS.

Comparison of sandwich solid-phase radioimmunoassay and two enzyme-linked immunosorbent assays for detection of Histoplasma capsulatum polysaccharide antigen.

Detection of a Histoplasma capsulatum polysaccharide antigen (HPA) has proved a useful approach to diagnosis of histoplasmosis. Two sandwich enzyme-linked immunosorbent assays (ELISAs) using alkaline phosphatase (AP)-or horseradish peroxidase (HRP)-conjugated antibodies were compared with solid-phase radioimmunoassay (RIA) for detection of HPA. The AP-ELISA and HRP-ELISA were each positive in 17 (89.5%) of 19 urine specimens from patients with disseminated histoplasmosis, while the RIA was positive in 18 (94.7%) of 19. Specimens from patients with nondisseminated histoplasmosis were positive by AP-ELISA in 8 of 32, by HRP-ELISA in 4 of 25, and by RIA in 12 of 25. Of control specimens from patients with other fungal infections, the AP-ELISA was negative in 22 (91.7%) of 24, the HRP-ELISA in 23 (95.8%) of 24, and the RIA in 23 (95.8%) of 24. Reproducibilities AP-ELISA and HRP-ELISA were 95.1% and 95.1%, respectively. Thus, AP-ELISA and HRP-ELISA appear less sensitive than RIA and may be falsely negative in specimens containing low levels of HPA.

The microbiology and chemotherapy of odontogenic infections.

This article reviews the complexity of the microbiological environment of odontogenic infections. With an understanding of the organisms involved, the appropriate antibiotic can be chosen. A review of current antibiotic choices and the rationale for their selection is also presented.

Significance of Histoplasma antigen in the cerebrospinal fluid of patients with meningitis.

A radioimmunoassay was previously developed for detection of Histoplasma capsulatum antigen in the blood and urine of patients with disseminated histoplasmosis. In this investigation, cerebrospinal fluid (CSF) specimens from 14 episodes of Histoplasma meningitis occurring in 12 patients were tested by radioimmunoassay. Histoplasma capsulatum antigen was detected in the CSF of five patients. Cerebrospinal fluid cultures were positive for H capsulatum in three of these five patients. Antibodies to H capsulatum were found in nine of the 13 CSF specimens tested. The radioimmunoassay for Histoplasma antigen was also positive in the CSF in one of 11 patients with coccidioidal meningitis but not in 17 patients with cryptococcal meningitis. It was concluded that Histoplasma antigen is present in the CSF of some patients with histoplasmosis and chronic meningitis, but cross-reactions may occur in patients with coccidioidal meningitis.

Immunologic diversity among serogroup 1 Legionella pneumophila urinary antigens demonstrated by monoclonal antibody enzyme-linked immunosorbent assays.

We tested urine specimens from 222 patients with serogroup 1 Legionella pneumophila pneumonia in two enzyme-linked immunosorbent assays (ELISAs) which used different monoclonal antibodies (A and B) as detector antibodies. Of 171 specimens which contained enough antigen to be detected in the ELISAs, 169 reacted in only one of the two assays. A total of 25 patients whose infections were acquired in any of three Indianapolis hospitals excreted antigen reactive with monoclonal antibody B, but 18 patients who were treated for infections acquired elsewhere reacted with monoclonal antibody A. The urinary antigen ELISA reactivity patterns correlated with the reactivity patterns of L. pneumophila isolates when a separate panel of seven monoclonal antibodies was used. The isolate patterns, in turn, correlated well with environmental isolate patterns from two of the hospitals with nosocomial cases. We conclude that at least two different epitopes exist on the antigen molecules in urine from patients with serogroup 1 L. pneumophila pneumonia and that the subtyping of urinary antigens can be useful epidemiologically.

Effect of a targeted antibiotic advisory consult program on hospital antibiotic costs.

The purpose of this 3-year study was to determine whether or not an antibiotic control program, designed by the P & T Committee, would affect the cost of antibiotics to this 473-bed general medical-surgical teaching hospital. Physician consultants from different clinical departments discussed each use of targeted antibiotics with prescribing physicians. Compared with the preprogram year, cost savings of $46,869 during the first program year, and $73,212 during the second program year were achieved. Many factors were taken into account before arriving at these figures, including cost analysis of both targeted and nontargeted antibiotics, impact of price changes, administration costs, and changes in hospital census. The authors conclude that in the absence of cost inflation, antibiotic costs would have risen considerably due to price escalation, but in the face of antibiotic controls, a modest cost savings over the 2 years analyzed was actually realized.

