- Email: [email protected]
Comparative study of cefotetan and cefoxitin in the treatment of intra-abdominal infections
Comparative study of cefotetan and cefoxitin in the treatment of intra-abdominal infections Ronald T. Lewis, MB, BS,• Richard J. Duma, MD," Roger M. Echols, MD,' Joseph G. Jemsek, M.D,d Ahmad z. Najem, MD,e Robert A. Press, MD,r H. Harlan Stone, MD,• Gia T. Ton, MD," and Samuel E. Wilson, MD," Montreal, Quebec, Canada, Richmond, Virginia. Albany and .'\'ew forh. Nn1• forh. Charlotte, North Carolina, East Orange. New Jersey, Clet>rlrwd. Ohio. and Palos Verdes Estates. California One hundred eighty-eight patients were enrolled in a multicenter, randomized clinical trial to compare the safety and effectiveness of 1 to 2 gm cefotetan every 12 hours with those of 1 to 2 gm cefoxitin every 6 hours in patients with intra-abdominal infections. Most of the infections were community acquired, were associated with gastrointestinal tract perforation, and were caused by both anaerobic and aerobic bacteria. The median duration of therapy was 6 days for each group. The clinical response rate for the 95 evaluable patients in the cefotetan group was 98%, and that for the 43 evaluable patients in the cefoxitin group was 95%. Bacteriologically, 97% of the 58 evaluable patients in the cefotetan group and 89% of the 27 evaluable patients in the cefoxitin group had a satisfactory or presumed satisfactory response; two patients in the cefotetan group and three in the cefoxitin group were considered bacteriologic failures. Cefotetan was as effective as cefoxitin in eradicating Bacteroides tragi/is and other species of Bacteroides, Clostridium sp., and gram-negative bacilli. The incidence of treatment-related adverse reactions for cefotetan (27%) was not statistically different from that for cefoxitin (17%). No clinically significant differences were detected between the treatment groups in changes in the results of clinical laboratory tests performed before and after treatment; a decrease in hematocrit among the cefotetan group was statistically greater (p = 0.04) than that for the cefoxitin group, and a decrease in serum creatinine level for the cefoxitin group was greater than that for the cefotetan group (p = 0.02). Cefotetan may represent an effective. safe, and cost-saving alternative to cefoxitin for the prompt treatment of community-acquired intra-abdominal infections. (AM J 0BSTET GYNECOL 1988;158:728-35.)
Key words: Cefotetan, cefoxitin. communitv-acquired infections. intra-abdominal infections Intra-abdominal infections are usually mixed infections, caused by both anaerobic and aerobic bacteria. Bacteroides fragilis and other species of Bacteroides are the major anaerobic pathogen, and Escherichw mli is u~ually the predominant aerobe.~." These infections result from peritoneal contamination from the perf(lrated viscus; and the degree of contamination depends on the number and type of bacteria in the viscus. In the large intestine B. f"agilis and E. coli are the most numerous, i.e., 10"' to 10'" and 1W to 10"' per gram of feces, respectively.' Experimental studies of intraabdominal infections in animals have shown that aerobes, primarily E. coli, predominate in the peritoneal
From the Queen Elizabeth Hospital," Montreal; the Medical College of Virginia'; the Albany Medical College'; the l..!nivnsity of North Carolina School of Medicine"; ti!R Veteran~ Admini.1tration Mediwl Center,' East Orange; the New York Universil)' J1,1ediml Crnter;1 the Fairoiew General Hospital.' Cleveland; and the Harhm'/ ['niversity of California at Los Angeles Medical Center." Supported by grants from Stuart PharmaceuticaL'. Diviswn of ICJ Americas, Inc. Reprint requests: Ronald T. Lewis, MB, BS, Queen Elizabeth Jimpita!, 2100 Marlowe, Montreal, Quebec, Canada H4A-JU5.
Huid and the blood and are the main cause of death, whereas B. fragilis is the common pathogen in intraabdominal abscesses.',.-, Thus facultative anaerobic and aerobic bacteria may have a svnergistic relationship in causing these infections. Treatment of intra-abdominal infections usually involves surgical drainage and debridement of necrotic tissue as well as antibiotic therapy. Clinical studies have shown that optimal results are achieved when the drug regimen has an antibacterial spectrum active against both aerobic and anaerobic organisms." Combinations of antibiotics, which may include an aminoglycoside, are often administered and are highly effective; however, these regimens are not without the risk of adverse effects.''·., Several clinical trials have compared the efficacy of single agents with that of antibiotic combinations in patients with intra-abdominal infections." "'·' Age and nutritional status of the patient, duration of the acute infection, previous administration of other antibiotics, and prolonged hospitalilation strongly influence the choice of antibiotic regimen and outcome of the infection," and it has been suggested that single ·tge .. , . __ , n ts ~..:
728
Volume l5H Number~. Part 2
may not be appropriate for patients with these risk factors. Cefotetan is a new semisynthetic cephamycin antibiotic with a broad spectrum of activity against gramnegative aerobic bacteria and most gram-positive and anaerobic organisms of clinical significance. Because of its extended half-life in plasma, cefotetan may be administered in a twice-daily regimen for the treatment of most infections. Cefoxitin is an older cephamycin agent with a similar antibacterial spectrum but a shorter half-life, therefore necessitating more frequent dosing. This study was undertaken to compare the effectiveness and safety of these two agents in the treatment of hospitalized patients with intra-abdominal infections caused by organisms susceptible to cephalosporin-type antibiotics. Material and Methods
In a multicenter, randomized, comparative study, nine investigators at eight centers evaluated the safety and effectiveness of cefotetan and cefoxitin in the treatment of intra-abdominal infections. 'The protocol for this study was approved by the institutional review board at each center. Hospitalized patients of either sex with known or suspected intra-abdominal infections necessitating parenteral antimicrobial therapy. who were IH years of age or older and whose infection was likely to be susceptible to cephalosporins, were eligible for enrollment. Ineligible patients included pregnant women and nursing mothers; patients with a history of hypersensitivity or allergy to cephalosporins or penicillins; those with renal functional impairment characterized by a serum creatinine level of ~2.5 mg/ 100 ml; patients with underlying disease or a rapidly fatal disease that would preclude completion of treatment with the study drug; those likely to require concomitant therapy with other antimicrobial agents; patients at identifiable risk of life-threatening pseudomona! disease, such as those with acute leukemia or granulocytopenia; and patients with evidence of infectious disease of the central nervous system. After giving written informed consent, each patient was given a physical examination. and blood and urine samples were obtained for the following laboratory determinations: hematocrit, hemoglobin, white blood cell count and differential count, red blood cell count, platelets, prothrombin time, blood urea nitrogen, serum creatinine, alanine aminotransferase, aspartate aminotransferase, lactate dehydrogenase, alkaline phosphatase, total bilirubin, and urinalysis. Appropriate intra-abdominal cultures and blood samples were obtained before treatment for the isolation, identification, and susceptibility testing of causative organisms. The specimens were plated on 5% sheep blood agar and MacConkey's agar for aerobic and facultative anaerobic
Cefotetan vs cefoxitin in intra-abdominal infections
729
bacteria and on trypticase soy agar with sheep blood with gentamicin and Brewer's meat broth for anaerobic bacteria. After a minimum of 48 hours' incubation at 37° C in anaerobic jars, the anaerobic plates were examined and bacteriologic colonies were subcultured onto trypticase soy agar with sheep blood, gram stained, and examined microscopically. Susceptibilitv testing of facultative anaerobic and aerobic bacteria was performed according to the Kirby-Bauer disk-diffusion methods,~<"'/"'' with :~0 J.Lg disks used for both cefotetan and cefoxitin, or dilution methods (minimum inhibitory concentration or broth disk). The patients were assigned to one of the two treatment groups according to a predetermined randomization schedule in a ratio of 2: I for the cefotetan and cefoxitin groups. respectively. Cefotetan was administered intravenously at a dosage of I or 2 gm every I2 hours, according to the investigator's assessment of the severity of infection and severity of the patient's condition. Cefoxitin was also administered intravenously but at a dosage of I or 2 gm every 6 hours. The duration of treatment was 48 hours to 10 days for most patients, at the discretion of the investigator and on the basis of the patient's general condition, underlying conditions, severity of infection, and response to therapv. l'\o other antimicrobial agent was administered concomitantlv with the study drugs. The patients were observed daily for their clinical status and any potential adverse reactions. The highest daily body temperature, local intolerance to the infusions, and observed adverse effects were recorded. Cultures were obtained during therapy as appropriate, and the clinical laboratory tests were repeated during and at the end of therapy. Plasma concentrations of cefotetan were determined from blood obtained just before and immediaely after one of the two daily intravenous doses on day 2, 3, or 4 of cefotetan therapy. With an aseptic technique and in the arm opposite to that used for cefotetan administration, blood samples were obtained by venipuncture in Vacutainer blood-collecting tubes containing potassium oxalate (Becton Dickinson Vacutainer Systems, Rutherford, N.J.). The specimens were centrifuged within I hour; 3 ml of plasma was drawn off. put in a dry plastic tube (Falcon, Oxnard, Calif.), and frozen for later assay. Urine specimens were obtained as collective samples from 0 to 6 hours also on day 2, 3, or 4. Each sample was homogenized. and I 0 ml aliquots were removed and frozen for later assav. Concentrations of cefotetan in plasma and urine were determined at Hazelton-Raltech, Inc., Madison, Wis. by means of a hig·h-pressure liquid chromatographic assay. The clinical response to therapy was considered satisfactory if local and systemic signs and symptoms of infection were resolved or subsequentlv improved after
730
Lewis et al.
