A US-based national surveillance study for the susceptibility and epidemiology of Clostridioides difficile isolates with special reference to ridinilazole: 2020-2021.

We have previously reported on the susceptibility and epidemiology of Clostridioides difficile isolates from six geographically dispersed medical centers in the United States. This current survey was conducted with isolates collected in 2020-2021 from six geographically dispersed medical centers in the United States, with specific attention to susceptibility to ridinilazole as well as nine comparators. C. difficile isolates or stools from patients with C. difficile antibiotic-associated diarrhea were collected and referred to a central laboratory. After species confirmation of 300 isolates at the central laboratory, antibiotic susceptibilities were determined by the agar dilution method [M11-A9, Clinical and Laboratory Standards Institute (CLSI)] against the 10 agents. Ribotyping was performed by PCR capillary gel electrophoresis on all isolates. Ridinilazole had a minimum inhibitory concentration (MIC) 90 of 0.25 mcg/mL, and no isolate had an MIC greater than 0.5 mcg/mL. In comparison, fidaxomicin had an MIC 90 of 0.5 mcg/mL. The vancomycin MIC 90 was 2 mcg/mL with a 0.7% resistance rate [both CLSI and European Committee on Antimicrobial Susceptibility Testing (EUCAST) criteria]. The metronidazole MIC 90 was 1 mcg/mL, with none resistant by CLSI criteria, and a 0.3% resistance rate by EUCAST criteria. Among the 50 different ribotypes isolated in the survey, the most common ribotype was 014-020 (14.0%) followed by 106 (10.3%), 027 (10%), 002 (8%), and 078-126 (4.3%). Ridinilazole maintained activity against all ribotypes and all strains resistant to any other agent tested. Ridinilazole showed excellent in vitro activity against C. difficile isolates collected between 2020 and 2021 in the United States, independent of ribotype.

Epidemiologic trends in Clostridioides difficile isolate ribotypes in United States from 2011 to 2016.

BACKGROUND: Geographic and temporal trends in the distribution of PCR ribotypes for Clostridioides difficile associated diarrheal isolates obtained in the United States (US) are changing. As part of a US national surveillance program of C. difficile susceptibility to fidaxomicin, we quantified the distribution of PCR ribotypes of stool isolates collected from 2011 to 2016. METHODS: C. difficile isolates or C. difficile toxin + stools from patients with C. difficile infection (CDI) were submitted for testing to Tufts Medical Center from 6 geographically distinct medical centers. Following isolation and confirmation as C. difficile, approximately 35% of the isolates were randomly sampled, stratified by center, for PCR ribotyping by capillary gel electrophoresis. Toxin gene profiling was performed on all isolates. RESULTS: 939 isolates from a total of 2814 (33.4%) isolated over the 6 years were analyzed. Seventy unique ribotypes were observed, including 19 ribotypes observed 10 or more times. Sixteen ribotypes were not previously observed in our data base. Ribotype 027 declined by more than 60% over the 6 years of the survey from 35.3% to 13.1% (p < 0.001). Ribotype 106 was the most common in 2016, followed by 027 and 014-020. There were strong correlations between 027 and binary toxin with the 18 base pair deletion of tcdC and ribotype 078-126 had 100% concordance with the previously described tcdC 39 base pair deletion. CONCLUSIONS: The frequency of ribotypes in the US has changed with a marked decline in 027. Each of the geographical areas had variations which differed from each other, but collectively, these results suggest that the changing epidemiology of C. difficile in the US is consistent with what is being seen in Europe. Continued surveillance and monitoring of changes in ribotype distributions of C. difficile are warranted.

U.S.-Based National Surveillance for Fidaxomicin Susceptibility of Clostridioides difficile-Associated Diarrheal Isolates from 2013 to 2016.

