Intravenous Ciprofloxacin or Ceftazidime in Selected Infections A Prospective, Randomized, Controlled Study JORGE VILLAVICENCIO, M.D., MARIA ESTHER ASENSIO DE FERNANDEZ, M.D., CLAUDIO A. RAMIREZ, M.D., F.A.C.P. Guatemala City, Guatemala

In a prospective, randomized, controlled study, the effectiveness and safety of intravenous cipro- floxacin was compared with those of ceftazidime in the treatment of tissue infections. A total of 52 patients received intravenous ciprofloxacin 200 mg twice daily (26 patients) or ceftazidime 1 to 2 g every eight to 12 hours (26 patients). This was followed, respectively, by oral ciprofloxacin or another suitable drug when improvement was seen. Informed consent was obtained from all patients. Cultures and laboratory determinations were performed initially and repeated periodically as indicated. Both groups were comparable in age, sex, number of infected sites, and instance of nosocomial infections (27 total). Severe infec- tions occurred in six ciprofloxacin and 12 ceftazi- dime patients (p = 0.056). There were 13 patients in the ciprofloxacin group and six in the ceftazi- dime group with accompanying diseases (p = 0.032). Patients were treated with ciprofloxacin or ceftazidime for infections of the urinary tract (eight and two patients, respectively), skin or soft tissues (15 and 17 patients), pelvis (one and zero patients), lower respiratory tract (one and one patients), intra-abdominal (zero and three patients), and bacteremia (three and five pa- tients). Two patients in each group had two sites of infection. Causative organisms were as fol- lows: 15 gram-positive cocci (90 percent minimal inhibitory concentration: ciprofloxacin, 0.5; cefta- zidime, 16.0 pg/ml) and 55 gram-negative rods (44 Enterobacteriaceae, nine Pseudomonas aerugin- osa; 90 percent minimal inhibitory concentration: ciprofloxacin, 0.25; ceftazidime, 8.0 pgiml). Re- sistance emerged in one patient treated with ciprofloxacin (Acinetobacter sp.> and 12 treated with ceftazidime (four Enterococcus, two Staphy- Zococcus aureus, six other). Intravenous treat- ment was longer with ceftazidime (11.5 days ver- sus 5.6 days for the ciprofloxacin group, p <0.0005), but the total duration of therapy was similar (12.9 versus 14.1 days, p value not signifi- cant). Resolution or improvement occurred in 23 ciprofloxacin and 26 ceftazidime sites of infection (p value not significant). Death occurred in two ceftazidime-treated patients (due to bacterial in- fection) and one ciprofloxacin-treated patient (at the induction of anesthesia). Adverse experiences

From the Sections of Infectious Diseases and Clinical MicrobIology, Roosevelt Hospi-

addressed to Dr. Claudlo A. Ramirez, Section of infectious Diseases, Roosevelt Hospi- tal, 14 Calle 7-13, Suite 82, Guatemala City, Guatemala 01009.

were more common in the ceftazidime group as compared with the ciprofloxacin group (22 versus 15 patients, p = 0.026). Ciprofloxacin eradicated 25 of 31 causative organisms, whereas ceftazi- dime eradicated 30 of 41 (p value not significant). Intravenous ciprofloxacin was at least as effec- tive as ceftazidime. Patients treated with cipro- floxacin may need added coverage for anaerobes, but the drug’s excellent activity against nosoco- mial pathogens and its availability in oral form allow for an early change to oral therapy without compromising effectiveness coupled with added savings and convenience.

S ince the introduction of the new oral fluoro- quinolone antimicrobials less than five years ago,

it has become clear that their high degree of in vitro and in vivo activity represents an advantage, espe- cially against difficult-to-treat nosocomial pathogens Cl]. In general, these agents have been well tolerated, even in patients with allergies to other commonly used antimicrobials. Several studies, including our own, demonstrated an equivalent degree of activity with parenteral third-generation cephalosporins in several clinical settings in mild to moderately ill patients who were still able to tolerate oral medication [2,3]. The development of intravenous formulations will now extend the use of these antimicrobials to severely ill patients who are unable to tolerate oral medication. In this study, using a prospective, controlled, random- ized protocol, we compared the effectiveness and safety of intravenous followed by oral ciprofloxacin with those of ceftazidime, a potent third-generation cephalosporin with a wide antibacterial spectrum in- cluding activity against Pseudomonas aeruginosa [4,5]. The patients studied had a variety of infections, all of which required intravenous therapy.

