Ertapenem Or Ticarcillin/Clavulanate for the Treatment of Intra-Abdominal Infections Or Acute Pelvic Infections in Pediatric Patients

Home , Ticarcillin

The American Journal of Surgery 194 (2007) 367–374 Clinical surgery–American Ertapenem or ticarcillin/clavulanate for the treatment of intra-abdominal infections or acute pelvic infections in pediatric patients

Albert E. Yellin, M.D.a,*, Jeffrey Johnson, M.D.b, Iliana Higareda, M.D.c, Blaise L. Congeni, M.D.d, Antonio C. Arrieta, M.D.e, Doreen Fernsler, B.S.f, Joseph West, Ph.D.f, Richard Gesser, M.D.f

aDepartment of Surgery, Keck School of Medicine of the University of Southern California, 1200 N State St, Rm 9610, Los Angeles, CA 90033, USA bDepartment of Pediatrics, Keck School of Medicine of the University of Southern California, 1240 N Mission Rd, L902, Los Angeles, CA 90089-9300, USA cHospital Civil de Guadalajara, Hospital 278, 4420 El Retiro St, Guadalajara, Jalisco Mexico dAkron Children’s Hospital, Division of Infectious Diseases, Locust Professional Bldg, 300 Locust St, Ste 4, Akron, OH 44308-1062, USA eChildren’s Hospital of Orange County, Pediatric Infectious Diseases, 455 South Main Street, Orange, CA 92868, USA fMerck Research Laboratories, Merck & Co., Inc., 126 East Lincoln Avenue, Rahway, NJ 07065, USA

Manuscript received June 20, 2006; revised manuscript January 20, 2007

Abstract Background: Ertapenem, a group I carbapenem antibiotic, has been shown to be safe and effective in treating adults with complicated intra-abdominal (cIAI) or acute pelvic infection (API). This study evaluated ertapenem for treating these infections in children. Methods: In an open-label study, children aged 2 to 17 years with cIAI or API were randomized 3:1 to receive ertapenem or ticarcillin/clavulanate. Children 13 to 17 years of age received 1 g parenterally daily, and those 2 to 12 years of age received 15 mg/kg twice daily. Patients Ͻ60 kg received ticarcillin/ clavulanate 50 mg/kg 4 to 6 times daily and 3.1g4to6times daily for those Ն60 kg. Patients were assessed for safety and tolerability throughout the study and for efficacy after the completion of therapy. Results: One hundred five patients, 72 (69%) with cIAI, received Ն1 dose of study drug and were included in the safety analysis. Eighty-one patients were treated with ertapenem. Infusion site pain was the most common drug-related adverse event in both groups. In the modified intent-to-treat analysis, the age-adjusted posttreatment clinical response rates were 87% (43/50 patients) and 100% (25/25 patients) in the cIAI and API patients, respectively, for ertapenem and 73% (11/15 evaluable patients) and 100% (8/8 evaluable patients), respectively, for ticarcillin/clavulanate. Overall age-adjusted response rates were 91% (68/75 evaluable patients) for ertapenem and 83% (19/23 evaluable patients) for the comparator. Conclusions: This study suggests that ertapenem is generally safe and efficacious for treating cIAI or API in pediatric patients. © 2007 Published by Excerpta Medica Inc.

Keywords: Ertapenem; Pediatric; Intra-abdominal; Acute pelvic infection

Complicated intra-abdominal infections (cIAIs), caused pri- adequately powered trials have shown the safety and efﬁ- marily by ruptured or gangrenous appendicitis with or with- cacy of ertapenem for treatment of cIAI and API in adults out peritonitis, are important causes of morbidity in children [4–6]. Because the pathophysiologic processes and patho- [1,2]. Acute pelvic infections (APIs), similar to cIAIs, are gens involved in cIAI and API are similar in adults and often mixed infections caused by enteric pathogens. These children, children should respond similarly to treatment include postpartum or obstetrical surgery infectious compli- with ertapenem when infections are caused by susceptible cations in adolescent females [3]. Previous well-controlled, bacteria. Ertapenem, a group 1 carbapenem, is active in vitro against most pathogens likely to be encountered in the treatment of cIAI (ie, infections requiring both operative * Corresponding author. Tel.: ϩ1-323-226-7737; fax: ϩ1-323-226- drainage and antimicrobial therapy) including Escherichia 5815. coli and other Enterobacteriaceae, Clostridium clostridio- E-mail address: [email protected] forme, Eubacterium lentum, Peptostreptococcus species,