In 2011, we initiated a sentinel surveillance network to assess changes in Clostridioides (formerly Clostridium) difficile antimicrobial susceptibility to fidaxomicin from 6 geographically dispersed medical centers in the United States. This report summarizes data from 2013 to 2016. C. difficile isolates or toxin-positive stools from patients were referred to a central laboratory. Antimicrobial susceptibility was determined by agar dilution. CLSI, EUCAST, or FDA breakpoints were used, where applicable. Toxin gene profiles were characterized by multiplex PCR on each isolate. A random sample of approximately 40% of isolates, stratified by institution and year, was typed by restriction endonuclease analysis (REA). Among 1,889 isolates from 2013 to 2016, the fidaxomicin MIC90 was 0.5 µg/ml; all isolates were inhibited at ≤1 µg/ml. There were decreases in metronidazole and vancomycin MICs over time. Clindamycin resistance remained unchanged (27.3%). An increase in imipenem resistance was observed. By 2015 to 2016, moxifloxacin resistance decreased in all centers. The proportion of BI isolates decreased from 25.5% in 2011 to 2012 to 12.8% in 2015 to 2016 (P < 0.001). The BI REA group correlated with moxifloxacin resistance (BI 84% resistant versus non-BI 12.5% resistant). Fidaxomicin MICs have not changed among C. difficile isolates of U.S. origin over 5 years post licensure. There has been an overall decrease in MICs for vancomycin, metronidazole, moxifloxacin, and rifampin and an increase in isolates resistant to imipenem. Moxifloxacin resistance remained high among the BI REA group, but the proportion of BI isolates has decreased. Continued geographic variations in REA groups and antimicrobial resistance persist.

Comparative Genome Analysis and Global Phylogeny of the Toxin Variant Clostridium difficile PCR Ribotype 017 Reveals the Evolution of Two Independent Sublineages.

The diarrheal pathogen Clostridium difficile consists of at least six distinct evolutionary lineages. The RT017 lineage is anomalous, as strains only express toxin B, compared to strains from other lineages that produce toxins A and B and, occasionally, binary toxin. Historically, RT017 initially was reported in Asia but now has been reported worldwide. We used whole-genome sequencing and phylogenetic analysis to investigate the patterns of global spread and population structure of 277 RT017 isolates from animal and human origins from six continents, isolated between 1990 and 2013. We reveal two distinct evenly split sublineages (SL1 and SL2) of C. difficile RT017 that contain multiple independent clonal expansions. All 24 animal isolates were contained within SL1 along with human isolates, suggesting potential transmission between animals and humans. Genetic analyses revealed an overrepresentation of antibiotic resistance genes. Phylogeographic analyses show a North American origin for RT017, as has been found for the recently emerged epidemic RT027 lineage. Despite having only one toxin, RT017 strains have evolved in parallel from at least two independent sources and can readily transmit between continents.

Susceptibility of Clostridium difficile isolates from a Phase 2 clinical trial of cadazolid and vancomycin in C. difficile infection.

OBJECTIVES: The aim of this study was to evaluate the susceptibilities of Clostridium difficile isolates to cadazolid, a novel antibiotic for the treatment of C. difficile infection. METHODS: Ribotyping and susceptibilities were determined for C. difficile isolates from a multicentre, double-blind, Phase 2 study of oral cadazolid in patients with C. difficile infection (NCT01222702, ClinicalTrials.gov; EudraCT 2010-020941-29, European Clinical Trials Database). Patients were randomized to receive 250, 500 or 1000 mg of cadazolid twice daily or 125 mg of vancomycin four times daily, for 10 days. MICs of cadazolid, vancomycin, fidaxomicin, linezolid and moxifloxacin were determined at baseline for all patients and post-baseline for patients with clinical failure or recurrence, using the agar dilution method. RESULTS: Seventy-eight of 84 patients had an evaluable toxigenic C. difficile isolate at baseline. The most frequent PCR ribotype was 027 (15.4%). Cadazolid MICs for baseline isolates (including epidemic strain 027) ranged from 0.06 to 0.25 mg/L. Baseline cadazolid MICs were similar to those of fidaxomicin and lower than those of vancomycin, linezolid and moxifloxacin. For each clinical outcome group (clinical cure, clinical failure, sustained clinical response and clinical failure or recurrence), the baseline cadazolid MIC range was 0.06-0.25 mg/L. Mean (min-max) cadazolid faecal concentration (µg/g) on day 5 was 884 (101-2710), 1706 (204-4230) and 3226 (1481-12 600) for the doses 250, 500 and 1000 mg, respectively. CONCLUSIONS: For all cadazolid doses, the faecal concentration was in excess of several thousand-fold the MIC90 for C. difficile. The MIC of cadazolid for all C. difficile isolates, including epidemic strains, was low and in the same narrow range regardless of treatment outcome.