Attitudes of house physicians concerning various antibiotic-use control programs.

Medical and surgical house physicians at three teaching hospitals with different antibiotic control programs (ACPs) were surveyed to determine their attitudes about and preferences for these programs. Questionnaires were mailed to resident physicians who had trained at all three hospitals. One hospital had no antibiotic prescribing policy (open prescribing), one employed an infectious-disease physician consultant who discussed antibiotic orders for certain reserved drugs with prescribing physicians but did not otherwise restrict access to the drugs (educational ACP), and the third hospital required approval of an infectious-disease physician for dispensing of reserved drugs by the pharmacy department (restrictive ACP). The survey response was 77% after one follow-up mailing to nonrespondents. Regardless of hospital type, physicians preferred the educational ACP to either open prescribing or the restrictive ACP. Based on personal experiences, significantly fewer physicians encountered patient-care problems with the educational ACP than with either open antibiotic prescribing or the restrictive ACP. Significantly more physicians perceived that the educational ACP was more beneficial for patient antibiotic therapy and will be more beneficial for future antibiotic prescribing than the restrictive ACP. Most respondents believed that ACPs save hospitals money, that ACPs can be implemented without compromising either patient care or physician performance, and that they would encounter similar ACPs in the future. Educational ACPs should be considered at teaching hospitals with interested infectious-disease consultants.

Development and evaluation of diagnostic radiometric assays for serogroup 4 L. pneumophila urinary antigens.

We previously reported that 80% of patients with serogroup 1 Legionella pneumophila pneumonia excrete detectable quantities of specific antigens in their urine. The purpose of this study was to determine whether specific antigens can be detected in urine from patients with serogroup 4 L. pneumophila pneumonia. Antisera were prepared in 15 rabbits and used to set up 25 solid-phase radioimmunoassays. The best of these assays detected specific antigens in urine from 4 of 5 patients with culture-proven serogroup 4 infections and from 1 of 2 patients with culture-proven serogroup 10 infections. None of 100 control specimens was positive in the assays. The antigens from serogroup 4 and 10 infections behaved differently in the assays, suggesting that they are not identical molecules. This study demonstrates that it is possible to detect specific antigens in the urine of patients with serogroup 4 L. pneumophila pneumonia. This may be a useful method for diagnosing these infections rapidly.

Production of monoclonal antibody against Pneumocystis carinii by using a hybrid of rat spleen and mouse myeloma cells.

It has been difficult to obtain pure Pneumocystis carinii antigen either from cultures or from infected lungs for use in producing a specific antibody against P. carinii. This report describes an approach toward producing a monoclonal antibody that bypasses the antigen purification steps. P. carinii infection was developed in Sprague-Dawley rats by the method of immunosuppression with cortisone. The infected lungs were homogenized, and the homogenate was used to immunize Sprague-Dawley rats. Rat spleen cells were then fused with SP2/0 mouse myeloma cells. Hybridoma clones were screened for antibody production against P. carinii by immunoperoxidase staining techniques and by enzyme-linked immunosorbent assay, using as antigens homogenates of normal rat lung, homogenates of P. carinii-infected rat lung, and harvests of P. carinii grown with WI-38 cells. Out of six hybridoma clones obtained that produced antibodies against P. carinii, one was able to produce ascitic fluid. This monoclonal antibody reacted with two P. carinii antigens with masses of about 35,000 and 65,000 daltons in P. carinii-infected lungs and three proteins with masses of about 35,000, 65,000, and 110,000 daltons in P. carinii that was harvested from a WI-38 cell culture.

Antigen detection for the rapid diagnosis of mycoplasma and Legionella pneumonia.

