INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion Appendix References

Skin and soft tissue infections (SSTIs), such as spontaneous lymphangitis or cellulitis, and especially complicated infections, such as wound and surgical infections or diabetic foot ulcers, often require hospitalization and intravenous (i.v.) antibacterial treatment. These infections are caused by a mixture of aerobic and anaerobic organisms and are responsible for increased morbidity, prolonged hospital stay, and increased health care costs (3, 13, 19).

Staphylococcus aureus and streptococci; gram-negative bacteria such as Pseudomonas aeruginosa, Enterobacteriaceae, and Enterococcus spp.; and anaerobes such as Bacteroides fragilis are frequently isolated (8, 12, 13, 18, 32). Methicillin resistance among S. aureus causing SSTI currently ranges from 11% in Canada to 19% in the United States, with rates as high as 25 to 50% in infected ulcers. Ciprofloxacin resistance among P. aeruginosa ranges from 19 to 36% (R. N. Jones, M. A. Schomberg, M. A. Pfaller, et al. and the SENTRY Participants Group, 38th Intersci. Conf. Antimicrob. Agents Chemother., abstr. E94, 1997).

Frequently used antimicrobial agents have included expandedspectrum cephalosporins (e.g., cefoxitin, cefotetan, cefmetazole), imipenem-cilastatin, -lactam--lactamase inhibitor combinations (e.g., piperacillin-tazobactam, ticarcillin-clavulanate), and fluoroquinolones (13, 14, 15, 18, 20, 35). However, emerging resistance, particularly among S. aureus, P. aeruginosa, and enterococci, is making the choice of treatment increasingly more difficult (23).

Clinafloxacin is an extended-spectrum fluoroquinolone antibacterial that is bactericidal against S. aureus, including methicillin-resistant strains, and most strains of ciprofloxacin-resistant S. aureus (7, 33). Like other fluoroquinolones, clinafloxacin has activity against Pseudomonas spp. and the Enterobacteriaceae; the MICs of clinafloxacin at which 90% of isolates are inhibited (MIC₉₀s) are similar to or lower than those of ciprofloxacin (17, 34). Clinafloxacin is also highly active against most species of anaerobes, and its MIC₉₀s are substantially lower than those of ciprofloxacin (6, 36).

Clinafloxacin steady-state maximum concentrations in plasma in humans averaged 2.6 µg/ml following dosing of 200 mg i.v. every 12 h (q12h) and oral (p.o.) bioavailability of 90% (31). Penetration into interstitial fluid and skin blister fluid is rapid and extensive, with mean ratios of concentration in blisters to concentration in plasma of 93.1% (37).

The primary objective of this study was to evaluate the efficacy and safety of clinafloxacin versus a regimen consisting of piperacillin-tazobactam with optional vancomycin among patients with complicated SSTIs.

(These data were previously presented as an abstract [Abstr. 39th Intersci. Conf. Antimicrob. Agents Chemother. abstr. 1106a, 1999].)

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion Appendix References

Adult patients with severe or limb-threatening SSTIs serious enough to require hospitalization and i.v. therapy were eligible for inclusion in this randomized, investigator-blind, multicenter, parallel-group trial. Patients with acute (5 days prior) physical findings of complicated SSTI of bacterial etiology and a diagnosis of spontaneous infection (e.g., phlegmon, cellulitis, lymphangitis), wound infections (e.g., trauma wound, surgical wound), or diabetic foot infection were included (5). Patients were also required to have material available for culture. Exclusion criteria included pregnancy or breastfeeding, significant hepatobiliary or renal dysfunction (total bilirubin levels of three or more times the upper limit of normal, alanine transaminase or aspartate transaminase levels of five or more times the upper limit of normal, or estimated creatinine clearance of <20 ml/min), immunodeficiency conditions, risk of convulsive disorders, hypersensitivity to study medications, septic shock, infected burns or decubitus ulcers, osteomyelitis, and major amputation. Also, patients were not allowed to have (i) been treated with more than a single dose of systemic antibacterial therapy for the current SSTI, (ii) had the infected site treated with topical antibiotics within 24 h prior to baseline culture collection, (iii) had prior treatment with any study medication within 7 days prior to study entry, or (iv) received treatment with any other investigational drug within 4 weeks prior to randomization. Also excluded were patients (i) receiving corticosteroids (>1 mg/kg of body weight/day, (ii) requiring concomitant topical antimicrobial agents for SSTIs, (iii) receiving other antibacterial therapy for concomitant infections, and (iv) known to have SSTI pathogens resistant to any study medication.