U.S.-Based National Sentinel Surveillance Study for the Epidemiology of Clostridium difficile-Associated Diarrheal Isolates and Their Susceptibility to Fidaxomicin.

In 2011 a surveillance study for the susceptibility to fidaxomicin and epidemiology of Clostridium difficile isolates in the United States was undertaken in seven geographically dispersed medical centers. This report encompasses baseline surveillance in 2011 and 2012 on 925 isolates. A convenience sample of C. difficile isolates or toxin positive stools from patients were referred to a central laboratory. Antimicrobial susceptibility was determined by agar dilution (CLSI M11-A8). Clinical and Laboratory Standards Institute (CLSI), Food and Drug Administration, or European Union of Clinical Antimicrobial Susceptibility Testing (EUCAST) breakpoints were applied where applicable. Toxin gene profiles were characterized by multiplex PCR on each isolate. A random sample of 322 strains, stratified by institution, underwent restriction endonuclease analysis (REA). The fidaxomicin MIC90 was 0.5 µg/ml for all isolates regardless of REA type or toxin gene profile, and all isolates were inhibited at ≤1.0 µg/ml. By REA typing, BI strains represented 25.5% of the isolates. The toxin gene profile of tcdA, tcdB, and cdtA/B positive with a tcdC 18-bp deletion correlated with BI REA group. Moxifloxacin and clindamycin resistance was increased among either BI or binary toxin-positive isolates. Metronidazole and vancomycin showed reduced susceptibility (EUCAST criteria) in these isolates. Geographic variations in susceptibility, REA group and binary toxin gene presence were observed. Fidaxomicin activity against C. difficile isolated in a national surveillance study did not change more than 1 year after licensure. This analysis provides baseline results for future comparisons.

Typing and susceptibility of bacterial isolates from the fidaxomicin (OPT-80) phase II study for C. difficile infection.

BACKGROUND: Clostridium difficile infection (CDI) has been increasing in incidence and severity in recent years, coincident with the spread of a "hypervirulent" strain, REA type BI (ribotype 027, PFGE NAP 1). Exacerbating the problem has been the observation that metronidazole may be showing decreased effectiveness, particularly in the more severe cases. Fidaxomicin is an 18-membered macrocycle currently in phase 3 trials for the treatment of C. difficile infection (CDI). An open-label, phase II study in CDI patients has been completed and the clinical results published. C. difficile organisms were isolated from patient stool specimens and typed by restriction endonuclease analysis (REA) in order to determine the frequency and susceptibility of the C. difficile isolates and their response to treatment. METHODS: Fecal samples were plated on CCFA agar for isolation of C. difficile. These isolates were tested for susceptibility to fidaxomicin, vancomycin, and metronidazole using CLSI agar dilution methods and were typed by REA. RESULTS: C. difficile was isolated from 38 of 49 subjects and 16 (42%) were the epidemic C. difficile BI group. The BI strain was distributed approximately equally in the three dosing groups. Overall antibiotic susceptibilities were consistent with the previously reported MIC(90) values for the three antibiotics tested, but the MIC(90) of BI strains was two dilutions higher than non-BI strains for metronidazole and vancomycin (for both antibiotics, MIC(90) was 2 microg/mL vs. 0.5 microg/mL, P<0.01 for metronidazole, P=NS for vancomycin). Clinical cure for BI isolates (11/14, 79%) was not significantly different from non-BI isolates (21/22, 95%). CONCLUSION: These results underscore the high prevalence of the BI epidemic strain and demonstrate that mild to moderate CDI infection as well as severe disease can be caused by these strains. Fidaxomicin cure rates for subjects with BI and with non-BI strains are similar, although the small numbers of subjects preclude a robust statistical comparison.