0002-9610/07/$ – see front matter © 2007 Published by Excerpta Medica Inc. doi:10.1016/j.amjsurg.2007.01.023 368 A.E. Yellin et al. / The American Journal of Surgery 194 (2007) 367–374 and Bacteroides spp including Bacteroides fragilis. In acute or cramping or tenderness or sonographic/radiographic ev- pelvic infections, such as postpartum endomyometritis, sep- idence suggesting a pelvic abscess; and (3) either fever (oral tic abortion, and postsurgical gynecologic infections, ertap- temperature Ͼ38°C), WBC Ͼ10,500/mm3, or a WBC dif- enem is effective against pathogens such as Streptococcus ferential with Ͼ10% band forms. agalactiae, E coli, B fragilis, Porphyromonas asaccharo- Enrollment was limited to immunologically competent lytica, Peptostreptococcus species, and Prevotella bivia. patients. Because specific ertapenem dosing guidelines for Unlike the group 2 carbapenems, imipenem and mero- pediatric patients with renal insufficiency are not available, penem, ertapenem has substantially less activity against children with significant renal impairment (serum creatinine nonfermentative gram-negative bacteria and thus is not in- Ͼ1.25 times the upper limits of normal) were excluded from dicated in the treatment of infections caused by Pseudomo- the study. nas aeruginosa or Acinetobacter species. Like other cur- rently approved beta-lactam agents, ertapenem is not active Interventions against methicillin/oxacillin-resistant staphylococci [7–9]. For children 3 months to 12 years of age, ertapenem was The aims of this study were to evaluate the safety, tol- administered twice daily as a 15-mg/kg parenteral dose to a erability, and efficacy of ertapenem versus a comparator maximum total daily dose of 1 g. Children 13 years of age ticarcillin/clavulanate control group in pediatric patients (3 or older received1gofertapenem as a single daily dose. months–17 years of age) with cIAI or API caused by sus- Treatment was always initiated with a 30-minute intrave- ceptible pathogens. nous (IV) infusion, but intramuscular dosing could be used if the patient met protocol-defined criteria (hemodynami- Methods cally stable and no coagulopathy). Dosing was based on preliminary results from a single-dose pharmacokinetic Study design study in pediatric patients and was chosen to approximate This was a prospective, multicenter, randomized, open- the exposure and pharmacokinetic /pharmacodynamic effi- label, comparative-controlled study conducted from March cacy determinants of 1 g daily dosing in adults [10]. 2002 to January 2004 at 15 sites in the United States, The comparator antibiotic regimen, IV ticarcillin/clavu- Mexico, and Brazil (12 sites in the United States). The lanate (50 mg/kg, based on the ticarcillin component), was protocol was approved by each center’s institutional review administered 4 or 6 times a day (for patients Ͻ60 kg) or committee. Written informed consent was obtained for each 3.1 g (3 g of ticarcillin and 100 mg of clavulanic acid) 4 or patient from his/her parent or guardian according to the 6 times a day (for patients Ն60 kg). Each infusion was guidelines of each institution. administered over a 30-minute period. These doses are ap- Children between 3 months and 17 years of age were proved for pediatric use in children 3 months of age and eligible to participate if they were diagnosed with cIAI with older for the indications of intra-abdominal infection and rupture and extension into the peritoneal cavity or API (in pelvic infections, including endometritis. the absence of pelvic inflammatory disease) that was likely Study therapy could be completed in the hospital, an to be treatable with 3 to 14 days of parenteral therapy and outpatient clinic, or with home infusion after at least 2 days was not complicated by preexisting conditions that could of in-hospital or clinic-based infusion therapy. Treatment confounded the evaluation of the study drugs. Patients with was to be continued for a minimum of 5 days for cIAI and intra-abdominal infection were either enrolled postopera- 3 days for API. The maximum suggested duration of ther- tively after visual confirmation of established intra-abdom- apy was 14 days. All therapy was to be administered par- inal infection including appendicitis complicated by perfo- enterally; the protocol did not allow for a switch to oral ration and/or abscess or were enrolled preoperatively based follow-up treatment. In addition to the antibiotic therapy, on the need for surgical intervention and the presence of medical and surgical treatment of the infectious process was each of the following criteria: (1) evidence of systemic performed according to guidelines established by the par- inflammatory response (at least 1 of the following: fever ticipating institution and was consistent with the investiga- [oral temperature Ͼ38.5°C], white blood cell [WBC] count tor’s usual clinical practice. Culture and sensitivity assays of Ͼ15,000/mm3, drop in blood pressure, increased pulse and appropriate bacterial specimens was requested at baseline, respiratory rate, hypoxemia, or altered mental status), (2) at any time that there was clinical or laboratory evidence of localized physical findings including pain/tenderness and persistence or progression of the infectious process (includ- rigidity consistent with intra-abdominal infection, and (3) ing persistent fever, elevated white blood cell count, or supportive radiologic findings in the abdomen such as de- significant changes in the patient’s clinical condition), and tection of an intraperitoneal abscess detected on computed at the time of any surgical or drainage procedure. tomography scan or ultrasound. Patients found to have sim- ple appendicitis or gangrenous appendicitis without rupture into the peritoneal cavity were excluded from enrollment. Assessment Patients with acute pelvic infection were required to have Safety and tolerability of the parenteral therapy were each of the following for a clinical diagnosis: (1) either evaluated daily while patients were receiving study therapy vaginal delivery, cesarean section, or gynecologic surgery and for 14 days after therapy. The primary objective of the Ͼ24 hours and within 1 month before enrollment (or within study was to assess the incidence of any serious drug-related 24 hours if oral temperature at least 38.6°C) or a diagnosis clinical and/or laboratory adverse experiences during the of septic abortion requiring at least 3 days of parenteral study drug therapy period and 14-day posttherapy follow- antibacterial