Patients were randomized in a 1:1 ratio at each site to receive either clinafloxacin (200 mg i.v. q 12h) or piperacillin-tazobactam (3.375 g i.v. q6h). To ensure blinding, patients in the clinafloxacin group also received placebo infusions every 12 h; both clinafloxacin and piperacillin-tazobactam were infused over a 30-min period. Patients in the piperacillin-tazobactam arm, but not in the clinafloxacin arm, could also receive i.v. vancomycin if methicillin-resistant staphylococci or enterococci were suspected or isolated. Following a minimum of 3 days of i.v. therapy, patients could be switched to p.o. therapy; patients in the clinafloxacin group received clinafloxacin (200 mg p.o. q12h), and those in the piperacillin-tazobactam group received amoxicillin-clavulanate (500 mg p.o. q8h).

Patients were advised to avoid direct or indirect sunlight and to apply sunscreen as well as wear protective clothing if sunlight exposure was unavoidable. Dosage of study drugs was adjusted in patients with impaired renal function (creatinine clearance, To maintain investigator blinding during outpatient therapy, a third-party member, who was not involved in assessing patient medical status, was responsible for dispensing study medication and retrieving patient diaries.

Dosing of the i.v. and p.o. courses combined was not to exceed 14 days unless approved by the sponsor. Following the completion of treatment, patients were assessed at a test-of-cure (TOC) visit (6 to 14 days posttherapy) and at a long-term follow-up (LTFU) visit (21 to 35 days post-therapy).

Within 48 h prior to the initiation of therapy, the baseline evaluation of patients was conducted, including a medical history, physical examination, electrocardiogram, clinical assessment of signs and symptoms, baseline culture, and clinical laboratory tests.

Pathogen susceptibility to clinafloxacin was determined by broth microdilution using trays manufactured by AccuMed and by Kirby-Bauer disk diffusion methods for determination of zones of inhibition.

Tentative clinafloxacin MIC susceptibility breakpoints that were used in this trial were defined as follows: susceptible, 1 µg/ml; intermediate, 2 µg/ml; and resistant, 4 µg/ml (22). Susceptibilities to piperacillin-tazobactam, amoxicillin-clavulanate, and vancomycin were determined by MIC testing. Susceptibility testing was performed by central laboratories (Covance, Indianapolis Ind.), which strictly conformed with National Committee for Clinical Laboratory Standards guidelines (27).

Efficacy and safety analyses. The primary efficacy parameters were the clinical cure rates and the by-pathogen microbiological eradication rates determined at the TOC visit, 6 to 14 days posttherapy. Secondary efficacy parameters included the clinical cure rate and the microbiological eradication rates at the LTFU visit, 21 to 35 days posttherapy. In addition, the development of resistance, the amputation rate, and the survival rate were determined at LTFU. An investigator assessment of clinical cure rate in all patients randomized to study treatment (intent-to-treat [ITT] population) was also conducted at the LTFU visit.

Statistical methodology. A sample size of 198 microbiologically evaluable patients (99 per treatment group) was selected to provide 80% power to assess the equivalence of treatments using a two-tailed 95% confidence interval (CI) with limits of ±20%, assuming a clinical response rate of 75%. With an estimated evaluability rate of 50%, a total enrollment of 396 patients was required.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion Appendix References

Four hundred nine patients were randomized to treatment, with 213 in the clinafloxacin group and 196 in the piperacillin-tazobactam group (Table 1). Patient characteristics and baseline diagnoses were similar in the two groups, although there were more patients 65 years of age in the piperacillin-tazobactam group than in the clinafloxacin group (32.7 versus 26.8%) (P = 0.196).

The numbers of patients who completed the study treatment were similar in the clinafloxacin and piperacillin-tazobactam groups (69.5 and 74.0%, respectively). More patients discontinued clinafloxacin prematurely because of adverse events than those receiving piperacillin-tazobactam) (11.4 versus 6.3%, respectively) (P = 0.05). More patients discontinued piperacillin-tazobactam prematurely due to treatment failure than those receiving clinafloxacin (8.7 versus 5.6%, respectively), though the difference was not statistically significant (P = 0.25).

Of the patients who completed study treatment, the median duration of treatment (primary study drug plus protocol-allowed add-ons) was 13 days in both groups. Six (2.8%) patients received clinafloxacin (300 or 400 mg q12h) at some point during treatment because of suspected antibacterial resistance (n = 5) or obesity (n = 1). Twelve (6.1%) of the patients in the piperacillin-tazobactam group received vancomycin for methicillin-resistant staphylococci or enterococci. A total of 290 patients switched to p.o. medication. Approximately 80% of patients received nondrug therapy for their SSTI during the study. Dressing changes, debridement, wound incisions, and drainage were common.