Treatment of Clostridium difficile-associated diarrhea and colitis.

Treatment of C. difficile diarrhea with metronidazole or vancomycin is highly effective at relieving symptoms. The high rate of diarrhea recurrence is concerning, but fortunately most patients respond to a second course of treatment. The problem of vancomycin resistance in hospital organisms has markedly reduced usage of this agent as a first-line treatment for C. difficile diarrhea, leaving metronidazole as the mainstay of treatment in the United States where teicoplanin and fusidic acid are not marketed. It is likely that any new antimicrobial agent used to treat C. difficile will be similarly plagued by a high rate of recurrence, presumably incurred as a result of disruption of normal bowel flora. There is a need for improved treatment and prevention of this increasingly frequent and debilitating nosocomial infection. Treatments that utilize passive antibodies, immunization, nontoxigenic C. difficile, or other forms of biotherapy may hold the key to improved treatment and prevention of C. difficile disease in the future. In the meantime, it behooves all practitioners to use antimicrobials judiciously in order to prevent as many cases of C. difficile diarrhea as possible.

Staphylococcus aureus bacteriuria.

One hundred twenty-seven episodes of Staphylococcus aureus bacteriuria were reviewed retrospectively in two hospitals to establish the rate of occurrence, clinical importance, and associated predisposing factors. Staphylococcus aureus was an infrequent urinary isolate, and accounted for only about 1% of all positive urine cultures. Although almost all cases in a Veterans Hospital occurred in elderly men, episodes in a community hospital were observed in women and children as well. Patients usually had pyuria (71%) but only 39% had urinary symptoms of fever. Among predisposing factors, serious underlying diseases were uncommon, but urinary tract manipulations or abnormalities were present in nearly two thirds of patients. Fifty-five percent of cases were nosocomial, and 73% of these were associated with urinary catheterization or other invasive urinary tract procedures. Most patients (61%) were not treated for their bacteriuria, and there was a secondary bacteremia rate of 5.5% in the Veterans Hospital. Although infrequently encountered, the presence of S aureus in urine should be treated with at least as much concern as more frequently encountered bacteria.

Fungal meningitis manifesting as hydrocephalus.

Four patients with fungal meningitis and hydrocephalus were treated by placement of intraventricular shunts prior to the diagnosis of infection. As a consequence, they were subjected to the risks of surgery as well as to shunt suprainfection. We suggest that chronic meningitis be ruled out in all patients prior to placement of shunts. Preoperative evaluation should include the examination of cisternal or ventricular CSF when a lumbar CSF specimen is nondiagnositc. When fungal meningitis is present, a course of amphotericin B should be initiated and the CSF sterilized prior to the placement of the permanent extracranial shunt. Where acute hydrocephalus supervenes, temporary ventricular drainage may be employed. In some cases of fungal meningitis, the symptoms of hydrocephalus will be resolved with antifungal therapy alone, obviating the need for ventricular decompression.

Serious staphylococcal infections with strains tolerant to bactericidal antibiotics.