therapy; (2) pelvic, abdominal, or uterine pain up. Secondary safety objectives for the study included eval- A.E. Yellin et al. / The American Journal of Surgery 194 (2007) 367–374 369 uating the incidence of any drug-related clinical and/or data as possible on ertapenem and still maintain a reason- laboratory adverse experiences, incidence of moderate-to- ably sized sample in the comparator treatment arm. Enroll- severe administration site reactions. ment was stratified based on indication (cIAI or API) and, For the evaluation of efficacy, patients were assessed by within the IAI indication, by age category (3–23 months, the investigator for clinical signs and symptoms of infection 2–12 years, or 13–17 years) to ensure comparability and at baseline, at the discontinuation of parenteral study ther- unbiased enrollment between treatments. Within each stra- apy, and at the 2- to 4-week posttreatment follow-up visit tum, patients were randomized in a 3:1 ratio according to for API and 3 to 5 weeks for cIAI patients. Bacterial cul- computer generated allocation schedules provided by the tures were to be obtained from appropriate infected material sponsor to a centralized randomization service. Screening at baseline and subsequently if clinically indicated. Overall, and consent activities were completed before randomiza- clinical and microbiologic outcome evaluations were made tion. No study site personnel had access to the centrally held at the discontinuation of parenteral study therapy and at the randomization schedules. posttreatment follow-up assessment. Clinical outcome included an assessment of “cure,” “failure,” or “indetermi- Statistical analyses nate.” At all investigative sites, patients were considered Both the clinical modified intent-to-treat (cMITT) and cured if there was complete resolution or substantial im- evaluable per protocol (EPP) populations were determined provement in the baseline signs and symptoms (including before analyses using prespecified criteria. The overall clin- pelvic, abdominal, or uterine pain/tenderness, nausea/vom- ical response was assessed by the investigator at the post- iting, fever, and elevated WBC) of the index infection with treatment follow-up visit. To include as many randomized no previous assessment of failure and no further antimicro- patients as possible in these analyses, the principal evalua- bial therapy indicated. A microbiologic response was as- tion of efficacy was based on the cMITT population, which sessed for each baseline pathogen. For patients in whom included all treated patients meeting baseline clinical diag- follow-up cultures were not performed, microbiologic re- nosis criteria. A microbiologic MITT population was also sponses for baseline pathogens were presumed based on defined and consisted of all cMITT patients with a baseline the clinical response. The overall microbiologic response pathogen. The MITT efficacy analyses took into consider- was considered favorable if all baseline pathogens were ation all posttreatment assessments and was based on as- eradicated or presumed eradicated. Emergent pathogens sessments within the protocol defined test-of-cure window were considered “superinfections” if first isolated during if available, (2 to 4 weeks after treatment for API patients therapy and “new infections” if isolated after completing and 3- to 5 weeks after treatment for cIAI patients) or were therapy. otherwise based on the last assessment available after the completion of study therapy. Patients missing a posttreat- Sample size and power calculations ment assessment were excluded from these analyses unless Based on estimates obtained from 2 adult phase IIb/III they had previously failed, in which case their outcome was clinical trials of cIAI [4] and API [6], the pooled average unfavorable. Analyses were also done on the EPP popula- rate of ertapenem patients having any clinical and/or labo- tion, which excluded patients with potential confounders ratory drug-related serious adverse experiences was 2% likely to have affected outcome in accordance with appli- with a range per study of 1% to 2%. It was anticipated that cable regulatory guidelines and accepted industry standards the experience in pediatric patients would parallel that of [11,12]. To be included in the EPP analyses, efficacy was adults. Assuming a total of 75 patients in the ertapenem required to have been assessed by the investigator within the group in this trial, if the observed incidence rate of any protocol defined test-of-cure window. Patients with 1 or clinical and laboratory drug-related serious adverse experi- more baseline pathogens were included in the EPP analyses ence during study therapy and 14-day follow-up period was if at least 1 baseline pathogen was susceptible to both 3%, approximately equal to the highest observed rate in the parenteral study therapies. adult studies, the 95% confidence interval (CI) about an The primary efficacy response for both cIAI and API incidence rate of 3% would be 0.4% to 10%. If a specific patients was based on the clinical efficacy assessment. A clinical or laboratory drug-related serious adverse experi- favorable clinical response (clinical cure) at the posttreat- ence was not observed in the 75 patients of the ertapenem ment assessment visit was prospectively defined as a com- treatment group, the true incidence rate of that clinical or plete resolution or substantial improvement in the signs and laboratory drug-related serious adverse experience would be symptoms of the index refraction with no previous assess- Ͻ5% with 95% confidence. ment of failure and no further antimicrobial therapy indicated. The efficacy summary statistics were computed adjusting Blinding and randomization for age within each stratum and adjusting for age “overall” The study drugs were administered in an unblinded man- (where strata were combined). ner. Because ertapenem and ticarcillin/clavulanate have dif- ferent dosing frequencies, the design of the study was by necessity open label so that matching placebo infusions Results would not have to be administered to acutely ill pediatric Patient accounting patients. One hundred twelve patients were randomized (Fig. 1). Patients were randomized 3:1 to receive ertapenem or One hundred five patients (cMITT population) received Ն1 ticarcillin/clavulanate to obtain as much safety and efficacy dose of study therapy and were included in the safety 370 A.E. Yellin et al. / The American Journal of Surgery 194 (2007) 367–374