MATERIALS AND METHODS Both male and female patients with bacteremia;

moderate to severe infections of the respiratory tract, skin or soft tissue, bone, or joints; or intra-abdominal infection originating from the gastrointestinal or geni- tourinary tract were candidates for this study. Pa- tients experiencing urinary tract infections were in- cluded only when their infections were complicated, severe, or caused by nosocomial pathogens or P. aem- ginosa resistant to other antimicrobials. Patients who were pregnant, had severe underlying diseases, a his- tory of allergy to the study drugs, renal impairment, or a need for other antimicrobials with similar spec- trum were excluded, as were patients with meningi- tis, endocarditis, immunosuppression, chronic in- dwelling urinary catheters, or gross obstruction of the

November 30, 1989 The Amencan Journal of Mediune Volume 87 (suppl 5A) 5A-191s

SYMPOSIUM ON CIPROFLOXACIN /VILLAVICENCIO ET AL

TABLE I

Patient Characteristics in Both Treatment Groups

Treatment Group

Ciprofloxacin Ceftazidime p Value*

Total patients Sex

Male Female

Age i SD (years) Duration of illness (days) Duration of hospitalization (days) Duration of hospitalization after

therapy started (days) Sites of infectlon Severity

Moderate Severe

Aczpanylng disease

Yes Duration of treatment (days)

Intravenous Oral

26 26

i: :: 44.2 + 19.8 38.3 i- 16.5 12.9 i: 11.2 11.5 t 7.4 39.4 i: 29.6 20.1 i 11.3 18.9 t 14.4 24.1 t 22.8

28 28 NS

20 6

0.056

13

12.9 :3,8 5.6 5 3.0 7.3 t 3.7

20 0.032

14.1 ,“5.2 NS 11.5 * 4.5 <0.0005 2.6 + 3.8 <0.0005

NS

Iis” <0.005

NS

S = not significant. jtudent t test or Fisher’s exact test, as indicated

urinary tract. Written informed consent was obtained from all subjects.

Following enrollment into the study, patients were randomly assigned to receive either ciprofloxacin 200 mg intravenously every 12 hours or ceftazidime 1 g intravenously every eight to 12 hours. After a mini- mum of two to five days, intravenous therapy was dis- continued and the patients were given oral therapy. Patients who received intravenous ciprofloxacin started receiving oral ciprofloxacin 500 mg every 12 hours. For patients assigned to ceftazidime, a suitable oral drug (depending on site of infection and suscepti- bility testing) was prescribed. Treatment was given for seven to 14 days, although more prolonged ther- apy was permitted if clinically indicated. Concomitant therapy with an agent,active against anaerobes (me- tronidazole or clindamycin) was allowed in selected cases, when involvement of Bacteroides fragilis was suspected.

A careful medical history was obtained and a physi- cal examination performed before the beginning of therapy. Baseline culture specimens and results of laboratory tests were also obtained. The latter in- cluded a complete blood count, Sequential Multiple Analyzer Computer-18 (blood chemistries), and uri- nalysis. A pregnancy test was performed in women of reproductive age. Other tests, including chest radio- graphic examinations, biopsies, and other culture specimens were obtained as clinically indicated. After therapy was initiated, patients were observed daily, and cultures and follow-up laboratory tests were re- peated periodically. Culture specimens were also ob- tained at the end of the intravenous therapy period, at the completion of oral therapy, and during the post- treatment period if there was material to be cultured.

Clinical isolates were identified using conventional methods 161; gram-negative rods were identified using the analytical profile index-20 enteric system (An- alytab Products, Plainview, New York). Susceptibil- ity testing was performed for all isolates both by a standardized, single-disk method [7], and by determi-

nations of the minimal inhibitory concentration (by a microbroth dilution method for ciprofloxacin and by an agar dilution method for ceftazidime) [8].

Clinical responses of disappearance, marked or moderate reduction, and insufficient lessening of all signs and symptoms were classified as complete reso- lution, improvement, or failure, respectively. The signs and symptoms to be evaluated were clearly de- fined for each type of infection. Bacteriologic response was graded as eradication with or without recurrence, marked reduction, reinfection, persistence, or super- infection, according to standard definitions. Clinical or bacteriologic response was classified as indeterminate when it could not be assigned to any of the groups.