\l;m [, ] •IH~ Am
Table I. Patient population
Age
;r
Race White Black Hispanic Other Weight (pounds) Mean± SD Range Risk of infection Hi).(ht Low
= (}06).
e1l
Ce{oxitin
122
titi
95 (789()
·!3 (ti5o/r)*
78 (82%) 17 (18%)
35 (Hio/r)
76 (80%.) 19 (209()
35 (RI%)
4 7 (509()
19(H%) 16 (37%) 7 (16%) I (2o/r.)
:l/(39'/() 1.)1:\"i'i;) Upper gastrointestinal tract ;) l ("i1'7r) ~~ (!i:ic;) Lower or both upper and lower gastrointestinal tract Peritonitis or intra-abdominal abscess 7 (7'i() I ~~~~!
H (19C)t)
H ( 199(.)
:~2 (34o/r)
14 (15o/r) ~ (2o/r)
66 to receive cefoxitin. Twenty-seven patients (22'1r) in
16:~.5
± 39.1 97-286
156.8 ± 29.3
ii-240
'27 (63'/r)
46 (4H'X) 49 (52C!r)
Hi (37%)
*Mean difference in rate of nunevaluabilitv was statisticallv
significant (p
(;yrJ<'r
Table II. Site of infection Cr:fatetan
:'\o. of patients enrolled I\o. of patients evaluable Sex l\Iale Female
.J ( )b,tel
·
·
tlntra-abdominal abscesses. lower gastrointestinal tract infection (except uncomplicated appendicitis). and peritonitis.
treatment. A satisfactory bacteriologic response was indicated by eradication of the causative organism on the posttreatment culture, and response was presumed satisfactory for infections in which no material was available for culture because of clinical improvement. Statistical analysis of the data was carried out with standard procedures. The Wilcoxon rank-sum test was used to compare treatment groups for differences in pretreatment continuous variables and duration of therapy. Data on each patient for maximum daily ternperature and individual laboratory parameters were summarized by calculation of linear-regression coefficients (slopes), with day of therapy as the independent variable. and the distribution of slopes for each treatment group was tested for a mean of zero. In instances where one mean slope was significantly different from zero, a comparison of slopes between the two treatment groups was made by a Wilcoxon rank-sum test. A x" test was used to detect differences in discrete demographic variables between the groups and to compare the incidences of adverse experiences; Fisher's exact test was substituted for the x" test for small sample sizes. Clinical and bacteriologic response rates were an<~lyzed according to an overall x" rest; logistic regression analysis was used in the comparison of the groups with respect to variables such as investigator, age, sex, race, alcohol abuse, characteristics of infection, and general health condition.
Results The total enrollment in this study was 188 patients. 122 of whom were randomized to receive cefotetan and
the cefotetan group and 23 patients (35t;it) in tht~ cefoxitin group were excluded from the evaluation of efficacy, generally because of dosing errors. an unacceptable clinical diagnosis. or isolation of a resistant pathogen. The resistant pathogens isolated before tr·eatment included Pseudorrumas al'ruginosa from the peritoneal fluid of one patient in the cefotetan group and }"'ntrrobacter doaml', Bacteroides I'ulgatus. or J>,,Ptulomonas aerugirwsa from intra-abdominal swabs or drains from four patients in the cefoxitin group. The rate of nonevaluability for the cefoxitin group was significantlv greater than that for cefotetan (p = 0.06), but the reasons did not appear to bias the results in bvor of cef(Jtetan. Of the 138 evaluable patients, 95 (69~/c) were given cefotetan and 43 (:11% ), cefoxitin; these proportions were consistent with the predetermined :2: 1 randomization ratio. Most of the patients in each group were men <60 years of age, with a mean weight of approximately 160 pounds: the distribution of patients by race was similar in both groups (Table 1). Approximately half of the cefotetan group and two thirds of the cefoxitin group were considered to be at high risk of infection. i.e .. those with intra-abdominal abscesses, lower gastrointestinal infection, or peritonitis. The groups were comparable with regard to location of infection, i.e .. upper or lower (gastrointestinal tract) (Table II). Perforation of the small intestine or appendix was the most common diagnosis for each group: for many patients the large intestine and another abdominal organ were also perforated (Table Ill). Bacteremia associated with the intra-abdominal infection occurred at a comparable incidence for both groups. For most patients the infections were community acquired and considered moderate or severe. Approximately half the patients in both groups had had the infections for ""'1 dav and had been treated previously with other antibiotics, including penicillins, cephalosporins. trimethoprirnsulfamethoxazole, and doxycycline and either had failed to respond or had received the drugs for a short time before operation. For the most part the patients were in fair condition and had normal renal function. Alcohol abuse was an underlying condition for approximately 22% of the patients in both groups: < ~JJX of the patients in either group had a historv of diabetes
Cefotetan vs cefoxitin in intra-abdominal infections
\olurne l?iH
731
:\umber :1. Part 2
Table III. Clinical diagnosis*
}>erforated viscera Ruptured appendix Small intestine Larg-e intestine Small and I or large intestine and stomach Small and large intestine Small and/ or large intestine and rectum. bladder and/ or ureter (;astric. pyloric. or duodenal ulcer Stomach Stomach and kidney Pancreatic cyst Perf(>rated viscus (appendix, sigmoid colon or sigmoid diverticulum) and bacteremia Intra-abdominal abscess Intra-abdominal abscess and bacteremia Appendicitis Cholecystitis Peritonitis Cholelithiasis Gangrenous appendix Ileocolonic fistula Diverticulitis Prepyloric ulcer Gangrenous colon
Cej(,tetan
Cefoxitin
(n = 95)
(n = 43)
:>5 (5R'7r)
22 (51'7r) 7 7 2
II II
5 4 3
()
0
2 15* 2 I I l (l'!i)
2 2 () ()
3 (7%)
14 (15%)*
6 (14'7r)
2 (2%) 12 (13%)
()
5 (5%)
I (2%)
4 (4%)* 1 (1%) l (1%)
3 (7%) 2 (4%) 0 3 (7'7r)
I(!%)
()
0
I (2%)
()
I (2%)
0
I (2%)
*One of these infections was known to be hospital acquired.