Fig. 1. Profile of study enrollment and summary of patients in each evaluable population. evaluation. Eighty-one and 24 patients were treated with tapenem group were bacteremic at baseline, 1 each with B ertapenem and ticarcillin/clavulanate, respectively. Overall, fragilis and Streptococcus constellatus. In the ticarcillin/ 94% of the patients randomized were included in the cMITT clavulanate comparator group, 1 patient was bacteremic population, and 72% (n ϭ 81) of all randomized patients with a Bacillus species at baseline. were included in the EPP population. In the API stratum, approximately half of the patients in each treatment arm had a diagnosis of endomyometritis associated with an obstetrical procedure; the remainder had Baseline characteristics a diagnosis of septic abortion. The most commonly isolated Baseline characteristics for all treated patients (cMITT pathogen in both treatment groups was E coli. There were population) are reported in Table 1. The study included 47 no bacteremic API patients in either treatment group at patients 2 to 12 years of age and 58 patients 13 to 17 years baseline. of age. All 33 patients in the API stratum were females between 13 and 17 years of age. Although patients between 3 months and 17 years of age were eligible to participate, Duration of treatment only patients 2 years of age and older were actually en- The duration of treatment for each treatment group by rolled. This age distribution accurately reflects the age disease indication is provided in Table 2. No patient in ranges for the infectious diseases under study in children. either treatment group missed a day of study therapy. Both One cIAI patient had a perforated Meckel’s diverticulum treatment groups appear similar with respect to extent of with localized peritonitis. All other cIAI patients had a exposure. Although the protocol allowed investigators the ruptured appendix at presentation; more than 30% of these option to administer ertapenem by the intramuscular (IM) or patients in each treatment group had Ն1 intra-abdominal IV route, only 1 patient in the study received IM therapy abscess, and 62% in each treatment group had evidence of (for 2 days). peritonitis. Seventy-five percent or more of the cIAI patients in each treatment group were initially managed by an open Incidence of adverse events surgical procedure; approximately 20% in each group had As shown in Table 3, 48 patients (46%) had 1 or more laparoscopic surgery, and the remaining patients were man- clinical adverse experiences as assessed by the investigator: aged by a percutaneous procedure only. The most common 32 (39%) in the ertapenem group and 16 (67%) in the pathogens isolated during surgery in both treatment groups comparator group. were E coli and Bacteroides species, including B fragilis One patient treated with ertapenem experienced a serious and Bacteroides thetaiotaomicron. Two patients in the er- drug-related clinical adverse experience. This was a 3-year A.E. Yellin et al. / The American Journal of Surgery 194 (2007) 367–374 371