RESULTS A total of 52 patients (26 in each group) participated

in the study. Patients were comparable based on the following parameters: age, sex, infection status (acute or chronic), general status, need for surgery, presence of foreign bodies, prior treatment or use of concomi- tant antimicrobials (metronizadole or clindamycin). More patients treated with ceftazidime were classified as having severe infections (12 of 26 patients versus six of 26 patients; p = 0.056, Fisher’s exact test). However, accompanying diseases were more common in the ciprofloxacin group (19 disorders in 13 of 26 pa- tients versus 11 in six of 26 patients; p = 0.032, Fish- er’s exact test) (Table I). These diseases included dia- betes mellitus (three patients), neurologic (three), uri- nary (three), cardiovascular (three), or respiratory disorders (two), alcoholism (two), and other (three) in the ciprofloxacin group, and gastrointestinal (four), diabetes mellitus (three), hepatobiliary (two), and other (two) in the ceftazidime group. Duration of hos- pitalization was longer in the ciprofloxacin group, but most of this duration was accounted for in the pre- treatment days, since duration of treatment and dura- tion of hospitalization after therapy was started were similar (Table I). However, duration of intravenous treatment was significantly shorter for the ciprofloxa- tin group (p <0.0005, two-tailed Student t test). Un- doubtedly, this happened because of the lack of a suit- able oral drug other than ciprofloxacin in many cases treated with ceftazidime. A significantly greater num- ber of patients in the ceftazidime group had gram- positive organisms isolated from their infection site (11 of 26 patients versus three of 26 patients; p = 0.0114, Fisher’s exact test). Otherwise there was no difference in the type or number of isolates or suscep- tibility between the two treatment groups. Suscepti- bility data appear in Table II.

There was no difference between both groups re- garding the need for discontinuation of therapy. There were three fatalities during the study: two occurred in the ceftazidime group and were considered to be due to infection, whereas one occurred in the ciprofloxacin group during induction of anesthesia for debridement of a decubitus ulcer. There were two reinfections with ceftazidime (P. aeruginosa in both, plus Escherichia coli and Chromobacterium violaceum) and none with ciprofloxacin. There were three superinfections with ciprofloxacin (all Acinetobacter cadcoaceticus, sub- species anitratus, one of them resistant) and two with ceftazidime (P. aeruginosa and Proteus rni?-abilis). A total of 12 bacterial isolates obtained during or after therapy with ceftazidime were found to be resistant,

5A-192s November 30, 1989 The American Journal of Medicine Volume 87 (suppl 5A)

SYMPOSIUM ON CIPROFLOXACIN / VILLAVICENCIO ET AL

TABLE II

Susceptibility Testing of Clinical Isolates for Both Antibiotics*

Ciprofloxacin MIC &/ml)

0.015 0.03 0.06 0.12 0.25 0.5 0.75 1.0

Gram-posrtrve cocci (MICeo = 0.5) 3

Gram-negative rods (MICqo = 0.25)

2; 1: :

b

z : :, 0

Enterobacteriaceae 26 11 4 1 2 0 A Nonfermenters 0 1 3 1 2 3 0 1

Ceftazidime MIC &/ml)

0.5 1.0 2.0 4.0 8.0 16 32 64 2128

Gram-positive cocci (MI&, = 16) 1 i : 0 2 1 0 0 Gram-negative rods (MI& = 8) Enferobacteriaceae ii

i i 2’ ; A i ;

E, 0

Nonfermenters 1 0 0 1 0 0 0 MIC = minimal inhibitory concentration; MI& = 90 percent minimal inhibitory concentration. *Data for both treatment groups combined for each antimicroblal.

TABLE Ill

Site of Infection and Overall Assessment for Each Treatment Group*

Completely Successful Partially Successful Not Successful Indeterminate Total

Ciprofloxacin Urinary

: i 8

Gynecologic A i Respiratory Skin/soft trssues : i i

0 !

Intra-abdominal ! I! i

b 15

Bacteremia 0 i Total 19 5 3 1 28

Ceftazidrme Urinary Gynecologic d b

0

1 ! ii

Respiratory Skin/soft tissues II i ! 1; Intra-abdominal

: 2 i

Bacteremia i

0 : ; Total 20 2 0 28

*Two sites of infection occurred in each treatment group (see text).

including one of the superinfecting P. aeruginosa, four Enterococcus, two Staphylococcus aureus, two Citrobacter freundii, and three others, but only one resistant strain (Acinetobacter sp.) was isolated dur- ing therapy with ciprofloxacin (p = 0.0004, Fisher’s exact test). There was no significant difference be- tween both groups regarding overall assessment per patient (chi-square, 1.311; p value not significant) (Table III). However, more patients treated with cef- tazidime suffered adverse experiences (22 of 26 pa- tients versus 15 of 26 patients; p = 0.026, Fisher’s exact test), mainly in the form of phlebitis and very likely due to the longer duration of intravenous admin- istration that they required. Adverse experiences observed are shown in Table IV.

