Netilmicin Sulfate as Single-Agent Therapy for Pseudomonas Infections RICHARD N. GREENBERG, MD, RICHARD R. LORBER, MD, and GEORGE H. MILLER, PhD, Kenilworth, NJ ABSTRACT: In a prospective multicenter clinical trial, 69 patients with Pseudomonas infections were treated with netilmicin sulfate as the only antipseudomonal antibiotic. Clinical resolution or improvement was observed for 81% of the infections, whereas 19% were considered treatment failures. The bacteriologic response, based on follow-up culture results, showed elimination of Pseudomonas from 62% of the infection sites, with Louis, Mo; JOHN N. HANSBROUGH, MD, San Diego, Calif persistence in 30%. All isolates were susce le by disk susceptibility testing (zone >15 mm), and by microdilution testing in unsupplemented broth. The majority of isolates, however, were res accounted for by factors other than ne effective as treatment for netilmici adults, A low incidence of nephrotoxic serum levels NETILMICIN SULFATE is an aminoglycoside anti- biotic approved for the treatment of infections caused by a variety of gram-negative organisms including Enterobacteriaceae and Pseudomonas species. Because the usefulness of an antibiotic is often derived by its in vitro susceptibility testing to the pathogen, proper interpretation of in vitro susceptibility results is imperative. Current suscep- tibility test guidelines published by the National Committee for Clinical Laboratory Standards (NCGLS) recommend Mueller-Hinton media with the addition of physiologic concentrations of calcium (50 mg/L) and magnesium (25 mg/L) for broth assays.! However, the in vitro activity of netilmicin is profoundly affected by such divalent cation concentrations.* For Pseudomonas spp without aminoglycoside in- activating enzymes, the minimal inhibitory concen- tration (MIC) ratio between cation supplemented, Mueller-Hinton broth (MHB) and unsupple- mented MHB for netilmicin is 21.1, as compared with 15.2 for gentamicin, 9.6 for amikacin, and 7.5 for tobramycin.’ In fact Barry et al,* report ing clinical outcome versus MICs of Pseudomonas ‘Cher: andthe Deprnte Cinieal Research and Neral Schering {creat Remy ok eh sey ws done ere Reprint reps o Richa iberg. MD, Division of Infectious bin Sane Univers Thine 14088 Gran vd Sr oui RIO 63108 stant in cation supplemented media. The clinical ceptible Pseudomonas infections in nonneutropenic (12%) oceurred despite careful monitoring of spp infection treated with netilmicin only, com: mented that cation-supplementation of MHB resulted in numerous '‘false-resistant™ results and found no evidence that susceptibility tests in un- supplemented broth (resulting in lower MICs for netilmicin to Pseudomonas spp) produced misleading clinical data. This paper is a com- panion paper to that of Barry et al, and examines in detail the safety and efficacy of netilmicin as a single agent in the treatment of netilmicin- susceptible Pseudomonas infections. PATIENTS AND METHODS ‘This open, multicenter trial was conducted at 21 sites in the United States, Europe, and South America, using a common protocol. The protocol .was approved by the institutional review board at cach study center, and informed consent we obtained from each patient before entry into the study Hospitalized patients of either sex and of any race and age with a known or suspected Pieudo- monas infection were eligible for enrollment. ‘The presence of Pseudomonas as a causative pathogen at an infection site had to be documented by ap- propriate culture and susceptibility testing done up to three days before initiating study medica- tion, or on the first day of therapy. For patients with suspected bacteremia/septicernia, two blood cultures taken one hour apart, from separate veni- punctures, and positive for Pseudomonas had to be obtained before initiating therapy. Greenberg et al + NETILMICIN FOR PSEUDOMONAS INFECTIONS 715Patients were excluded from the study if they were pregnant, known to be allergic to any amino- glycosides, receiving dialysis, had a diagnosis of meningitis, osteomyelitis, endocarditis, or leuko- penia, or were enrolled in other clinical trials. susceptibility of Pseudomonas isolates (0 was determined by the standard anti- biotic disk technique (Bauer-Kirby method) using a 30 ug netilmicin disk. A zone diameter of = 15 mm was indicative of susceptibility of Pseudomonas to netilmicin; a zone of $12 mm was indicative of resistance. To monitor bacteriologic responses to therapy, repeat cultures were obtained during and after therapy (within 48 hours of discontinu- ing antibiotics). All isolates were labeled Pseudomonas aeruginosa by the reporting physicians, The participating investigators were required to submit to the Schering Corporation subcultures of all Preudomonas organisms isolated. ‘These isolates were reidentified and tested at the Scher ing Corporation (Bloomfield, NJ), the Clinical Microbiology Institute (Tualatin, Ore), and the Centers for Disease Control (Atlanta, Ga) on a variety of culture media to determine the influence of various concentrations of calcium and magnesium in Mueller-Hinton broth on the MIC of netilmicin against these organisms.