Table 1 adverse event. This was the only patient to discontinue Demographic and baseline characteristics: clinical MITT population ertapenem therapy because of a drug-related adverse expe- (primary efﬁcacy population) rience. Diarrhea was reported as a drug-related adverse Ticarcillin/ event with a similar frequency in the 2 treatment groups and Ertapenem clavulanate with a similar frequency to that reported in previous adult (n ϭ 81) (n ϭ 24) studies [4,5]. n (%) n (%) In total, 4 patients discontinued therapy because of adverse clinical experiences, 2 in each group. The other pa- Gender tient in the ertapenem group had an abdominal abscess and Male 30 (37) 10 (42) discontinued therapy on day 10. In the comparator group, 1 Female 51 (63) 14 (58) patient developed a rash and discontinued therapy on day Age 2 to 12 years 37 (46) 10 (42) 28, and the other, with a wound infection, discontinued 13 to 17 years 44 (54) 14 (58) therapy on day 9. Stratum by diagnosis and age Eleven patients (14%; 95% CI, 7.0–23.0) in the ertap- Acute pelvic infection 25 (31) 8 (33) enem group and 8 patients (33%; 95% CI, 15.6–55.3) in the 13 to 17 years 25 (31) 8 (33) comparator group reported drug-related clinical and/or lab- Complicated intra-abdominal infection 56 (69) 16 (67) oratory adverse experiences. Infusion-site pain was the most 2 to 12 years 37 (46) 10 (42) common drug-related adverse experience reported in both 13 to 17 years 19 (23) 6 (25) treatment groups: 6 (7%) patients in the ertapenem treat- Intra-abdominal infection stratum ment group and 3 (12%) patients in the comparator group. Overall, 103 (98%) treated patients had Ն1 laboratory (n ϭ 56) (n ϭ 16) test postbaseline during study therapy and the 14-day fol- Anatomic site of origin of current infection low-up period. Of these, 10 (10%) had laboratory adverse Small intestine 1 (2) 0 (0) experiences. Five patients (3 in the ertapenem group and 2 Vermiform appendix 55 (98) 16 (100) in the comparator group) had drug-related laboratory ad- Infectious process* verse experiences. However, none of the laboratory adverse Abscess 17 (30) 6 (37) experiences was serious or resulted in study discontinua- Gangrene 5 (9) 3 (19) tion. All of the drug-related laboratory adverse experiences Pelvic abscess 1 (2) 0 (0) reported occurred during parenteral therapy. Peritonitis 35 (62) 10 (62) Drug-related clinical and laboratory adverse events that Appendicitis perforated 7 (12) 2 (12) Ն Gastrointestinal perforation 4 (7) 0 (0) occurred with an incidence of 2% in either treatment Procedure† group are shown in Table 4. Laparoscopic 11 (20) 3 (19) Open 43 (77) 12 (75) Tolerability Percutaneous 2 (4) 1 (6) Assessment of tolerability (cMITT population) at the IV Irrigation Entire abdomen 34 (61) 14 (87) and IM study drug infusion-site was performed daily while Local 16 (29) 2 (12) the patient was on study therapy. Of the 105 treated patients, None 6 (11) 0 (0) 9 (11%; 95% CI, 5.2–20.0) in the ertapenem group and 6 Wound (25%; 95% CI, 9.8–46.7) in the comparator group experi- Delayed primary closure 8 (14) 1 (6) enced ϭ 1 local reactions of any intensity at the infusion/ Primary closure 48 (86) 15 (94) injection site (data not shown). Two patients (2%; 95% CI, Acute pelvic infection stratum –8.6) in the ertapenem group and 2 (8%; 95% CI, 1.0–27.0) in the comparator group experienced Ն1 local reactions of (n ϭ 25) (n ϭ 8) Infection details Table 2 Postoperative obstetric infection 8 (32) 4 (50) Treatment characteristics: clinical MITT population Spontaneous obstetric infection 17 (68) 4 (50) Infectious process Duration of parenteral study therapy (d) Endomyometritis 12 (48) 4 (50) Septic abortion 13 (52) 4 (50) Ertapenem Ticarcillin/clavulanate