The clinical response in 20 cases of serious staphylococcal infection was compared with the in vitro resistance or "tolerance" of the infecting Staphylococcus to killing by antibiotics used in treatment. Cases were divided into two groups: (1) patients who initially received nonbactericidal antibiotics (ten cases), and (2) patients who initially received bactericidal antibiotics with or without nonbactericidal antibiotics. Mortality due to uncontrolled staphylococcal infection was 40% (4/10) in group 1 as compared with no mortality (1/10) in group0) in group 1 as compared with no mortality (0/10) in group 2. The duration of positive cultures after start of therapy in group 1 (mean, 6.1 days) was significantly longer than that in group 2 (mean, 1.3 days). The duration of fever after start of therapy in group 1 was not significantly different when compared with group 2.

Nosocomial febrile illnesses in patients on an internal medicine service.

Febrile illnesses commonly arise in hospitalized patients after admission, but most previous studies have been of specific subsets of febrile illnesses. To provide practical information about the problem as a whole, we studied febrile illnesses arising after admission (nosocomial febrile illnesses [NFls]) in 123 inpatients of an internal medicine service who had been afebrile for the preceding week. We compared them with 123 randomly selected patients without NFl. Causes of NFl included infections in 83 cases; noninfectious, inflammatory states in 15; malignancy in 12; ischemia in eight; and procedures in three. Evidence for the cause of the NFl was present at onset in at least 110 of the 123 patients. Despite this, antimicrobial agents were administered to 23 (58%) of 40 patients without infections. Thirty-four patients with NFl died; the NFl contributed to death in 26. In contrast, only eight comparison patients died. "Do not resuscitate" status was present in 32 patients with NFl compared with only 12 comparison patients, and 19 (59%) of the former died. The data from this study provide a comprehensive description of NFl arising in hospitalized internal medicine patients, indicate that the occurrence of a new febrile illness signifies a poor prognosis, and provide a rational basis for management.

Clostridium difficile-associated diarrhea and colitis in adults. A prospective case-controlled epidemiologic study.

In a one-year period, 149 adult cases of Clostridium difficile-associated diarrhea and colitis were compared with 148 diarrhea-free controls. Eighty-seven percent were nosocomial and 75% were on surgical services. Endoscopy revealed pseudomembranes in 51% of the 109 cases in which stool cytotoxin was present, compared with 11% of the 40 cases that were culture-positive but cytotoxin-negative. Cases diagnosed only by stool culture showed essentially no differences from controls, 21% of whom had asymptomatic stool colonization. We estimate that only 20% of these cases had diarrhea due to C difficile. Compared with controls, cases diagnosed by the presence of cytotoxin or pseudomembranes were found to have been hospitalized longer at diarrhea onset, to have had more antecedent infections, and to have received clindamycin, multiple antimicrobials, and therapeutic antimicrobials more often than controls, but controls received prophylactic antimicrobials more frequently than cases. Cultures of the environment, patients, and personnel failed to detect a mechanism of acquisition.

Emergence of ciprofloxacin resistance in nosocomial methicillin-resistant Staphylococcus aureus isolates. Resistance during ciprofloxacin plus rifampin therapy for methicillin-resistant S aureus colonization.

We initiated a randomized, single-blinded trial of ciprofloxacin plus rifampin vs sulfamethoxazole and trimethoprim plus rifampin in the therapy for patients who underwent colonization with methicillin-resistant Staphylococcus aureus (MRSA). Patients who were colonized with MRSA received 2 weeks of either regimen. The study was terminated after the enrollment of 21 subjects due to the recognition of ciprofloxacin resistance in 10 of 21 new MRSA isolates during the last 2 months of the study. Five of the 10 patients with ciprofloxacin-resistant MRSA isolates had never received ciprofloxacin. Long-term (6-month) eradication had been achieved in only three of 11 ciprofloxacin plus rifampin and four of 10 sulfamethoxazole and trimethoprim plus rifampin recipients. The use of this new fluoroquinolone for the eradication of MRSA colonization is usually not effective and may risk the development of ciprofloxacin resistance in MRSA within the hospital environment.