* Patients who qualified for enrollment into the study received netilmicin sulfate parenterally at a daily dose of 4.0 to 6.5 mg/kg. The adequacy of dosage was monitored by periodic determina- tions of the netilmicin serum level. Peak serum levels were to be maintained between 4 and 10 mg/ml; trough levels were to be kept at <2 pg/ml. Other antibiotics could be administered in addi- tion to netilmicin, provided they did not possess in vitro antipseudomonal activity. ‘The patients’ response to therapy was assessed both clinically and bacteriologically. Clinical responses were categorized as complete resolution, tm- provement, failure, or indeterminate. Bacteriologic responses, based on the results of cultures obtained belore, during, and after therapy, were categor- ized as elimination, persistence, or indeterminate Adverse reactions were monitored before and dur~ ing therapy by laboratory and clinical evaluations, including complete blood count, liver function studies, renal function studies, urinalysis, and serum electrolyte determinations. Attempts to monitor for vestibular and ototoxicity were made by evaluation for nystagmus and by audiograms (when possible) done before, during, and after therapy. Nephrotoxicity was defined as a 50% increase of serum creatinine above the prestudy serum creatinine to a level greater than 1.3 mg/dl RESULTS Of the total of 220 patients enrolled into the study, 69 were considered evaluable for efficacy determinations in that they had a documented Pseudomonas infection treated with netilmicin as the only antipseudomonal agent, and the isolate was submitted to Schering. One hundred fifty-one pa- tients were excluded from the efficacy population for the following reasons: no Pseudomonas isolated or no growth in initial culture (69), patient received netilmicin for less than 72 hours or had an interrupted course of therapy (58), isolate was not submitted to Schering (16), inappropriate culture was obtained (six), and organism was resis- tant to netilmicin (two), ‘The efficacy population consisted of 48 men and 21 women, with a mean age of 30.5 (+ 22.1) years, and mean weight of 69.9 (+ 23.2) kg. The mean duration of therapy was 10.9 (+ 4.9) days, and the mean daily netilmicin dose was 4.3 (+ 1.9) mg/kg. In these patients, the sites of infection were identified as skin and skin structures (51), blood (13), abdomen (four), and lower respiratory tract (one). 51 infections of the skin and skin structures included a variety of categories, the major category being surgical or traumatic wounds (20), followed by infected burns (15), documented by biopsy cultures or manifested by sloughing of skin grafts. Eight patients had infected decubitus ulcers, and another eight had miscellaneous infections, Approximately 40% of the 69 evaluable infec- tion ‘sites were reported to have pathogens in addition to Pseudomonas. The most frequent gram- negative pathogen was Escherichia coli, and the most common gram-positive organisms’ were entero- cocei and Staphylococcus aureus. Concomitant antimicrobials were administered to 71% (49/69) of the patients, the remaining 20 patients having received only netilmicin. The most frequently used agents were penicillins (38%), cephalosporins (38%), and clindamycin (24%). Other agents included metronidazole (12%), trimethoprim-sulfamethoxazole (4%), vancomycin (1%), and erythromycin (1%). In each instance the only agent active against the Pseudomonas was netilmicin, Complete clinical resolution or improvement was observed for 81% of the infection sites, with 19% considered treatment failures (Table 1). To determine whether the therapeutic responses in pa- tients with monomicrobial infections (Pseudomonas only) differed from the responses in those with polymicrobial infections, we examined the clinical response for the 45 sites in which Pseudomonas was 716 June 1989 + SOUTHERN MEDICAL JOURNAL * Vol, 82, No. 6TABLE 1, Cincal Responses ‘TABLE 2. Gactovlogic Responses Chal Hepa the only causative organism isolated. In these infections, complete clinical resolution or improve- ment occurred in 88% of the sites, with 12% deemed treatment failures. To further determine whether there were discernible differences between patients who responded favorably to treatment and those in whom therapy failed, we compared the demo- graphic characteristics of these groups and found no. statistically significant differences in age, weight, duration of treatment, or mean daily dose of netilmicin, ‘The bacteriologic data, including susceptibility to netilmicin in various media, have been pub- lished.’ Bacteriologic elimination of Pseudomonas from the infection site was documented in 62% of the sites, with persistence occurring in 30% (Table 2). While receiving therapy two patients had elimination of the infecting pathogen, but superinfections developed with another genus of Pseudomonas; three patients whose infection appeared to have resolved did not have a second culture during or after treatment and could not be fully evaluated. For sites with monomicrobial infections, Pseudomonas was eliminated from 70% and persisted in 30% Adverse reactions were documented in eight pa- tients. All adverse reactions were significant rises in serum creatinine. In four instances only serum| trough levels of netilmicin were elevated, and in one instance both peak and trough levels were elevated. Levels were within the normal range in two patients, and in one patient no serum levels were available. No audiologic toxicity was iden- tified which could be clearly related to netilmicin. We identified 