* Patients could have more than 1 infectious process recorded. IAI † Patients could have more than 1 procedure recorded. N5616 Mean 7.1 8.5 Median 6.0 7.0 old Asian male with a cIAI and moderate diarrhea before Range 1.0–17.0 6.0–28.0 enrollment who developed progressively more frequent di- API arrhea over a 4-day course of ertapenem therapy resulting in N258 discontinuation. Clostridium difﬁcile testing was negative in Mean 3.3 3.9 this patient. On study day 5, the diarrhea resolved, and the Median 3.0 4.0 patient was discharged from the hospital. Because the diar- Range 3.0–5.0 3.0–4.0 rhea prolonged hospitalization, it was considered a serious N ϭ number of patients in each treatment group. 372 A.E. Yellin et al. / The American Journal of Surgery 194 (2007) 367–374

Table 3 Summary of safety during parenteral therapy and the 14-day follow-up period Clinical adverse experiences Ertapenem Ticarcillin/Clavulanate Number of treated patients (n ϭ 81) (n ϭ 24) Number (%) of patients: n (%) n (%) With 1 or more adverse experiences 32 (39) 16 (66.7) With no adverse experience 49 (60) 8 (33.3) With drug-related adverse experiences* 10 (12) 6 (25.0) With serious adverse experiences 11 (14) 3 (12.5) With serious drug-related adverse experiences 1 (1) 0 (0.0) Who died 0 (0) 0 (0.0) Discontinued because of adverse experiences 2 (2) 2 (8.3) Discontinued because of drug-related adverse experiences 1 (1) 1 (4.2) Discontinued because of serious adverse experiences 2 (2) 0 (0.0) Discontinued because of serious drug-related adverse experiences 1 (1) 0 (0.0)

Laboratory adverse experiences Ertapenem Ticarcillin/clavulanate Number of patients with at least 1 laboratory test postbaseline (n ϭ 79) (n ϭ 24) Number (%) of patients: n (%)† n (%)† With 1 or more adverse experiences 6 (8) 4 (17) With no adverse experience 73 (92) 20 (83) With drug-related adverse experiences* 3 (4) 2 (8) With serious adverse experiences 0 (0) 0 (0) With serious drug-related adverse experiences 0 (0) 0 (0) Who died 0 (0) 0 (0) Discontinued because of adverse experiences 0 (0) 0 (0) Discontinued because of drug-related adverse experiences 0 (0) 0 (0) Discontinued because of serious adverse experiences 0 (0) 0 (0) Discontinued because of serious drug-related adverse experiences 0 (0) 0 (0)

* Determined by the investigator to be possibly, probably, or deﬁnitely drug related. † The percent ϭ number of patients within the laboratory adverse experience category/number of patients with 1 or more laboratory tests post baseline.

moderate to severe intensity at the infusion/injection site. Table 4 All of the tolerability symptoms reported were from patients Most commonly reported drug-related* adverse experiences during receiving IV therapy. One patient received ertapenem IM parenteral therapy (incidence Ն2% in either treatment group) therapy for 2 days with no injection site reactions reported. Adverse experience Ticarcillin/ Ertapenem clavulanate (n ϭ 81) (n ϭ 24) Efficacy n (%) n (%) The proportion of patients in the cMITT and EPP populations and within each study drug group with a favorable Clinical clinical response assessment (cure but not failure or inde- Diarrhea 3 (4) 1 (4) terminate) at the post treatment visit are shown in Table 5. Infusion-site burning 1 (1) 1 (4) Clinical efficacy response rates in the EPP population were Infusion-site pain 6 (7) 3 (12) consistent with the cMITT results; overall response rates Infusion-site reaction 0 (0) 1 (4) Infusion-site warmth 1 (1) 1 (4) were 89% (59/66 evaluable patients) for ertapenem and Vaginal candidiasis 0 (0) 1 (4) 73% (11/15 evaluable patients) for the comparator. Compa- Insomnia 0 (0) 1 (4) rable rates were seen across each of the age groups studied. Rash 0 (0) 1 (4) Although both drugs had 100% efficacy in the API stratum in both cMITT and EPP populations, the lack of difference n/m (%) n/m (%) in clinical efficacy between treatments may be because of Laboratory the relatively small number of patients in the cMITT and ALT increased 2/74 (3) 1/23 (4) EPP populations of this stratum. Clinical efficacy in the AST increased 2/76 (3) 1/23 (4) cMITT population of the cIAI stratum was 87% (95% CI, Platelet count increased 1/76 (1) 1/23 (4) 77%–96%) for ertapenem versus 73% (95% CI, 45%–92%) n ϭ number of patients with adverse experience; m ϭ number of for the comparator. The overall age adjusted response rate in patients with laboratory test postbaseline. these patients was 68/75 patients (91%; 95% CI, 85%–98%) ALT ϭ alanine ainotransferase; AST ϭ aspartate aminotransferase. for ertapenem and 19 of 23 patients (83%; 95% CI, 1%– * Determined by the investigator to be possibly, probably, or definitely 95%) for the comparator. In both treatment groups, some- drug related. A.E. Yellin et al. / The American Journal of Surgery 194 (2007) 367–374 373