mellitus or malignancy. No statistically significant differences were detected between the groups for any of the variables for demographics, clinical diagnosis and infection history, or past medical history and pretreatment health status (p > 0.05). The differences for most pretreatment laboratory values between the groups were not statistically significant; the mean level of serum creatinine for the patients in the cefoxitin group ( 1.3 mg/ dl) was slightly higher than the value for those in the cefotetan group ( 1.1 mg/ dl), a difference that approached statistical significance (p = 0.07). The mean oral body temperatures taken before treatment were similar for the two groups, i.e., I 00.0° ± !. 7o F for the cefotetan group and I 00.0° ± 1.8° F for the cefoxitin group. The respective mean estimates of daily temperature change (slope) were -0.31° ± 0.29° and -0.29° ± 0.37° F. Among the evaluable patients 86 of the 95 in the cefotetan group received 2 gm every 12 hours and nine received 1 gm every 12 hours; 40 patients in the cefoxitin group were given 2 gm every 6 hours and the other three received I gm every 6 hours. The median total dose of cefotetan was 20 gm (mean = 23 gm), and that for cefoxitin was 42 gm (mean = 45 gm). The median duration of therapy was 6 days for each group. The clinical response to treatment with both drugs was excellent; 93 of the 95 evaluable patients in the cefotetan group (98%) and 41 of the 43 evaluable patients in the cefoxitin group (95%) either had resolution
of clinical signs and symptoms or had improved substantially (Table IVA). No statistically significant difference in response rates was detected between the groups (p = 0.59). Of the 138 evaluable patients 53 were not included in the bacteriologic evaluation as no pathogens could be isolated before treatment. These included 37 of the cefotetan group, in whom the signs and symptoms of infection cleared in 32 and improved in five, and 16 of the cefoxitin group, all of whom were clinically cleared. These patients were treated primarily for appendicitis, gangrenous appendix, perforated small bowel, or perforated prepyloric or duodenal ulcer. Fifty-six of the 58 bacteriologically evaluable patients in the cefotetan group (97%) had a satisfactory or presumed satisfactory bacteriologic response, as did 24 of the 27 (89%) in the cefoxitin group (Table IVB). This difference in bacteriologic response rate was not statistically significant (p = 0.32). Table V gives the bacteriologic response according to the pathogens isolated before treatment. A total of 177 bacterial isolates was cultured among the 58 bacteriologically evaluable patients in the cefotetan group, an average of 3.0 isolates per patient; 84 were anaerobic bacteria, an average of 1.4 per patient. In the cefoxitin group the 27 bacteriologically evaluable patients had a total of 101 isolates, an average of 3.7 per patient and 49 were anaerobes. an average of 1.8 per patient. E. coli and Streptococcus sp. were the most common
732
Lewis et al.
M.nch I~JKI<
Am
J
Ohstet Cnw< ol
Table IVA. Clinical responses Ccfoxitiu .JJJ
Cejotf'lrw (n = 95)
Cleared Improved Unchanged or worse
ill (85.3%)} ')'' 12 (12.6%)
• .>
Ill =
("~ ()CI) CI/ .• /(
39(90.7'.il)41 2 (4.7'/()
f
"'J'l
(9.7 .. ;(.)
2 (4.7'7r)
2 (2.1%)
Table IVB. Bacteriologic responses*
Satisfactory or presumed satisfactor~· Unsatisfactory
Cefotetrm
Cefoxitin
(11 = 58)
(II=
56 (96.6'7r)
2·4 (88.9%)
2 (3.4%)
3 ( 11.1 '7r)
27)
*Bacteriologic response not evaluated for 53 patients.
aerobic bacteria isolated before treatment; B. fragilis. other species of Bacteroides, and Clostridium sp. were the most frequent anaerobic pathogens. Bacteroides ovatus was isolated from the blood of each of the three bacteremic patients in the cefotetan group; B. fragilis, Bacteroides sp .. or gram-positive cocci were the pathogens in the three patients with bacteremia who received cefoxitin. In the cefotetan group 89 of the 93 aerobic isolates (96%) were eradicated, as were 77 of the 84 anaerobes (92% ), including all three isolates from blood. In the cefoxitin group 38 of 46 aerobes (83%) were eradicated, as were 44 of 49 anaerobic bacteria (90%), including all three pathogens isolated from the blood. Two patients in the cefotetan group showed no clinical improvement and had an unsatisfactory bacteriologic response. One was a 27-year-old Hispanic man with an appendiceal abscess from which Streptococcus sp. (group D), B. fragilis, Clostridium sp., and Eubacterium limosum were cultured from peritoneal fluid before therapy. At the end of treatment the B. fragilis and Clostridium sp. had been eradicated, but he developed a subfascial abscess from which Streptococcus sp., Eubacterium sp., and coagulase-negative Staphylococcus epiderrnidis were cultured. The other clinical and bacteriologic failure occurred in a 22-year-old white man admitted with a ruptured appendix from which Streptocornts sp. (group D), E. roli, B. ji·agilis, Bacteroides sp., and Jo:u/)(Jcteriurn sp. were isolated. He subsequently developed a supralevator abscess, but no pathogens could be isolated from it. Two patients in the cefoxitin group were clinical failures, and three were considered bacteriologic failures. A 49-year-old Hispanic man was admitted with an appendiceal abscess from which Streptococcus sp. (group D) and£. coli were cultured before treatment. He subsequently developed a subfascial abscess that grew Streptococcus sp. (group D). The other clinical failure in the cefoxitin group occurred in a 65-year-old white woman with a perforated sigmoid diverticulum
and bacteremia. Before therapy, Staphylococcus aureus, E. coli, Klebsiella oxytoca, B. fragilis, and Clostridium sp. were isolated from peritoneal fluid, and B. fragilis and Bacteroides sp. were isolated from the blood. The patient developed an abdominal abscess and underwent a laparotomy; B (ragilis, Lactobacillus sp .. and Pseudomonas sp. were isolated from a culture of peritoneal Huid at that time. The third patient was considered a bacteriologic failure but a clinical success. This 34-year-old white man was admitted with a perforatitd appendix with abscess. Pretreatment cultures of peritoneal fluid grew Strep· tococcus sp .. E. coli, Pseudomona.\ aeruginosa, Bacteroides sp., Bacteroides vulgatus, and Peptocorrus sp. The patient improved slowly, but cultures from the drain site continued to grow StTeptococcus sp. (group D and nongroup D), E. coli, Pseudomonas aeruginosa, Staphylococcus au reus, B. fragilis, and Bacteroides sp. after cefoxitin therapy was stopped. The mean concentrations of cefotetan in serum from 30 patients after drug administration on days 2. 3, and 4 ranged from 117.7 to 223.2 f,lg/ml, and the per· centage of the dose recovered as tau to mer was bet ween 8.2 and 15.4 (Table VI). In urine samples obtained from two patients 0 to 6 hours after drug administration on day 4, the mean level of cefotetan was 2673.5 f,lg/ ml, and its tautomer accounted for II. I 7c of the dose. The only statistically significant differences in mean slopes for the clinical laboratory determinations performed before and after treatment were those for hematocrit and serum creatinine. The fall in hematocrit for cefotetan (- 1. 122) was greater than that for cefoxitin (- 0.444) (p = 0.04), and the decrease in serum creatinine for cefoxitin (- 0.038) was almost twice that for cefotetan (- 0.020) (p = 0.02). Neither difference was considered clinically significant. All 121 patients in the cefotetan group and 66 in the cefoxitin group treated during the swdy were included in the evaluation of safe tv; no safety information was
Cefotetan vs cefoxitin in intra-abdominal infections
Volume 158 Number 3, Part 2
733
Table V. Bacteriologic response of pathogens isolated before treatment Crj(;/etan
Cefoxitin
isolated
Satisfactory or presumed 1atis{artm-y
4
4
I H 26
I ~4
4 12
2
2
1
31 5 0 2 I I
30
20
5
4
I l
.~.Vo.
1.Vo. isolated
Satisjrtctory or presumed satisfactory
Aerobic bacteria Gram-positive cocci Staphylococcus aureus (coagulase positive) Streptococcus pneumoniae Streptococcus sp. group D Streptococcus sp.
Unspecified gram-positive cocci Gram-negative bacilli Escherichia coli Klebsiella pneumoniae Klebsiella ox~toca Enterobacter cloacae Enterobacter gergoviae Enterobacter sp. Serratia marcescens Proteus mirabilis Pseudomonas aeruginosa Pseudomonas vesicularis Eikenella corrodens Haemophilus irifluenzae
Unspecified gram-negative rods Gram-positive bacilli Diphtheroids Anaerobic bacteria Gram-positive cocci Peptococcus sp. PeptostreptococcUI sp. Gaffkya anaerobius
Unspecified anaerobic gram-positive cocci Gram-negative cocci
4 I
0 2
7
I 0
()
()
:I I! I 17 4
()
2
!
2
0
()
I I
0 1
()
I I ·l:
0 0
0
3
2
I 0 2
()
()
1
! ()
I
I
()
()
()
()
2
2
()
()
I
l 4 I I
2 3 0
~~ ()
1
I
()
()
4 I I
V eillonella parvula
!