There were 279 patients (144 receiving clinafloxacin and 135 receiving piperacillin-tazobactam) who were clinically evaluable and 204 patients (108 receiving clinafloxacin and 96 receiving piperacillin-tazobactam) who were microbiologically evaluable at the TOC visit. The most common reasons for exclusion from the clinically evaluable population were insufficient treatment duration (<5 days of therapy for evaluable cure or <3 days therapy for evaluable failure) and missed clinical assessment (which includes patients who had telephone contact assessments). The most common reason for exclusion from the microbiologically evaluable group was a lack of a proven pathogen at baseline (Table 2).

A total of 587 pathogens were isolated from 264 patients (141 receiving clinafloxacin, and 123 receiving piperacillin-tazobactam) in the modified ITT population. Multiple pathogens were isolated from 79 clinafloxacin-treated patients and 65 piperacillin-tazobactam-treated patients (55% overall). Of the patients with single pathogens, S. aureus was the most common, isolated from 29 clinafloxacin-treated and 33 piperacillin-tazobactam-treated patients. Additional single isolates included Streptococcus pyogenes (8 and 6 clinafloxacin- and piperacillin-tazobactam-treated patients, respectively) and P. aeruginosa (6 and 1 clinafloxacin- and piperacillin-tazobactam-treated patients respectively). Pathogen susceptibility to clinafloxacin was determined in 560 baseline isolates, and that to piperacillin-tazobactam was determined in 552 isolates. Overall the MICs of clinafloxacin for most isolates were 1, and most isolates were susceptible to piperacillin-tazobactam. Twenty-five S. aureus isolates were methicillin resistant. The other commonly isolated pathogens were Enterococcus faecalis (n = 37), P. aeruginosa (n = 34), Streptococcus agalactiae (n = 25), S. pyogenes (n = 25), and Enterobacter cloacae (n = 23) (Table 3).

Fewer baseline isolates were resistant to clinafloxacin (10 of 560 [1.8%] than to piperacillin-tazobactam (34 of 552 [6.2%]; P = 0.001) or amoxicillin-clavulanate (113 of 552 [20.5%]; P = 0.001). Isolates resistant to clinafloxacin included methicillin-resistant S. aureus and E. faecalis, while those resistant to piperacillin-tazobactam included methicillin-resistant S. aureus, E. cloacae, and P. aeruginosa.

Efficacy. Clinical cure rates in the clinafloxacin-treated patients (68.8%) and the piperacillin-tazobactam-treated patients (65.2%) were similar. Cure rates were also similar between treatment groups for each baseline diagnosis category (Table 4).

Safety. Of 409 patients randomized to treatment, 400 received study drug and were evaluable for safety. The majority of adverse events were mild to moderate in intensity (Table 6). There were 82 (39.0%) clinafloxacin-treated patients with drug-associated adverse events compared to 57 (30.0%) piperacillin-tazobactam-treated patients with drug-associated events (P = 0.050). The difference in rates of associated adverse events was primarily due to photosensitivity in clinafloxacin-treated patients. The most common drug-associated adverse events reported in at least 5% of patients were phototoxicity (23 of 210 [11.0%]) in the clinafloxacin group and nausea (10 of 190 [5.3%]) and diarrhea (17 of 190 [8.9%]) in the piperacillin-tazobactam group.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion Appendix References

In the present study, microbiologic eradication rates were similar for clinafloxacin and piperacillin-tazobactam, with by-pathogen eradication rates of 61.5 and 57.2%. These rates are somewhat lower than the pathogen eradication rates previously reported in a comparison of piperacillin-tazobactam (76%) and ticarcillin-clavulanate (82.8%) (35). The lower rates may reflect the greater number of baseline isolates of P. aeruginosa and enterococci in the present study and the strict criteria used to assess pathogen response, in that persistence (rather than unevaluability) was assessed if the patient received nonantibiotics or was otherwise felt to be clinically failing protocol.

Clinafloxacin exhibited higher by-pathogen eradication rates compared to piperacillin-tazobactam for all three of the most commonly isolated pathogens, S. aureus (including methicillin-resistant S. aureus), E. faecalis, and P. aeruginosa, although no differences were statistically significant.