Table 5 Clinical efﬁcacy response (cures): clinical EPP and cMITT Disease stratum Ertapenem Ticarcillin/clavulanate EPP (n ϭ 66) cMITT (n ϭ 81) Adjusted EPP (n ϭ 15) cMITT (n ϭ 24) Adjusted Observed response* response* (95% CI†) Observed response* response* (95% CI†) n/m % n/m % n/m % n/m % cIAI 36/43 84 43/50 88 (77–96) 7/11 64 11/15 73 (45–92) API 23/23 100 25/25 100 (86–100) 4/4 100 8/8 100 Overall 59/66 90 68/75 91 (85–98) 11/15 73 19/23 83 (61–95)

n/m ϭ number of patients with favorable assessment (cure)/number of patients with a posttreatment assessment. *IfmՆ 30, the response was computed by using Cochran-Mantel-Haenszel weights, adjusting for age within each disease stratum and adjusting for age for “overall.” If m Ͻ 30, then the observed response was computed. †IfmՆ 30, then the CI was calculated by using the normal approximation to the binomial distribution, and Cochran-Mantel-Haenszel weights were used to adjust for stratification (age within each disease stratum, and age for “overall”). If 10 Յ m Ͻ 30, then the exact CI was computed based on Clopper and Pearson’s method. If m Ͻ 10, then no CI was calculated. what lower response rates were observed in the EPP anal- data are consistent with the results obtained previously for ysis. This resulted primarily from the exclusion of patients these indications in adults [4–6]. whose final assessment of “cure” in the cMITT analysis Clinical and microbiological efficacy response rates were occurred before the test-of-cure window. Consistent with comparable in this study between pediatric patients treated the results of the cMITT analysis, however, the overall with ertapenem and patients treated with ticarcillin/clavu- response rates observed in the EPP analysis were also lanate. Patients treated for intra-abdominal infection were higher in the ertapenem treatment group. required to have had a ruptured viscous; all but 1 patient had Microbiological efficacy of treatments against the major a ruptured appendicitis with extension of infection into the pathogens identified (E coli and Bacteroides species) in the peritoneal space. Similar to the study in adults, approxi- cMITT population are shown in Table 6. Ertapenem was mately 63% of patients had evidence of generalized or effective against these pathogens in both disease strata, with localized peritonitis and approximately one third had one or favorable responses ranging from 83% (E coli in cIAI) to more intra-abdominal abscesses identified at the time of 100% (E coli in API). The comparator had favorable re- surgery. A slightly higher percentage of adult patients sponses between 67% (B thetaiotaomicron in cIAI) and (89%) were managed with open surgical intervention as 100% (E coli in API). compared with approximately 76% of patients in this pediatric study. The difference may be a reflection of the fact Comments that virtually all patients in the pediatric study were treated Guidelines from both the Infectious Diseases Society of for infectious complications of a ruptured appendix, America and the Surgical Infection Society support and whereas this represented only about half of the patients recommend the use of ertapenem as an empirical treatment treated in the adult study (the remainder having infections option for adult patients with community-acquired cIAI originating from the colon or other intra-abdominal site) [4]. [13,14]. The results of this small randomized, controlled Adolescent patients treated for acute pelvic infection had study extend the available data on the safety, tolerability, either septic abortion or postpartum endomyometritis. Overall, and efficacy of ertapenem in a pediatric population. These the efficacy results observed for ertapenem in this pediatric study were consistent with that observed previously in adult patients treated with ertapenem for these indications [4–6]. Table 6 Ertapenem was generally well tolerated in pediatric pa- Favorable response by pathogen (cMITT population) tients. Only 1 ertapenem patient reported a serious drug- Ticarcillin/ related clinical adverse experience (moderately intense di- Ertapenem clavulanate arrhea) which required discontinuation. There were no Disease Pathogen Observed Observed serious drug-related laboratory adverse experiences re- response response ported in any patient receiving ertapenem therapy, and no patients discontinued ertapenem therapy as the result of a n/m % n/m % laboratory adverse experience. There were no deaths re- IAI E coli 29/35 83 6/8 75 ported in any patient during the study, and overall the rate B fragilis 12/13 92 3/4 75 of serious adverse experiences was low and of the type B thetaiotaomicron 10/11 91 2/3 67 generally expected in hospitalized pediatric patients treated API E coli 8/8 100 2/2 100 for these types of infectious diseases. n/m ϭ number of pathogens associated with a favorable clinical assess- The overall and the drug-related incidence of clinical and ment/number of pathogens with an assessment. laboratory adverse experiences was low and similar in the 2 374 A.E. Yellin et al. / The American Journal of Surgery 194 (2007) 367–374 treatment groups. The overall safety profile for ertapenem in References pediatric patients in this study was similar to that described [1] Dougherty SH, Sirinek KR, Schauer PR, et al. Ticarcillin/clavulanate for adults. compared with clindamycin/gentamicin (with or without ampicillin) The drug-related adverse experiences were generally for the treatment of intraabdominal infections in pediatric and adult patients. Am Surg 1995;61:297–303. mild and considered of minimal clinical significance. The [2] Stevenson RJ. Appendicitis. In: Rudolph AM, Hoffman JIE, Rudolph most commonly reported drug-related clinical adverse ex- CD, eds. Rudolph’s Pediatrics. 