Gram-positive bacilli Actinobacillus israeli Bifidobacterium adolescentis Bifidobacterium sp. Clostridium peifringens Clostridium symbiosum Clostridium tertium Clostridium sp. Eubacterium sp.
Anaerobic gram-positive non-spore-forming rods Gram-negative bacilli B. fragilis Bacteroides ovatus Bacteroides vulgatus Bactermdes distasonis Bacteroides thetaiotaomicron Bacteroides melaninogenicus Bacteroides sp. Fusobacterium sp. Fusobacterium necrophorum Fusobacterium variurn
Anaerobic gram-negative rods Other pathogens CDC GR 3452
2
2
()
()
I
()
()
I 4
! !
0
()
4
3
:>
0
()
()
0 .S
1 1 !2
I I
6
7
9 0
()
2 !
!0 !0
8 10
H 4
5
5
2
4
4
3 1
;)
1 I 1
14
4 0 1 I 0
I 13 4 ()
I
I
II 2
7 4
I I I I
8
7
()
()
1 0 0
()
! 0
()
*Some patients had more than one pathogen isolated.
available for one patient in the cefoxitin group. Adverse reactions occurred in 33 patients (27%) in the cefotetan group and in II patients (I7%) in the cefoxitin group. The difference in these incidences was not statistically
significant (p = O.IO) (Table VII). These reactions were considered related to the drug only in five patients in the cefotetan group (4%) and in four patients in the cefoxitin group (6%). This difference was not statisti-
73L-
r_ewis et al.
\larch 1~lkS .\111 .J Ol»il'l ( ;\'lll'cul
Table VI. Plasma and urine levels of cefotetan ,\'o. fif patient.\ Plasma Dav ~ Da} :> Day 4 Urine
5 13 9
Day~
Day 4
~
JI,J ean level (J.Lglm/)
Table VII. Adverse reactions (:tji>ll'lllll
and range
Tautomer (%! and range
223.3 (167-298) 179.9 (11-255) 117.7 (52-231)
8.2 (1.6-25.2) 15.4 (0-29.0) 15.0 (3.7-22.1)
974 2674 (657-4690)
11.1 (3.7-18.7)
0
cally significant (p = 0.72). One patient receJVmg cefotetan developed hives on the face and neck after receiving his second dose of the drug and experienced pruritus of the face and neck, without rash, after the third dose. No symptoms developed after the fourth dose of cefotetan, but therapy was discontinued. Three patients in the cefotetan group died during the studv, but no death was considered related to treatment. One death occurred in a 71-year-old man with a history of angina and pulmonary fibrosis of unknown cause who was treated for cholecystitis; he died 1 month after completion of therapy from end-stage pulmonary fibrosis and cardiac arrest. The second death occurred in a 49-year-old man with a history of alcoholic cirrhosis and end-stage liver disease who received two doses of cefotetan because of spontaneous bacterial peritonitis before he died of hepatorenal failure, septic shock, and aspiration. The third patient was a 94-year-old woman with a history of coronary artery disease who was treated for a perforated duodenal ulcer and peritonitis and died on the fifth day of cefotetan therapy of a suspected myocardial infarction or pulmonary embolus.
Comment In this population of predominantly young men with community-acquired intra-abdominal infections after gastrointestinal tract perforation, prompt treatment with the less complicated dosage regimen of cefotetan of 2 gm every 12 hours was as effective as cefoxitin administered at 2 gm every 6 hours. With sample sizes of 95 evaluable patients in the cefotetan group and 43 in the cefoxitin group and an assumed rate of success for cefoxitin of 95%, the probability is 0.37 that a 10% lower success rate for cefotetan would have been detected and 0.65 that a 15% lower rate would have been detected. Of the total of 138 patients evaluated, only two patients in each group failed to respond. These treatment success rates are higher than those reported in previous studies of cefotetan in patients with established intraabdominal infections but without adjunct surgeri 11 "
Patienh with no adverse reactions Patients with at least one adverse reaction Patients thought to have a drug-related reaction No. reported (treatment related) Symptoms Phlebitis at the injection site Edema Itching
Rash l'rticaria I lematuria I'Hlria Diagnoses Death Laboratory results Increased eosinophils lncreased partial t h rmnboplastin time Increased prot It rom bin time Thrombocvthemia or thro~bocytosis Increased white blood cell count I ncrcased liver function tests Urinarv casts
R:-l/1~1 (7~.7~;;.)
')!)/(if)
:r11Jn
I lihti i 1!1. ,f.,,:)
')I[~]
(~7.:1~0
H.IVr l
(:•(!.:~(_;;)
4/()1) jl), [
~(~I
0
l (II :l (1) I I (II I I
0 II
<;
i
I)
II I) I)
:1
II
:l (II l
:l (I i 0
R
:l
II (]I
:1 (I)
2:1 !:"\)
li(:l)
0
*:-.lo intinmation available for one patient in the ce!cJxitin group.
and similar to the results of other studies in patients with peritonitis.'' Therapeutic response rates in studies of third-generation cephalosporins in patients with polymicrobial peritonitis and peritoneal abscess were lower.'· The bacteriologic cause of these infections was similar to that in other reports of intra-abdominal infections.~. "· '' Most infections were polymicrobial; approximately half the isolates were anaerobes, and E. coli was the most common aerobic pathogen, and B. fragilis and other Bacternides sp. were the most frequently isolated anaerobic pathogens. Cefotetan was as effective as cefoxitin in achieving a satisfactory bacteriologic response in patients with infections caused by these bacteria. All 4 7 isolates of B. jr·agilis and its subtypes were eradicated after cefotetan therapy, as were all but one of the 50 gram-negative bacilli. These clinical results support the good in vitro activitY reported for cefotetan against such organisms.,~. ,-, Cefotetan was also well tolerated in this study. The incidence of adverse reactions related to the drug was slightly lower than that of reactions related to treatment with cefoxitin, although this difference was
Volume 158 Number :1. Part 2
not statistically significant, With its less complicated dosing schedule, equivalent effectiveness, and marginally lower toxicity, cefotetan may represent an advantageous alternative to cefoxitin in the treatment of promptly diagnosed polymicrobial, communityacquired intra-abdominal infections. REFERENCES L Dunn DL, Simmons RL The role of anaerobic bacteria in intraabdominal infections. Rev Infect Dis 1984;6(suppl 1):5139-46. 2. Lorber B, Swenson RM. The bacteriology of intraabdominal infections. Surg Clin North Am 1975;55: 1349-4. 3. Ho JL. Barza M. Role of aminoglycoside antibiotics in the treatment of intra-abdominal infection. Amimicroh Agents Chemother 1987;31:485-91. 4. Onderdonk AB, Shapiro ME, Finberg RW, Zaleznick DF, Kasper DL. Use of a model of intraabdominal sepsis for studies of the pathogenicity of Bacteroides fragilis. Rev Infect Dis 1984;6(suppl I):S91-5. 5. Weinstein WM, Onderdonk AB, BartlettJG, Gorbach SL. Experimental intra-abdominal abscesses in rats: development of an experimental model. Infect Immun 1974; 10:1250-9. 6. Tally FP. Gorbach SL Therapy of mixed anaerobicaerobic infections. Lessons from studies of intraabdominal sepsis. Am J Med 1985;78(suppl 6A): 145-53. 7. Drusano GL. Warren JW, Saah AJ, et a!. A prospective randomized controlled trial of cefoxitin versus clindamycin-aminoglycoside in mixed anaerobic-aerobic infections. Surg Gynecol Obstet 1982;154:715-20.