COMMENTS In this study, ciprofloxacin was at least as effective

as ceftazidime in the treatment of a variety of moder- ate to severe infections. This result is not surprising in view of the accumulated body of information regard- ing the comparability of ciprofloxacin levels after oral and intravenousadministration 191, the maintenance

of adequate serum levels with the dose utilized in this study [lO,ll], the adequate tissue penetration of the drug [12,131, and the observation of comparable re- sults with oral ciprofloxacin and intravenous cefotax- ime mentioned previously. Although, by chance, more patients were classified as having severe infections in the ceftazidime group, underlying disorders were more frequently seen in the ciprofloxacin-treated pa- tients, who had a longer duration of hospitalization prior to therapy; this latter fact may explain why more gram-positive organisms were seen in the cefta- zidime group! since gram-negative colonization tends to increase with increasing duration of hospitalization. However, hospital-acquired infection was considered to occur in only 14 of 26 ciprofloxacin-treated eom- pared with 13 of 26 ceftazidime-treated patients (p value not significant).

No differences were detected between the groups regarding clinical outcome. The eight bacteremias reported here happened to be due to Salmonella typhi; one patient treated with ceftazidime developed enterorrhagia after becoming afebrile and treatment was considered a partial success. Other third-genera-

November 30, 1989 The American Journal of Medicine Volume 87 (suppl 5A) 5A-193s

SYMPOSIUM ON CIPROFLOXACIN / VILLAVICENCIO ET AL

TABLE IV

Adverse Experiences Observed in Both Treatment Groups

Treatment Group

Ciprofloxacin Ceftazidime

Patients with adverse experiences 15 22 Clinical

Phlebitis 13 19 Nausea or vomitine. 0 1 Tachycardia ” Death

Laboratory abnormalities Considered significant Total laboratory abnormalitres observed Leukopenia Increased AST Hyperglycemia Increased creatinine Other

3T = aspartate aminotransferase.

tion cephalosporins have been tried in typhoid fever [141, but the lack of an oral alternative enabling early discharge from the hospital may be a drawback to their use. Ciprofloxacin orally has been used with suc- cess in typhoid fever [151, and typhoid carriers [16], and the availability of the intravenous preparation may allow for its use in severely ill patients, with the prospect of an early discharge on the oral formulation: results of ongoing trials may help to elucidate the role of short courses of therapy in this situation [17].

Most failures observed during this study occurred in patients with soft-tissue infections: in two patients with staphylococcal myositis, one with involvement of the shoulder joint, ciprofloxacin therapy was deemed a failure when persistently positive culture specimens did not clear; both resolved promptly with methicillin therapy. Infections of a leg stump and an abdominal wound occurred in two patients treated with ceftazi- dime who died of progressive infection while receiving therapy. The other failure was in the patient receiving ciprofloxacin for a complicated urinary tract infection, who died during induction of anesthesia for debride- ment of a decubitus ulcer; his urine culture specimen had been negative during therapy.

Another study has reported no significant differ- ence between ciprofloxacin and another broad-spec- trum agent, imipenem, in severely ill patients, having mainly lower respiratory tract infections [X3]. As in this study, no significant differences could be shown regarding clinical and bacteriologic efficacy and safety between ciprofloxacin and the comparative drug; the failures observed occurred in patients with fatal or ultimately fatal underlying disease.

There was no significant difference between the treatment groups with respect to the occurrence of reinfection or superinfection in this study. However, the appearance of resistant strains was more com- monly seen in ceftazidime-treated patients than in those treated with ciprofloxacin. This fact may have important implications in the development of overall resistance in the hospital environment.

Although it is true that there was no difference in total duration of hospitalization between groups after treatment was started, we were reluctant to send pa- tients home even when they were receiving oral ther- apy because of previous experience with difficulties in follow-up once the patients were discharged. Even if the patients are not discharged? the fact that oral therapy is possible in patients infected with gram- negative organisms resistant to other oral medications represents an advantage in terms of both savings and convenience to patients [19]. We conclude that intra- venous ciprofloxacin is as effective as a broad-spec- trum cephalosporin in the treatment of moderate to severe infections in patients requiring parenteral therapy, and has the advantage of providing an oppor- tunity for early change to the oral route.

ACKNOWLEDGMENT We thank Ileana Gonzalez, R.Ph., for her invaluable

help in the preparation and distribution of medications used in this study.