20th ed. Stamford, CT: Appleton & periences during ertapenem parenteral therapy and 14-day Lange; 1996:1105–7. follow-up period included infusion site pain and diarrhea, [3] Hemsell DL, Solomkin JS, Sweet R, et al. Evaluation of new anti- similar to the experience in adult patients receiving ertap- infective drugs for the treatment of acute pelvic infections in hospitalized women. Clin Infect Dis 1992;15(suppl 1):S43–52. enem [4–6]. The incidence of both of these adverse expe- [4] Solomkin JS, Yellin AE, Rotstein OD, et al. Ertapenem versus pip- riences was less in the ertapenem group than in the com- eracillin/tazobactam in the treatment of complicated intraabdominal parator group. No seizures were reported, and no patients infections: results of a double-blind, randomized comparative phase discontinued ertapenem therapy because of a drug-related III trial. Ann Surg 2003;237:235–45. rash. Overall, the incidence of rash was low in both treat- [5] Yellin AE, Hassett JM, Fernandez A, et al. Ertapenem monotherapy ment groups. versus combination therapy with ceftriaxone plus metronidazole for treatment of complicated intraabdominal infections in adults. Int J Drug-related laboratory abnormalities were uncommon Antimicrob Agents 2002;20:165–73. in both treatment groups, and none were considered serious. [6] Roy S, Higareda I, Angel-Muller E, et al. Ertapenem once a day The most commonly reported drug-related laboratory ad- versus piperacillin/tazobactam every 6 hours for treatment of acute verse experiences were mild transaminase elevation, occur- pelvic infections: a prospective, multicenter, randomized, double- ring in 2 patients in the ertapenem group and 1 patient in the blind study. Infect Dis Obstet Gynecol 2003;11:27–37. comparator group. These resolved on discontinuation of [7] Zhanel GG, Johanson C, Embil JM, et al. Ertapenem: review of a new carbapenem. Expert Rev Anti Infect Ther 2005;3:23–39. study drug at completion of treatment. There were no re- [8] Keating GM, Perry CM. Ertapenem. A review of its use in treatment ports of decreased neutrophil counts as an adverse experi- of bacterial infections. Drugs 2005;65:2151–78. ence, and other drug-related laboratory adverse experiences [9] Shah PM, Issacs RD. Ertapenem, the first of a new group of carbap- were also uncommon. Overall, the laboratory safety profile enems. J Antimicrob Chemother 2003;52:538–42. of ertapenem in pediatric patients was similar to that of [10] Mistry G, Blumer J, Topelburg S, et al. Single dose pharmacokinetics (PK) of ertapenem (ERT) in pediatric patients. Poster presented at the ticarcillin/clavulanate and to that reported in adults treated 45th Annual International Conference on Antimicrobial Agents and with ertapenem [4–6]. Chemotherapy, December 17, 2005, Washington, DC. Similar proportions of patients in both treatment groups [11] Beam TR Jr, Gilbert DN, Kunin CM. General guidelines for the reported drug-related infused vein complications during clinical evaluation of anti-infective drug products. Infectious Diseases study therapy. In a separate assessment designed to examine Society of America and the Food and Drug Administration. Clin local tolerability, similar proportions of patients in the treat- Infect Dis 1992;15(suppl 1):S5–12. [12] Solomkin JS, Hemsell DL, Sweet R, et al. Evaluation of new anti- ment groups had signs or symptoms of local intolerance, infective drugs for the treatment of intraabdominal infections. Infec- most of which were considered mild. Overall, infusion of tious Diseases Society of America and the Food and Drug Adminis- ertapenem appeared to be at least as well tolerated as ticar- tration. Clin Infect Dis 1992;15(suppl 1):S33–42. cillin/clavulanate. [13] Solomkin JS, Mazuski JE, Baron EJ, et al. Guidelines for the selec- In summary, results from this study, similar to those tion of anti-infective agents for complicated intra-abdominal infec- observed in adults, show that ertapenem administered 1 g tions. Clin Infect Dis 2003;37:997–1005. [14] Mazuski JE, Sawyer RG, Nathens AB, et al. The Surgical Infection parenterally per day to children 13 to 17 years of age or 15 Society Guidelines on antimicrobial therapy for treatment of compli- mg/kg twice daily to children 2 to 12 years of age is a safe cated intra-abdominal infections: an executive summary. Surg Infect and effective therapeutic regimen for treating cIAI or API. 2002;3:161–73.