Resistance surveillance programs and the incidence of gram-negative bacillary resistance to amikacin from 1967 to 1985.

Data relating to amikacin resistance among gram-negative bacilli were obtained by means of a review of published literature and resistance surveillance studies. Data from the first several years of amikacin use are difficult to interpret because the 10-micrograms disk used for Kirby-Bauer susceptibility testing resulted in apparent greater resistance than the present 30-micrograms disk. A large United States susceptibility surveillance program that monitors antibiotic use has shown a trend since 1977 of greater susceptibility of Serratia species and greater resistance among Pseudomonas aeruginosa for all the aminoglycosides. Pseudomonas resistance to amikacin has shown the smallest increase of any aminoglycoside. Several hospitals (Strong Memorial Hospital, University of Maryland Cancer Center, and Minneapolis Veterans Administration Medical Center) have reported either no significant change or a decrease in resistance to amikacin when it was the most frequently used aminoglycoside. In a large, 14-center, prospective study, high-level use of amikacin resulted in a significant decrease in resistance to gentamicin and tobramycin (p less than 0.01) and a marginal increase (p less than 0.05) in amikacin resistance. Significantly increased amikacin resistance has been reported from two institutions, neither of which used amikacin as the predominant aminoglycoside. Overall, the high-level use of amikacin in large multi-center surveillance programs for as long as five years has not resulted in a significant increase in amikacin resistance rates at any of the individual institutions surveyed.

Aminoglycoside resistance in gram-negative bacilli during increased amikacin use. Comparison of experience in 14 United States hospitals with experience in the Minneapolis Veterans Administration Medical Center.

Resistance to amikacin, gentamicin, and tobramycin was surveyed prospectively during controlled aminoglycoside usage in 14 hospitals. Following an initial baseline period (minimum, three months) during which gentamicin use predominated, gentamicin and tobramycin were placed on restrictive control, establishing amikacin as the aminoglycoside of general use (86 percent of total aminoglycoside usage). During an average of 35 months' restriction of gentamicin and tobramycin, significant reductions in gram-negative resistance to gentamicin (8.4 to 7.0 percent, p less than 0.001) and tobramycin (6.0 to 5.3 percent, p less than 0.01) were observed. The most significant decreases in resistance to gentamicin and tobramycin (p less than 0.001) were found for Pseudomonas aeruginosa, Klebsiella species, Serratia species, and Proteus species. Amikacin resistance among gram-negative bacilli was observed to increase marginally from 1.4 to 1.7 percent (p less than 0.05) during the surveillance period, primarily due to a significant increase in resistance among P. aeruginosa (3.0 to 3.9 percent, p = 0.05). These data were compared with data from a similar surveillance program at the 700-bed Minneapolis Veterans Administration Medical Center. Over a period of 54 months, both gentamicin and tobramycin resistance decreased significantly when amikacin was used (p less than 0.001), then increased with reintroduction of gentamicin (p less than 0.05), and decreased significantly with reintroduction of amikacin (p less than 0.001). Despite predominant amikacin use for a total of 38 months, amikacin resistance did not increase and actually decreased significantly (p less than 0.05) in the last 12 months.

The association between Staphylococcus aureus bacteremia and bacteriuria.

The relationship between Staphylococcus aureus bacteremia and bacteriuria was studied over a five year period in three hospitals. In a Veterans Administration Hospital, 59 patients with Staph, aureus bacteremia had a urine culture within 48 hours of a positive blood culture. In 16 of 59 (27 per cent), greater than 10(5) Staph. aureus was recovered from the urine in pure culture. Six of these patients had apparent primary staphylococcal urinary tract infection. Clinical and laboratory parameters in the patients with staphylococcal bacteremia and bacteriuria were compared with those in 31 patients with staphylococcal bacteremia and sterile urine cultures. The two groups differed only in the more frequent occurrence of pyuria and proteinuria in the bacteriuric patients. In two other hospitals, staphylococcal bacteriuria occurred in 7 per cent of patients with Staph. aureus bacteremia and in 13 per cent of cases of staphylococcal endocarditis. Review of autopsy records for 33 patients who died within one month of their bacteremia failed to show a correlation between bacteriuria and the presence of renal abscess. Staphylococcal bacteriuria is a frequent and unexplained concomitant of Staph. aureus bactremia.