Cefotetan vs cefoxitin in intra-abdominal infections
735
8. Tally FP, Kellum JM, Ho JL. O'Donnell TF, Barza M, Gorbach SL. Randomized prospective stud} comparing moxalactam and cefoxitin with or without tobramycin for the treatment of serious surgical infections. Antimicrob Agents Chemother !986:29:244-9. 9. Schentag JJ, Wels PB, Reitberg DP, Walczak P. Van Tyle JH, Lascola RJ. A randomized clinical trial of moxalactam alone versus tobramycin plus clindamvcin in abdominal sepsis. Ann Surg 1985:198:35-41. lOa. Bauer WW. Kirby WMM. Sherris .JC, Turck :VI. Antibiotic susceptibility testing by a standardized single disc method. Am .J Clin Pathol 1966:45:493-6. lOb. Standardized disc susceptibility test. Fed Reg- 1972;37: 20527-9. JOe. National Committee for Clinical Laboratorv Standards. Approved standard: ASM-2, Performance standards for antimicrobial disc susceptibility tests, .July 1975. II. Fujimoto M, Ueda T, Sakai K. et al. A clinical trial on cefotetan (YM 09330) in the field of surger~. Chemotherapy 1982:30:817-26. 12. YuraJ, Shinagawa N. Ishikawa S. et al. Fundamental and clinical studies of ceforetan (YM 09330) on the surgical field. Chemotherapy 1982:30:786-795. 13. Wilson SE, Lewis RT, Ton GT. Comparative clinical evaluation of cefotetan with moxalactam in the treatment of hospitalised patients with serious intra-abdominal infections. [Abstracts]. In: Proceedings of the fourteenth international congress of chemotherapy. Kvoto. Tokvo: University of Tokyo Press, 1985:86. 14. Ruckdeshel G. Activity in 1•itru of cefotetan against nonsporing anaerobes: a comparative study . .J Antimicrob Chemother 1983; ll(suppl A): II 7-24. 15. Watt B, Brown FV. The comparative in-1•itro activity of cefotetan against anaerobic bacteria . .J Antimicrob Chemother 1985;15:671-7.
Key words: Cefotetan, cefoxitin. communitv-acquired infections. intra-abdominal infections Intra-abdominal infections are usually mixed infections, caused by both anaerobic and aerobic bacteria. Bacteroides fragilis and other species of Bacteroides are the major anaerobic pathogen, and Escherichw mli is u~ually the predominant aerobe.~." These infections result from peritoneal contamination from the perf(lrated viscus; and the degree of contamination depends on the number and type of bacteria in the viscus. In the large intestine B. f"agilis and E. coli are the most numerous, i.e., 10"' to 10'" and 1W to 10"' per gram of feces, respectively.' Experimental studies of intraabdominal infections in animals have shown that aerobes, primarily E. coli, predominate in the peritoneal
From the Queen Elizabeth Hospital," Montreal; the Medical College of Virginia'; the Albany Medical College'; the l..!nivnsity of North Carolina School of Medicine"; ti!R Veteran~ Admini.1tration Mediwl Center,' East Orange; the New York Universil)' J1,1ediml Crnter;1 the Fairoiew General Hospital.' Cleveland; and the Harhm'/ ['niversity of California at Los Angeles Medical Center." Supported by grants from Stuart PharmaceuticaL'. Diviswn of ICJ Americas, Inc. Reprint requests: Ronald T. Lewis, MB, BS, Queen Elizabeth Jimpita!, 2100 Marlowe, Montreal, Quebec, Canada H4A-JU5.
Huid and the blood and are the main cause of death, whereas B. fragilis is the common pathogen in intraabdominal abscesses.',.-, Thus facultative anaerobic and aerobic bacteria may have a svnergistic relationship in causing these infections. Treatment of intra-abdominal infections usually involves surgical drainage and debridement of necrotic tissue as well as antibiotic therapy. Clinical studies have shown that optimal results are achieved when the drug regimen has an antibacterial spectrum active against both aerobic and anaerobic organisms." Combinations of antibiotics, which may include an aminoglycoside, are often administered and are highly effective; however, these regimens are not without the risk of adverse effects.''·., Several clinical trials have compared the efficacy of single agents with that of antibiotic combinations in patients with intra-abdominal infections." "'·' Age and nutritional status of the patient, duration of the acute infection, previous administration of other antibiotics, and prolonged hospitalilation strongly influence the choice of antibiotic regimen and outcome of the infection," and it has been suggested that single ·tge .. , . __ , n ts ~..:
728
Volume l5H Number~. Part 2
may not be appropriate for patients with these risk factors. Cefotetan is a new semisynthetic cephamycin antibiotic with a broad spectrum of activity against gramnegative aerobic bacteria and most gram-positive and anaerobic organisms of clinical significance. Because of its extended half-life in plasma, cefotetan may be administered in a twice-daily regimen for the treatment of most infections. Cefoxitin is an older cephamycin agent with a similar antibacterial spectrum but a shorter half-life, therefore necessitating more frequent dosing. This study was undertaken to compare the effectiveness and safety of these two agents in the treatment of hospitalized patients with intra-abdominal infections caused by organisms susceptible to cephalosporin-type antibiotics. Material and Methods
In a multicenter, randomized, comparative study, nine investigators at eight centers evaluated the safety and effectiveness of cefotetan and cefoxitin in the treatment of intra-abdominal infections. 'The protocol for this study was approved by the institutional review board at each center. Hospitalized patients of either sex with known or suspected intra-abdominal infections necessitating parenteral antimicrobial therapy. who were IH years of age or older and whose infection was likely to be susceptible to cephalosporins, were eligible for enrollment. Ineligible patients included pregnant women and nursing mothers; patients with a history of hypersensitivity or allergy to cephalosporins or penicillins; those with renal functional impairment characterized by a serum creatinine level of ~2.5 mg/ 100 ml; patients with underlying disease or a rapidly fatal disease that would preclude completion of treatment with the study drug; those likely to require concomitant therapy with other antimicrobial agents; patients at identifiable risk of life-threatening pseudomona! disease, such as those with acute leukemia or granulocytopenia; and patients with evidence of infectious disease of the central nervous system. After giving written informed consent, each patient was given a physical examination. and blood and urine samples were obtained for the following laboratory determinations: hematocrit, hemoglobin, white blood cell count and differential count, red blood cell count, platelets, prothrombin time, blood urea nitrogen, serum creatinine, alanine aminotransferase, aspartate aminotransferase, lactate dehydrogenase, alkaline phosphatase, total bilirubin, and urinalysis. Appropriate intra-abdominal cultures and blood samples were obtained before treatment for the isolation, identification, and susceptibility testing of causative organisms. The specimens were plated on 5% sheep blood agar and MacConkey's agar for aerobic and facultative anaerobic
Cefotetan vs cefoxitin in intra-abdominal infections
729
bacteria and on trypticase soy agar with sheep blood with gentamicin and Brewer's meat broth for anaerobic bacteria. After a minimum of 48 hours' incubation at 37° C in anaerobic jars, the anaerobic plates were examined and bacteriologic colonies were subcultured onto trypticase soy agar with sheep blood, gram stained, and examined microscopically. Susceptibilitv testing of facultative anaerobic and aerobic bacteria was performed according to the Kirby-Bauer disk-diffusion methods,~<"'/"'' with :~0 J.Lg disks used for both cefotetan and cefoxitin, or dilution methods (minimum inhibitory concentration or broth disk). The patients were assigned to one of the two treatment groups according to a predetermined randomization schedule in a ratio of 2: I for the cefotetan and cefoxitin groups. respectively. Cefotetan was administered intravenously at a dosage of I or 2 gm every I2 hours, according to the investigator's assessment of the severity of infection and severity of the patient's condition. Cefoxitin was also administered intravenously but at a dosage of I or 2 gm every 6 hours. The duration of treatment was 48 hours to 10 days for most patients, at the discretion of the investigator and on the basis of the patient's general condition, underlying conditions, severity of infection, and response to therapv. l'\o other antimicrobial agent was administered concomitantlv with the study drugs. The patients were observed daily for their clinical status and any potential adverse reactions. The highest daily body temperature, local intolerance to the infusions, and observed adverse effects were recorded. Cultures were obtained during therapy as appropriate, and the clinical laboratory tests were repeated during and at the end of therapy. Plasma concentrations of cefotetan were determined from blood obtained just before and immediaely after one of the two daily intravenous doses on day 2, 3, or 4 of cefotetan therapy. With an aseptic technique and in the arm opposite to that used for cefotetan administration, blood samples were obtained by venipuncture in Vacutainer blood-collecting tubes containing potassium oxalate (Becton Dickinson Vacutainer Systems, Rutherford, N.J.). The specimens were centrifuged within I hour; 3 ml of plasma was drawn off. put in a dry plastic tube (Falcon, Oxnard, Calif.), and frozen for later assay. Urine specimens were obtained as collective samples from 0 to 6 hours also on day 2, 3, or 4. Each sample was homogenized. and I 0 ml aliquots were removed and frozen for later assav. Concentrations of cefotetan in plasma and urine were determined at Hazelton-Raltech, Inc., Madison, Wis. by means of a hig·h-pressure liquid chromatographic assay. The clinical response to therapy was considered satisfactory if local and systemic signs and symptoms of infection were resolved or subsequentlv improved after
730
Lewis et al.