REFERENCES 1. Ramirez CA, Bran JL, Mejia CR, Garcia JF: Open, prospective study of the clinical efficacy of ciprofloxacin. Antimicrob Agents Chemother 1985; 28: 128-132. 2. Ramirez-Ronda CH, Saavedra S, Rivera-Vasquez CR: Comparative, double blind study of oral crorofloxacin and intravenous cefotaxime in skrn and skin structure infectrons. Am J Med i987; 82 (suppI 4A): 220-223. 3. Self PL, Zeluff BA, Sollo D, Gentry LO: Use of crprofloxacrn in the treatment of serious skrn and skin structure infections. Am J Med 1987; 82 (suppl 4A): 239-241. 4. Young LS: Ceftazrdime: summary of the First International Symposrum. J Antimicrob Chemother 1981; 8 (suppl B): 349-352. 5. Fong IW, Tomkins KB: Review of Pseudomonas aeruginosa meningitis with special em phasis on treatment with ceftazidime. Rev Infect Dis 1985; 7: 604-612. 6. lsenberg HD, WashIngton II JA, Balows A, Sonnenwirth AC: Collection, handling and processrng of specimens. In: Lennette EH, Balows A, Hausler WJ, Shadomy HJ, eds. Man- ual of clinical microbiology. Washington, DC.: American Society for Mrcrobiology, 1985; 73-98. 7. National Committee for Clinrcai Laboratory Standards: Standards for antimrcrobral disk susceptibility tests. M2-A2S2. Second supplement. Villanova, Pennsylvania: National Com- mittee for Clrnical Laboratory Standards, 1982. 8. National Committee for Cknical Laboratory Standards: Tentative standard: methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. M7-T. Villanova, Pennsylvania: National Committee for Clinical Laboratory Standards, 1983. 9. Drusano GL, Standrford HC, Plaisance K, Forrest A, Leslre J, Caldwell J: Absolute bioa- vailability of ciprofloxacin. Antimicrob Agents Chemother 1986; 30: 444-446. 10. Drusano GL, Plaisance Ki, Forrest A, Standrford HC: Dose rangrng study and constant Infusion evaluation of crprofloxacrn. Antimicrob Agents Chemother 1986; 30: 440-443. 11. Dudley MN, Ericson J, Zinner SH: Effect of dose on serum pharmacokinetrcs of intrave- nous ciprofloxacrn with identification and characterizabon of extravascular compartments using noncompartmental and compartmental pharmacokinetics models. Antimicrob Agents Chemother 1987; 31: 1782-1786. 12. LeBel M, Vallee F, Bergeron MG: Tissue penetration of ciprofloxacin after single and multiple doses. Antimrcrob Agents Chemother 1986; 29: 501-505. 13. Fong IW, Ledbetter WH, Vandenbroucke AC, Simbul M, Rahm V: Ciprofloxacin concen- trations in bone and muscle after oral dosing. Antimrcrob Agents Chemother 1986; 29: 405-408. 14. Soe GB, Overturf GD: Treatment of typhoid fever and other systemic salmonelloses with cefotaxime. ceftriaxone. cefoaerazone and other newer ceohalosoorins. Rev Infect DiS 1987; 9: 719-736. 15. Ramirez CA, Bran JL, Mejia CR, Garcia JF: Cknical efficacy of ciprofloxacin In typhoid fever. In: Neu HC, Weuta H, eds. Proceedings of the 1st International Ciprofloxacin Work- shop. Excerpta Medica, 1985; 365-369. 16. Ferreccro C, Morris JG, Valdivieso C, et at Efficacy of ciprofloxacrn in the treatment of chronic typhoid carriers. J Infect Dis 1988; 157: 1235-1239. 17. Carbon C, Weber P, Levy M, Boussougant Y, Cerf M: Shori-term crprofloxacrn therapy for typhoid fever (letter). J Infect DIS 1987; 155: 833. 18. Lode H, Wiley R, Hofken G, Wagner J, Borner K: Prospective randomrzed controlled study of ciprofloxacin versus imipenem-crlastatrn in severe clinical infecbons. Antimicrab Agents Chemother 1987; 31: 1491-1496. 19. Quintikani R, Cooper SW, Briceland LL, Nightrngale CH: Economic impact of streamlin- ing antrbiotic adminlstration. Am J Med 1987; 82 (suppl 4A): 391-394.

5A-194s November 30, 1989 The American Journal of Medicine Volume 87 (suppl 5A)