Osteomyelitis in the feet of diabetic patients. Long-term results, prognostic factors, and the role of antimicrobial and surgical therapy.

Fifty-one diabetic patients with osteomyelitis of the foot were studied to determine potential prognostic factors and the role of antimicrobial therapy. Most of the patients were elderly, with diminished pulses, a sensory neuropathy, and a polymicrobial infection. Twenty-seven patients had a good outcome, defined as clinical resolution at the time of the last follow-up examination, without the need for amputation. The mean duration of follow-up for these patients was 19 months. Fifteen patients had a below-knee amputation, and nine had a toe amputation. The absence of necrosis and/or gangrene, the presence of swelling, and the use of antimicrobial therapy active against the isolated pathogens for at least four weeks intravenously, or combined intravenously and orally for 10 weeks, predicted a good outcome. Diabetic foot osteomyelitis, in the absence of extensive necrosis or gangrene, usually responds to antimicrobial therapy without the need for an ablative surgical procedure.

Evaluation of ciprofloxacin's synergism with other agents by multiple in vitro methods.

The efficacies of ciprofloxacin, ceftizoxime, azlocillin, mezlocillin, and amikacin (minimal inhibitory concentration and minimal bactericidal concentration) against six Pseudomonas aeruginosa, six Enterobacteriaceae, and six group D streptococcal strains were evaluated using both agar and broth susceptibility methods, two inoculum sizes (5.7 log10 colony-forming units (cfu)/ml and 7.7 log10 cfu/ml), and aerobic and anaerobic incubation conditions. The results showed agreement between broth and agar methods of susceptibility determination; inoculum effects with beta-lactam antimicrobials; and decreased susceptibility to amikacin under anaerobiasis. Ciprofloxacin combined with azlocillin, ceftizoxime, or aminoglycosides in broth microdilution checkerboards against 100 gram-negative bacilli and gram-positive cocci demonstrated that ciprofloxacin combined with azlocillin or ceftizoxime was synergistic against at least 50 percent of P. aeruginosa and Serratia marcescens isolates and that ciprofloxacin combined with amikacin was synergistic against at least 50 percent of S. marcescens and Staphylococcus aureus isolates. Ciprofloxacin and azlocillin in combination were evaluated by microdilution checkerboard, agar dilution, and broth macrodilution time-kill methods at two inoculum sizes to assess antibacterial activity. Comparison between in vitro combination methods showed the following: the presence or absence of checkerboard synergism (as defined by the fractional inhibitory concentration index and the fractional bactericidal concentration index) with ciprofloxacin and azlocillin did not correlate with time-kill results; and good agreement between methods when comparing broth macrodilution time-kill (3 log10 cfu/ml or more decrease) with antimicrobial combinations at a single concentration in both agar and microdilution broth for ciprofloxacin and azlocillin. Rabbit studies using subcutaneous dialysis membrane chambers inoculated with six P. aeruginosa, six Enterobacteriaceae, and six group D streptococcal strains were performed using ciprofloxacin, azlocillin, ceftizoxime, and amikacin alone and in combination as therapy. In vitro testing of antibiotic combinations that provided the best prediction of in vivo outcome were combination antibacterial activity (3 log10 cfu/ml or more decrease) at 24 hours using either broth macrodilution time-kill or antimicrobial combinations at a single concentration in either agar or broth (microdilution). For the most efficacious in vivo combination, ciprofloxacin plus azlocillin, there was in vitro correlation with in vivo outcome for 17 of 18 isolates.