\l;m [, ] •IH~ Am
Table I. Patient population
Age
;r
Race White Black Hispanic Other Weight (pounds) Mean± SD Range Risk of infection Hi).(ht Low
= (}06).
e1l
Ce{oxitin
122
titi
95 (789()
·!3 (ti5o/r)*
78 (82%) 17 (18%)
35 (Hio/r)
76 (80%.) 19 (209()
35 (RI%)
4 7 (509()
19(H%) 16 (37%) 7 (16%) I (2o/r.)
:l/(39'/() 1.)1:\"i'i;) Upper gastrointestinal tract ;) l ("i1'7r) ~~ (!i:ic;) Lower or both upper and lower gastrointestinal tract Peritonitis or intra-abdominal abscess 7 (7'i() I ~~~~!
H (19C)t)
H ( 199(.)
:~2 (34o/r)
14 (15o/r) ~ (2o/r)
66 to receive cefoxitin. Twenty-seven patients (22'1r) in
16:~.5
± 39.1 97-286
156.8 ± 29.3
ii-240
'27 (63'/r)
46 (4H'X) 49 (52C!r)
Hi (37%)
*Mean difference in rate of nunevaluabilitv was statisticallv
significant (p
(;yrJ<'r
Table II. Site of infection Cr:fatetan
:'\o. of patients enrolled I\o. of patients evaluable Sex l\Iale Female
.J ( )b,tel
·
·
tlntra-abdominal abscesses. lower gastrointestinal tract infection (except uncomplicated appendicitis). and peritonitis.
treatment. A satisfactory bacteriologic response was indicated by eradication of the causative organism on the posttreatment culture, and response was presumed satisfactory for infections in which no material was available for culture because of clinical improvement. Statistical analysis of the data was carried out with standard procedures. The Wilcoxon rank-sum test was used to compare treatment groups for differences in pretreatment continuous variables and duration of therapy. Data on each patient for maximum daily ternperature and individual laboratory parameters were summarized by calculation of linear-regression coefficients (slopes), with day of therapy as the independent variable. and the distribution of slopes for each treatment group was tested for a mean of zero. In instances where one mean slope was significantly different from zero, a comparison of slopes between the two treatment groups was made by a Wilcoxon rank-sum test. A x" test was used to detect differences in discrete demographic variables between the groups and to compare the incidences of adverse experiences; Fisher's exact test was substituted for the x" test for small sample sizes. Clinical and bacteriologic response rates were an<~lyzed according to an overall x" rest; logistic regression analysis was used in the comparison of the groups with respect to variables such as investigator, age, sex, race, alcohol abuse, characteristics of infection, and general health condition.
Results The total enrollment in this study was 188 patients. 122 of whom were randomized to receive cefotetan and
the cefotetan group and 23 patients (35t;it) in tht~ cefoxitin group were excluded from the evaluation of efficacy, generally because of dosing errors. an unacceptable clinical diagnosis. or isolation of a resistant pathogen. The resistant pathogens isolated before tr·eatment included Pseudorrumas al'ruginosa from the peritoneal fluid of one patient in the cefotetan group and }"'ntrrobacter doaml', Bacteroides I'ulgatus. or J>,,Ptulomonas aerugirwsa from intra-abdominal swabs or drains from four patients in the cefoxitin group. The rate of nonevaluability for the cefoxitin group was significantlv greater than that for cefotetan (p = 0.06), but the reasons did not appear to bias the results in bvor of cef(Jtetan. Of the 138 evaluable patients, 95 (69~/c) were given cefotetan and 43 (:11% ), cefoxitin; these proportions were consistent with the predetermined :2: 1 randomization ratio. Most of the patients in each group were men <60 years of age, with a mean weight of approximately 160 pounds: the distribution of patients by race was similar in both groups (Table 1). Approximately half of the cefotetan group and two thirds of the cefoxitin group were considered to be at high risk of infection. i.e .. those with intra-abdominal abscesses, lower gastrointestinal infection, or peritonitis. The groups were comparable with regard to location of infection, i.e .. upper or lower (gastrointestinal tract) (Table II). Perforation of the small intestine or appendix was the most common diagnosis for each group: for many patients the large intestine and another abdominal organ were also perforated (Table Ill). Bacteremia associated with the intra-abdominal infection occurred at a comparable incidence for both groups. For most patients the infections were community acquired and considered moderate or severe. Approximately half the patients in both groups had had the infections for ""'1 dav and had been treated previously with other antibiotics, including penicillins, cephalosporins. trimethoprirnsulfamethoxazole, and doxycycline and either had failed to respond or had received the drugs for a short time before operation. For the most part the patients were in fair condition and had normal renal function. Alcohol abuse was an underlying condition for approximately 22% of the patients in both groups: < ~JJX of the patients in either group had a historv of diabetes
Cefotetan vs cefoxitin in intra-abdominal infections
\olurne l?iH
731
:\umber :1. Part 2
Table III. Clinical diagnosis*
}>erforated viscera Ruptured appendix Small intestine Larg-e intestine Small and I or large intestine and stomach Small and large intestine Small and/ or large intestine and rectum. bladder and/ or ureter (;astric. pyloric. or duodenal ulcer Stomach Stomach and kidney Pancreatic cyst Perf(>rated viscus (appendix, sigmoid colon or sigmoid diverticulum) and bacteremia Intra-abdominal abscess Intra-abdominal abscess and bacteremia Appendicitis Cholecystitis Peritonitis Cholelithiasis Gangrenous appendix Ileocolonic fistula Diverticulitis Prepyloric ulcer Gangrenous colon
Cej(,tetan
Cefoxitin
(n = 95)
(n = 43)
:>5 (5R'7r)
22 (51'7r) 7 7 2
II II
5 4 3
()
0
2 15* 2 I I l (l'!i)
2 2 () ()
3 (7%)
14 (15%)*
6 (14'7r)
2 (2%) 12 (13%)
()
5 (5%)
I (2%)
4 (4%)* 1 (1%) l (1%)
3 (7%) 2 (4%) 0 3 (7'7r)
I(!%)
()
0
I (2%)
()
I (2%)
0
I (2%)
*One of these infections was known to be hospital acquired.
mellitus or malignancy. No statistically significant differences were detected between the groups for any of the variables for demographics, clinical diagnosis and infection history, or past medical history and pretreatment health status (p > 0.05). The differences for most pretreatment laboratory values between the groups were not statistically significant; the mean level of serum creatinine for the patients in the cefoxitin group ( 1.3 mg/ dl) was slightly higher than the value for those in the cefotetan group ( 1.1 mg/ dl), a difference that approached statistical significance (p = 0.07). The mean oral body temperatures taken before treatment were similar for the two groups, i.e., I 00.0° ± !. 7o F for the cefotetan group and I 00.0° ± 1.8° F for the cefoxitin group. The respective mean estimates of daily temperature change (slope) were -0.31° ± 0.29° and -0.29° ± 0.37° F. Among the evaluable patients 86 of the 95 in the cefotetan group received 2 gm every 12 hours and nine received 1 gm every 12 hours; 40 patients in the cefoxitin group were given 2 gm every 6 hours and the other three received I gm every 6 hours. The median total dose of cefotetan was 20 gm (mean = 23 gm), and that for cefoxitin was 42 gm (mean = 45 gm). The median duration of therapy was 6 days for each group. The clinical response to treatment with both drugs was excellent; 93 of the 95 evaluable patients in the cefotetan group (98%) and 41 of the 43 evaluable patients in the cefoxitin group (95%) either had resolution
of clinical signs and symptoms or had improved substantially (Table IVA). No statistically significant difference in response rates was detected between the groups (p = 0.59). Of the 138 evaluable patients 53 were not included in the bacteriologic evaluation as no pathogens could be isolated before treatment. These included 37 of the cefotetan group, in whom the signs and symptoms of infection cleared in 32 and improved in five, and 16 of the cefoxitin group, all of whom were clinically cleared. These patients were treated primarily for appendicitis, gangrenous appendix, perforated small bowel, or perforated prepyloric or duodenal ulcer. Fifty-six of the 58 bacteriologically evaluable patients in the cefotetan group (97%) had a satisfactory or presumed satisfactory bacteriologic response, as did 24 of the 27 (89%) in the cefoxitin group (Table IVB). This difference in bacteriologic response rate was not statistically significant (p = 0.32). Table V gives the bacteriologic response according to the pathogens isolated before treatment. A total of 177 bacterial isolates was cultured among the 58 bacteriologically evaluable patients in the cefotetan group, an average of 3.0 isolates per patient; 84 were anaerobic bacteria, an average of 1.4 per patient. In the cefoxitin group the 27 bacteriologically evaluable patients had a total of 101 isolates, an average of 3.7 per patient and 49 were anaerobes. an average of 1.8 per patient. E. coli and Streptococcus sp. were the most common
732
Lewis et al.