During the early 1990s, fluoroquinolones such as ciprofloxacin became widely used in the treatment of SSTIs, with reported cure rates between 70 and 85% (18, 19, 20). In a randomized trial, i.v. fleroxacin administered once daily was shown to be as efficacious as ceftazidime administered two or three times daily in the treatment of patients with severe SSTIs, achieving cure rates of 82 and 73%, respectively (29). Likewise, a randomized, multicenter trial compared sequential i.v.-to-p.o. ofloxacin (400 mg) with i.v. ampicillin-sulbactam (1 to 2 g and 0.5 to 1 g) followed by p.o. amoxicillin-clavulanate (500 and 125 mg) in the treatment of diabetic patients hospitalized with foot infections; clinical success rates (cure and improvement) were 85% in the ofloxacin group and 83% in the aminopenicillin group, and microbiological eradication rates were 78 and 88%, respectively (26).

Despite these encouraging results, the development of high and rapidly increasing rates of ciprofloxacin resistance among isolates of methicillin-resistant S. aureus as well as P. aeruginosa has discouraged the role of fluoroquinolones in the empiric therapy of severe SSTIs (8, 21, 30). This has led to the development of newer fluoroquinolones such as levofloxacin, moxifloxacin, and trovafloxacin, with higher potencies against quinolone-resistant S. aureus, P. aeruginosa, and anaerobes (4, 25). In a randomized, multicenter study with p.o. trovafloxacin (200 mg once daily) versus p.o. amoxicillin-clavulanic acid (500 and 125 mg three times daily) in the treatment of complicated SSTIs, high pathogen eradication rates (eradication plus presumed eradication) were observed for S. aureus (80 versus 73%), E. faecalis (93 versus 85%), and P. aeruginosa (70 versus 60%), respectively (9).

In the present study the effectiveness of clinafloxacin given alone, i.v. (200 mg q 12 h) with an optional switch to p.o. therapy, was similar to that of piperacillin-tazobactam (3.375 g i.v. q 6 h), followed by oral amoxicillin-clavulanate (500 mg q8 h), with the optional addition of vancomycin for methicillin-resistant staphylococci or enterococci. Clinical cure rates were 68.8% for the clinafloxacin group and 65.2% for the piperacillin-tazobactam group. Overall clinical cure rates were somewhat lower than those in previously published reports (9, 10, 20) and may be a reflection of differences in methodology. The clinical cure rate reported in the present study was determined 6 to 14 days posttherapy rather than the more common endpoint at end-of-therapy. Patient outcomes were assessed bimodally as cure or failure; thus, the patient with only improved clinical status with partial resolution of symptoms would conservatively be considered a failure. This approach is in contrast to the more common approach which allows for a cure, improved, or failed outcome as assessment options. A multicenter study with ticarcillin-clavulanate versus piperacillin-tazobactam in patients with complicated skin and skin structure infections (half of which were graded severe) demonstrated a 61% cure rate in both treatment groups (similar to the present study), while another 16 and 15% were considered improved, respectively (35). Stricter criteria for patient evaluability and handling of nonstudy antibacterial therapy may also have resulted in fewer patients being classified as evaluable cures, since patients who received nonprotocol antibacterials were considered treatment failures rather than nonevaluable. As in other recently published reports, the most commonly isolated baseline pathogens in the present study were S. aureus, E. faecalis, and P. aeruginosa (14, 24); B. fragilis was the most predominate anerobic pathogen, being found 11 of 95 isolates.

In this study a there was a statistically significant difference in the rate of resistance in baseline pathogens, favoring clinafloxacin (P = 0.001). The baseline susceptibility of >95% of pathogens to clinafloxacin confirms the broad spectrum of activity of this antibacterial; 4 of the 10 clinafloxacin-resistant isolates were enterococci. In contrast, 21 of the 34 isolates resistant to piperacillin-tazobactam were staphylococci, supporting this antibacterial's recognized lack of activity against methicillin-resistant cocci (1).

Fluoroquinolones containing a halogen group at position 8 of the quinolone ring, including lomefloxacin, fleroxacin, and sparfloxacin, have been associated with high rates of photosensitivity reactions (11, 28). Photosensitivity, at the rate of 11% in patients treated with clinafloxacin, was the most frequently reported drug-associated adverse event in the clinafloxacin treatment group. The majority of cases of photosensitivity were of mild to moderate intensity; however, four cases were considered severe and required treatment discontinuation. Outpatients were more likely to experience photosensitivity reactions, although two hospitalized patients exposed to direct or indirect sunlight developed photosensitivity reactions. Photosensitivity appears to be a manageable adverse event in clinafloxacin-treated inpatients. The rates of other adverse events were similar between the two treatment groups.

In summary, clinafloxacin's broad range of activity against the pathogens most frequently associated with SSTIs, including many resistant species that have become increasingly problematic in recent years, allows clinafloxacin monotherapy to achieve efficacy equivalent to that of a regimen of piperacillin-tazobactam plus optional vancomycin in the treatment of hospitalized patients with severe SSTIs.