Immunologic attempts to detect mycoplasma antigens in fluids of infected patients have been rare and largely unsuccessful. Nucleic acid hybridization procedures appear promising on the basis of successes in detecting mycoplasmal contamination of tissue culture cells; results of attempts to apply these techniques to human infections have not been reported. Antigens can be detected in the urine of about 80% of patients with serogroup 1 Legionella pneumophila pneumonia and of some patients with serogroup 4 Legionella pneumophila and Legionella dumoffii pneumonia. The specificity of these assays is greater than 99%. In a test population in which the prevalence of Legionella pneumophila was 4%, the posterior probabilities of positive and negative results of tests for antigen were 86.5% and 99.3%, respectively. Antigen is detectable within the first 3 days of illness approximately as often as at later periods, and antigen may remain detectable for a few days to 1 yr after successful therapy. Antigen is detectable in serum, but the concentrations are considerably lower than in urine. Combining urinary antigen detection with direct fluorescent antibody examination of secretions increases the rapid diagnostic yield by 10%-20%. Monoclonal antibody studies demonstrate that subgroup specificities are present among the serogroup 1 urinary antigens. Radiometric and enzyme immunoassays detect antigen in equal proportions of patients. Latex agglutination results are positive in about 80% of those cases positive by the other methods.

Diagnosis of disseminated histoplasmosis by detection of Histoplasma capsulatum antigen in serum and urine specimens.

The diagnosis of Histoplasma capsulatum infection by serologic testing for the presence of antibodies is limited by a high rate of false positive and false negative results and by the requirement that the patient have a normal immune response. We have developed a radioimmunoassay for the detection of H. capsulatum antigen in urine and serum specimens. Antigenuria was noted in 20 of 22 episodes of disseminated histoplasmosis that occurred in 16 patients, in 6 of 32 patients with self-limited infection, in 2 of 32 patients with cavitary histoplasmosis, and in 4 of 8 patients with a sarcoid-like illness caused by H. capsulatum. The detection of antigen in urine was reproducible in 38 of 41 (93 percent) retests of specimens. H. capsulatum antigen was also detected in the serum during 11 of the 22 episodes of disseminated histoplasmosis, in none of the 12 episodes of other types of histoplasmosis in patients with antigenuria, in 1 of the 33 patients with histoplasmosis who lacked the urinary antigen, and in none of the 50 controls. Antigenemia and antigenuria decreased after initiation of antifungal therapy and recurred in patients who had a relapse. We conclude that this radioimmunoassay for H. capsulatum antigen represents a useful new method for the rapid diagnosis of disseminated histoplasmosis.

Cross-reactive urinary antigens among patients infected with Legionella pneumophila serogroups 1 and 4 and the Leiden 1 strain.

A solid-phase radioimmunoassay was developed to detect antigens of Legionella pneumophila serogroup 4. The assay detected antigen in urine from four of seven patients with pneumonia caused by L. pneumophila serogroup 4, two of two patients with L. pneumophila Leiden 1 strain (proposed new serogroup), and 26 of 55 patients with L. pneumophila serogroup 1 (proven by culture or direct fluorescent-antibody assay). The antigen of two patients with serogroup 4, one with Leiden 1 strain, and 49 with serogroup 1 infections could be detected in a previously described assay for antigens of L. pneumophila serogroup 1. Serogroup specificity of the assays could be demonstrated if serial dilutions of urine were tested. None of 347 urine specimens used as controls were positive in the radioimmunoassay. We conclude that antigens are excreted in the urine of at least some patients with Legionnaires' disease caused by serogroup 4 and Leiden 1 strain. The urinary antigens in patients with serogroup 4 and Leiden 1 strain infections, although different from those in patients with serogroup 1 infections, serologically contain cross-reactive components.

Onset and duration of urinary antigen excretion in Legionnaires disease.

The purposes of this study were to determine whether antigen is excreted by patients with Legionnaires disease early enough after the onset of symptoms to be useful for making therapeutic decisions and whether antigen excretion ends when successful treatment is concluded. Specific antigen was detected in the urine of 14 (88%) of 16 patients with Legionnaires disease during days 1 to 3 of symptoms, 33 (80%) of 41 patients during days 4 to 7, 25 (89%) of 28 patients during days 8 to 14, and 11 of 11 patients after day 14, by solid-phase immunoassays for serogroup 1 Legionella pneumophila antigen. Antigen excretion persisted for 42 days or longer after the onset of treatment in at least 15 patients. The longest documented duration of excretion was 326 days. We conclude that antigen can be detected approximately as often early after symptoms begin as later, allowing meaningful therapeutic decisions to be made, but that prolonged antigen excretion may negate the diagnostic value of urinary antigen detection for relapsing or recurrent L. pneumophila pneumonia.