M.nch I~JKI<
Am
J
Ohstet Cnw< ol
Table IVA. Clinical responses Ccfoxitiu .JJJ
Cejotf'lrw (n = 95)
Cleared Improved Unchanged or worse
ill (85.3%)} ')'' 12 (12.6%)
• .>
Ill =
("~ ()CI) CI/ .• /(
39(90.7'.il)41 2 (4.7'/()
f
"'J'l
(9.7 .. ;(.)
2 (4.7'7r)
2 (2.1%)
Table IVB. Bacteriologic responses*
Satisfactory or presumed satisfactor~· Unsatisfactory
Cefotetrm
Cefoxitin
(11 = 58)
(II=
56 (96.6'7r)
2·4 (88.9%)
2 (3.4%)
3 ( 11.1 '7r)
27)
*Bacteriologic response not evaluated for 53 patients.
aerobic bacteria isolated before treatment; B. fragilis. other species of Bacteroides, and Clostridium sp. were the most frequent anaerobic pathogens. Bacteroides ovatus was isolated from the blood of each of the three bacteremic patients in the cefotetan group; B. fragilis, Bacteroides sp .. or gram-positive cocci were the pathogens in the three patients with bacteremia who received cefoxitin. In the cefotetan group 89 of the 93 aerobic isolates (96%) were eradicated, as were 77 of the 84 anaerobes (92% ), including all three isolates from blood. In the cefoxitin group 38 of 46 aerobes (83%) were eradicated, as were 44 of 49 anaerobic bacteria (90%), including all three pathogens isolated from the blood. Two patients in the cefotetan group showed no clinical improvement and had an unsatisfactory bacteriologic response. One was a 27-year-old Hispanic man with an appendiceal abscess from which Streptococcus sp. (group D), B. fragilis, Clostridium sp., and Eubacterium limosum were cultured from peritoneal fluid before therapy. At the end of treatment the B. fragilis and Clostridium sp. had been eradicated, but he developed a subfascial abscess from which Streptococcus sp., Eubacterium sp., and coagulase-negative Staphylococcus epiderrnidis were cultured. The other clinical and bacteriologic failure occurred in a 22-year-old white man admitted with a ruptured appendix from which Streptocornts sp. (group D), E. roli, B. ji·agilis, Bacteroides sp., and Jo:u/)(Jcteriurn sp. were isolated. He subsequently developed a supralevator abscess, but no pathogens could be isolated from it. Two patients in the cefoxitin group were clinical failures, and three were considered bacteriologic failures. A 49-year-old Hispanic man was admitted with an appendiceal abscess from which Streptococcus sp. (group D) and£. coli were cultured before treatment. He subsequently developed a subfascial abscess that grew Streptococcus sp. (group D). The other clinical failure in the cefoxitin group occurred in a 65-year-old white woman with a perforated sigmoid diverticulum
and bacteremia. Before therapy, Staphylococcus aureus, E. coli, Klebsiella oxytoca, B. fragilis, and Clostridium sp. were isolated from peritoneal fluid, and B. fragilis and Bacteroides sp. were isolated from the blood. The patient developed an abdominal abscess and underwent a laparotomy; B (ragilis, Lactobacillus sp .. and Pseudomonas sp. were isolated from a culture of peritoneal Huid at that time. The third patient was considered a bacteriologic failure but a clinical success. This 34-year-old white man was admitted with a perforatitd appendix with abscess. Pretreatment cultures of peritoneal fluid grew Strep· tococcus sp .. E. coli, Pseudomona.\ aeruginosa, Bacteroides sp., Bacteroides vulgatus, and Peptocorrus sp. The patient improved slowly, but cultures from the drain site continued to grow StTeptococcus sp. (group D and nongroup D), E. coli, Pseudomonas aeruginosa, Staphylococcus au reus, B. fragilis, and Bacteroides sp. after cefoxitin therapy was stopped. The mean concentrations of cefotetan in serum from 30 patients after drug administration on days 2. 3, and 4 ranged from 117.7 to 223.2 f,lg/ml, and the per· centage of the dose recovered as tau to mer was bet ween 8.2 and 15.4 (Table VI). In urine samples obtained from two patients 0 to 6 hours after drug administration on day 4, the mean level of cefotetan was 2673.5 f,lg/ ml, and its tautomer accounted for II. I 7c of the dose. The only statistically significant differences in mean slopes for the clinical laboratory determinations performed before and after treatment were those for hematocrit and serum creatinine. The fall in hematocrit for cefotetan (- 1. 122) was greater than that for cefoxitin (- 0.444) (p = 0.04), and the decrease in serum creatinine for cefoxitin (- 0.038) was almost twice that for cefotetan (- 0.020) (p = 0.02). Neither difference was considered clinically significant. All 121 patients in the cefotetan group and 66 in the cefoxitin group treated during the swdy were included in the evaluation of safe tv; no safety information was
Cefotetan vs cefoxitin in intra-abdominal infections
Volume 158 Number 3, Part 2
733
Table V. Bacteriologic response of pathogens isolated before treatment Crj(;/etan
Cefoxitin
isolated
Satisfactory or presumed 1atis{artm-y
4
4
I H 26
I ~4
4 12
2
2
1
31 5 0 2 I I
30
20
5
4
I l
.~.Vo.
1.Vo. isolated
Satisjrtctory or presumed satisfactory
Aerobic bacteria Gram-positive cocci Staphylococcus aureus (coagulase positive) Streptococcus pneumoniae Streptococcus sp. group D Streptococcus sp.
Unspecified gram-positive cocci Gram-negative bacilli Escherichia coli Klebsiella pneumoniae Klebsiella ox~toca Enterobacter cloacae Enterobacter gergoviae Enterobacter sp. Serratia marcescens Proteus mirabilis Pseudomonas aeruginosa Pseudomonas vesicularis Eikenella corrodens Haemophilus irifluenzae
Unspecified gram-negative rods Gram-positive bacilli Diphtheroids Anaerobic bacteria Gram-positive cocci Peptococcus sp. PeptostreptococcUI sp. Gaffkya anaerobius
Unspecified anaerobic gram-positive cocci Gram-negative cocci
4 I
0 2
7
I 0
()
()
:I I! I 17 4
()
2
!
2
0
()
I I
0 1
()
I I ·l:
0 0
0
3
2
I 0 2
()
()
1
! ()
I
I
()
()
()
()
2
2
()
()
I
l 4 I I
2 3 0
~~ ()
1
I
()
()
4 I I
V eillonella parvula
!
Gram-positive bacilli Actinobacillus israeli Bifidobacterium adolescentis Bifidobacterium sp. Clostridium peifringens Clostridium symbiosum Clostridium tertium Clostridium sp. Eubacterium sp.
Anaerobic gram-positive non-spore-forming rods Gram-negative bacilli B. fragilis Bacteroides ovatus Bacteroides vulgatus Bactermdes distasonis Bacteroides thetaiotaomicron Bacteroides melaninogenicus Bacteroides sp. Fusobacterium sp. Fusobacterium necrophorum Fusobacterium variurn
Anaerobic gram-negative rods Other pathogens CDC GR 3452
2
2
()
()
I
()
()
I 4
! !
0
()
4
3
:>
0
()
()
0 .S
1 1 !2
I I
6
7
9 0
()
2 !
!0 !0
8 10
H 4
5
5
2
4
4
3 1
;)
1 I 1
14
4 0 1 I 0
I 13 4 ()
I
I
II 2
7 4
I I I I
8
7
()
()
1 0 0
()
! 0
()
*Some patients had more than one pathogen isolated.
available for one patient in the cefoxitin group. Adverse reactions occurred in 33 patients (27%) in the cefotetan group and in II patients (I7%) in the cefoxitin group. The difference in these incidences was not statistically
significant (p = O.IO) (Table VII). These reactions were considered related to the drug only in five patients in the cefotetan group (4%) and in four patients in the cefoxitin group (6%). This difference was not statisti-
73L-
r_ewis et al.