25 patients in whom therapy failed or whose pathogen persisted. In most instances, the ‘‘pathogen’” either persisted (15 patients) or appeared to become resistant (four patients) in surface cultures of open skin wounds. Of the 15 patients with persistence of the putative pathogen, nine had clinical improvement. Those in whom therapy failed had apparent superinfec- tions (two), severe skin burns (three), and decubi- tal ulcer with a decreased level of consciousness (one) In one instance, netilmicin therapy was discon- tinued because severe renal failure developed. One intra-abdominal mixed infection with a fistula to a colectomy site failed (0 respond to netilmicin and metronidazole, and treatment was changed after ten days. Four patients died with se Pseudomonas aeruginosa. One of them re y six doses of netilmicin, and blood cultures just before death yielded no growth (this patient was hot included in evaluation of overall data). Three other patients clearly had treatment failure: one was a leukemic receiving steroids, one was an alcoholic in whom treatment with ticarcillin and tobramycin also failed, and one had severe chronic obstructive lung disease with a tracheostomy and died after seven days of netilmicin therapy. DISCUSSION This report describes results from an interna- tional study to evaluate the safety and efficacy of netilmicin as single-agent therapy for Pseudomonas infections. Sixty-nine patients with such infections, were studied. In each instance, the pathogen was susceptible 10 netilmicin. Among nonleukopenic patients, 81% showed clinical resolution or improvement, 62% having bacteriologic elimina- tion of Pseudomonas from the infection site. A. critical analysis of patients in whom therapy failed or Pieudomonas persisted disclosed no data to su gest lack of antimicrobial effect. There were four instances of resistance developing during netil- micin therapy, two superinfections of open skin wounds, several failures in severe burns (three pa tients), one decubitus ulcer failing to respond, and one abdominal fistula failing to heal, Four patients died of sepsis; one failed to respond not only to netilmicin but also to combination therapy with ticarcillin and tobramycin, one died too quickly to allow evaluation (afier six doses), one with active leukemia died while receiving steroids, and one died with severe chronic obstructive lung disease. In each instance, extenuating factors were involved in drug failure, Greenberg et al © NETILMICIN FOR PSEUDOMONAS INFECTIONS 717Ototoxicity was not found in the study. Eight instances of nephrotoxicity were documented (12%), and most of these patients had elevated serum levels of netilmicin. Our 12% incidence of netilmicin-induced nephrotoxicity is similar to that reported in other aminoglycoside studies.> No other unusual adverse reactions were described. Collection of nonblood or nontissue isolates for this study is one major concern. Open wounds are often colonized with Pseudomonas spp, and surface cultures are not definitive. In the review of the data for this report, all patients with skin infec- tion identified by the investigator as having persistence of the pathogen had cultures obtained from surface “‘pus.”” Hence, it is not certain that, in some instances they did not isolate a coloniz- ing Pseudomonas, but the data were analyzed as if every culture did grow the pathogen. This retrospective analysis may actually bias the data against netilmicin efficacy. Nevertheless, persis- tence of Pseudomonas was found in only 33% of skin, infections. ‘This study did not compare netilmicin to other antipseudomonal agents and therefore we can make no claim that the agent is any better or worse than other single agents in the treatment of Pseudomonas infection. Furthermore, in the initial treatment of patients with Pseudomonas infections, combination therapy with antimicrobials selected from various groups (8-lactams, aminoglycosides, quinolones, etc) may improve the patient’s chance of survival. In a study of 200 consecutive patients with Pseudomonas aeruginosa bacteremia (58% of Continued fom page 714 22 Sehuthaler EP, Cosma ME, Eagin ta Sind Kapon Seton in pat nh AIDS se) Med Ba 25. Blackwood 1 Penningian JE: Dowdepeaden tet a aco tuts skibes nates et, Be ep whom were immunocompromised), Hilf et al? found significantly different mortality rates between patients receiving combination therapy (27%) and those receiving monotherapy (47%). Because of the possibility of multiple antimicrobial resistance by the organism or immunocompromise in the host, combination therapy is recommended, at least initially, and until susceptibility results are available. Acknowledgments. Cases reviewed in this paper were evaluated by Drs. Austin (Houston), Colby (Brookdyn, NY), Conway (New Orleans), Hansbrough (San Diego), Shlacs (Cleveland) Siebert (Houston), Slama (Indianapolis), Moreno (Colt Colombia), Noone (London), Prada (Bogotd, Colombia), ‘Trujillo (Medellin, Colombia), Dinick (Birmingham, Ala), Gerding (Minneapolis), Ryan. Westhaven, Conn), and Greenberg (St Louis). We thank Sue Stevens for seretaial assistance Aetorence mittee for Cinical Laboratory Standards, RY " Seger a ace 21. Moers HU: Meehan accom ation in pulmonary deat Cha pisiee nS Pe 45. Begin P Ser yoe in Promo tcvn pcan, Cm Md A J 718 June 1989 + SOUTHERN MEDICAL JOURNAL * Vol. 62, No. 6