\larch 1~lkS .\111 .J Ol»il'l ( ;\'lll'cul
Table VI. Plasma and urine levels of cefotetan ,\'o. fif patient.\ Plasma Dav ~ Da} :> Day 4 Urine
5 13 9
Day~
Day 4
~
JI,J ean level (J.Lglm/)
Table VII. Adverse reactions (:tji>ll'lllll
and range
Tautomer (%! and range
223.3 (167-298) 179.9 (11-255) 117.7 (52-231)
8.2 (1.6-25.2) 15.4 (0-29.0) 15.0 (3.7-22.1)
974 2674 (657-4690)
11.1 (3.7-18.7)
0
cally significant (p = 0.72). One patient receJVmg cefotetan developed hives on the face and neck after receiving his second dose of the drug and experienced pruritus of the face and neck, without rash, after the third dose. No symptoms developed after the fourth dose of cefotetan, but therapy was discontinued. Three patients in the cefotetan group died during the studv, but no death was considered related to treatment. One death occurred in a 71-year-old man with a history of angina and pulmonary fibrosis of unknown cause who was treated for cholecystitis; he died 1 month after completion of therapy from end-stage pulmonary fibrosis and cardiac arrest. The second death occurred in a 49-year-old man with a history of alcoholic cirrhosis and end-stage liver disease who received two doses of cefotetan because of spontaneous bacterial peritonitis before he died of hepatorenal failure, septic shock, and aspiration. The third patient was a 94-year-old woman with a history of coronary artery disease who was treated for a perforated duodenal ulcer and peritonitis and died on the fifth day of cefotetan therapy of a suspected myocardial infarction or pulmonary embolus.
Comment In this population of predominantly young men with community-acquired intra-abdominal infections after gastrointestinal tract perforation, prompt treatment with the less complicated dosage regimen of cefotetan of 2 gm every 12 hours was as effective as cefoxitin administered at 2 gm every 6 hours. With sample sizes of 95 evaluable patients in the cefotetan group and 43 in the cefoxitin group and an assumed rate of success for cefoxitin of 95%, the probability is 0.37 that a 10% lower success rate for cefotetan would have been detected and 0.65 that a 15% lower rate would have been detected. Of the total of 138 patients evaluated, only two patients in each group failed to respond. These treatment success rates are higher than those reported in previous studies of cefotetan in patients with established intraabdominal infections but without adjunct surgeri 11 "
Patienh with no adverse reactions Patients with at least one adverse reaction Patients thought to have a drug-related reaction No. reported (treatment related) Symptoms Phlebitis at the injection site Edema Itching
Rash l'rticaria I lematuria I'Hlria Diagnoses Death Laboratory results Increased eosinophils lncreased partial t h rmnboplastin time Increased prot It rom bin time Thrombocvthemia or thro~bocytosis Increased white blood cell count I ncrcased liver function tests Urinarv casts
R:-l/1~1 (7~.7~;;.)
')!)/(if)
:r11Jn
I lihti i 1!1. ,f.,,:)
')I[~]
(~7.:1~0
H.IVr l
(:•(!.:~(_;;)
4/()1) jl), [
~(~I
0
l (II :l (1) I I (II I I
0 II
<;
i
I)
II I) I)
:1
II
:l (II l
:l (I i 0
R
:l
II (]I
:1 (I)
2:1 !:"\)
li(:l)
0
*:-.lo intinmation available for one patient in the ce!cJxitin group.
and similar to the results of other studies in patients with peritonitis.'' Therapeutic response rates in studies of third-generation cephalosporins in patients with polymicrobial peritonitis and peritoneal abscess were lower.'· The bacteriologic cause of these infections was similar to that in other reports of intra-abdominal infections.~. "· '' Most infections were polymicrobial; approximately half the isolates were anaerobes, and E. coli was the most common aerobic pathogen, and B. fragilis and other Bacternides sp. were the most frequently isolated anaerobic pathogens. Cefotetan was as effective as cefoxitin in achieving a satisfactory bacteriologic response in patients with infections caused by these bacteria. All 4 7 isolates of B. jr·agilis and its subtypes were eradicated after cefotetan therapy, as were all but one of the 50 gram-negative bacilli. These clinical results support the good in vitro activitY reported for cefotetan against such organisms.,~. ,-, Cefotetan was also well tolerated in this study. The incidence of adverse reactions related to the drug was slightly lower than that of reactions related to treatment with cefoxitin, although this difference was
Volume 158 Number :1. Part 2
not statistically significant, With its less complicated dosing schedule, equivalent effectiveness, and marginally lower toxicity, cefotetan may represent an advantageous alternative to cefoxitin in the treatment of promptly diagnosed polymicrobial, communityacquired intra-abdominal infections. REFERENCES L Dunn DL, Simmons RL The role of anaerobic bacteria in intraabdominal infections. Rev Infect Dis 1984;6(suppl 1):5139-46. 2. Lorber B, Swenson RM. The bacteriology of intraabdominal infections. Surg Clin North Am 1975;55: 1349-4. 3. Ho JL. Barza M. Role of aminoglycoside antibiotics in the treatment of intra-abdominal infection. Amimicroh Agents Chemother 1987;31:485-91. 4. Onderdonk AB, Shapiro ME, Finberg RW, Zaleznick DF, Kasper DL. Use of a model of intraabdominal sepsis for studies of the pathogenicity of Bacteroides fragilis. Rev Infect Dis 1984;6(suppl I):S91-5. 5. Weinstein WM, Onderdonk AB, BartlettJG, Gorbach SL. Experimental intra-abdominal abscesses in rats: development of an experimental model. Infect Immun 1974; 10:1250-9. 6. Tally FP. Gorbach SL Therapy of mixed anaerobicaerobic infections. Lessons from studies of intraabdominal sepsis. Am J Med 1985;78(suppl 6A): 145-53. 7. Drusano GL. Warren JW, Saah AJ, et a!. A prospective randomized controlled trial of cefoxitin versus clindamycin-aminoglycoside in mixed anaerobic-aerobic infections. Surg Gynecol Obstet 1982;154:715-20.
Cefotetan vs cefoxitin in intra-abdominal infections
735
8. Tally FP, Kellum JM, Ho JL. O'Donnell TF, Barza M, Gorbach SL. Randomized prospective stud} comparing moxalactam and cefoxitin with or without tobramycin for the treatment of serious surgical infections. Antimicrob Agents Chemother !986:29:244-9. 9. Schentag JJ, Wels PB, Reitberg DP, Walczak P. Van Tyle JH, Lascola RJ. A randomized clinical trial of moxalactam alone versus tobramycin plus clindamvcin in abdominal sepsis. Ann Surg 1985:198:35-41. lOa. Bauer WW. Kirby WMM. Sherris .JC, Turck :VI. Antibiotic susceptibility testing by a standardized single disc method. Am .J Clin Pathol 1966:45:493-6. lOb. Standardized disc susceptibility test. Fed Reg- 1972;37: 20527-9. JOe. National Committee for Clinical Laboratorv Standards. Approved standard: ASM-2, Performance standards for antimicrobial disc susceptibility tests, .July 1975. II. Fujimoto M, Ueda T, Sakai K. et al. A clinical trial on cefotetan (YM 09330) in the field of surger~. Chemotherapy 1982:30:817-26. 12. YuraJ, Shinagawa N. Ishikawa S. et al. Fundamental and clinical studies of ceforetan (YM 09330) on the surgical field. Chemotherapy 1982:30:786-795. 13. Wilson SE, Lewis RT, Ton GT. Comparative clinical evaluation of cefotetan with moxalactam in the treatment of hospitalised patients with serious intra-abdominal infections. [Abstracts]. In: Proceedings of the fourteenth international congress of chemotherapy. Kvoto. Tokvo: University of Tokyo Press, 1985:86. 14. Ruckdeshel G. Activity in 1•itru of cefotetan against nonsporing anaerobes: a comparative study . .J Antimicrob Chemother 1983; ll(suppl A): II 7-24. 15. Watt B, Brown FV. The comparative in-1•itro activity of cefotetan against anaerobic bacteria . .J Antimicrob Chemother 1985;15:671-7.