- Email: [email protected]
Selection of first-line i.v. antibiotics for acute pyelonephritis in patients requiring emergency hospital admission
J Infect Chemother (2011) 17:231–237 DOI 10.1007/s10156-010-0116-y
ORIGINAL ARTICLE
Selection of first-line i.v. antibiotics for acute pyelonephritis in patients requiring emergency hospital admission Tomihiko Yasufuku • Katsumi Shigemura Masuo Yamashita • Soichi Arakawa • Masato Fujisawa
•
Received: 14 December 2009 / Accepted: 27 July 2010 / Published online: 14 September 2010 Ó Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases 2010
Abstract Febrile urinary tract infections (UTIs) often require the intravenous infusion of antibiotics and/or hospitalization. Acute pyelonephritis (AP) is one of the most severe forms of UTI, and the antibiotics we should use as the first line and the risk factors for treatment failure remain controversial. The objective of this study was to investigate the efficacy of i.v. antibiotics selected for the treatment of febrile AP and to examine the risk factors for antibiotic resistance. We set risk factors for antibiotic treatment failure such as age, sex, and the presence of underlying urinary tract disease. We classified all cases into 49 cases of complicated AP and 24 cases of uncomplicated AP according to the presence of underlying urinary tract diseases, and examined the characteristics of the patients and the efficacy of the antibiotics used in this study. We investigated risk factors which relate to initial treatment failure and the duration of antibiotic treatment. Initial antibiotic treatment failure was significantly correlated to C-reactive protein in complicated AP and to positive blood culture in uncomplicated AP. We revealed a significant correlation between the duration of the given antibiotics and diabetes mellitus or positive blood culture in uncomplicated AP, and tazobactam/piperacillin was significantly related to prolongation of antibiotic treatment in complicated AP. In conclusion, in this study, a positive blood
T. Yasufuku K. Shigemura (&) S. Arakawa M. Fujisawa Division of Urology, Department of Surgery, Kobe University Graduate School of Medicine, Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan e-mail: [email protected] K. Shigemura M. Yamashita Department of Urology, Akashi Municipal Hospital, Akashi, Japan
culture was the representative risk factor that related to both initial treatment failure and longer duration of the given antibiotics in uncomplicated AP. Keywords
Pyelonephritis Emergency Empiric therapy
Introduction Acute pyelonephritis (AP) [1–3] is a common urinary tract infection (UTI) which often requires emergency hospitalization. Infections are mostly caused by bacteria in the urinary tract, and the most common pathogens belong to the Enterobacteriaceae family, especially Escherichia coli (E. coli) [4]. AP is a frequent and possibly severe infection, and accounts for more than 100,000 hospitalizations per year in the United States [5]. The severity of AP varies from relatively mild to life-threatening [4]. It is necessary to diagnose and evaluate the severity of the disease correctly, and the decision to hospitalize should be based an the evaluation of the patient’s clinical status and the estimated risk factors, such as diabetes mellitus (DM) [6]. Treatment strategies in AP include the use of antibiotics with or without surgical procedures such as ureteral catheterization, percutaneous drainage, or nephrectomy, especially for life-threatening renal abscess [1, 5, 7]. It is important to select the proper antibiotics to achieve treatment success and to avoid the development of antimicrobial resistance [4]. The first-line antibiotic that should be used and how long it should be administered remain controversial issues [2, 4]. Previous reports demonstrated the efficacy of cephems in UTI cases [6, 8] and recommended that severe cases should be treated with fluoroquinolones or extended-spectrum cephalosporin with or without associated with aminoglycoside [5, 9].
123
232
J Infect Chemother (2011) 17:231–237
In this study, we investigated the statistical correlation of the antibiotics used or the risk factors with treatment success or duration of antibiotic treatment in both complicated and uncomplicated febrile AP cases which required hospitalization with intravenous antimicrobial agents.
and examined the difference in risk factors for complicated and uncomplicated AP. Previous use of an antibiotic was defined as the administration of an antibiotic for more than 2 days during the previous 1 year [10].
Patients and methods
There were patients who were cured by the initial intravenous antibiotic and cases in which the initial antibiotics used were less effective at improving fever and pyuria, so a change to another antibiotic was required. The cases that needed a change to a second-line antibiotic were defined as ‘‘initial treatment failure.’’ In addition, we evaluated the duration of the intravenous antibiotic used initially. We investigated the difference between complicated and uncomplicated AP regarding the rate of decrease of white blood cells (WBC), C-reactive protein (CRP), and causative bacteria.
Patient selection We examined 73 adult patients diagnosed with febrile AP who were hospitalized in Akashi Municipal Hospital in Japan from August 2002 to November 2008. AP was defined on the basis of clinical criteria (fever, costovertebral pain, and dysuric symptoms) and microbiologic criteria (pyuria and/or positive urine culture with C104 cfu/ ml) [4]. Three cases needing drainage were excluded from our study, and 70 cases of AP treated with antibiotics only were analyzed retrospectively. All cases were febrile (C37.0°C) and treated with i.v. antibiotics only, such as cephems, penicillins, carbapenems, or fluoroquinolones. Our cases were classified as 49 cases of complicated AP and 24 cases of uncomplicated AP, and risk factors and the difference between each group were investigated. Antibiotics The antibiotics used in this study were ceftazidime (CAZ), cefotiam (CTM), ceftriaxone (CTRX), cefozopran (CZOP), sulbactam/cefoperazone (SBT/CPZ), flomoxef (FMOX), piperacillin (PIPC), tazobactam/piperacillin (TAZ/PIPC), sulbactam/ampicillin (SBT/ABPC), meropenem (MEPM), doripenem (DRPM), imipenem/cilastatin (IPM/CS), ciprofloxacin (CPFX), and vancomycin (VCM).
Evaluation of antibiotic efficacy of the initial antibiotic treatment and duration of the treatment
Statistical analysis Statistical analyses were performed using STATA (StataCorp LP, Lakeway Drive, College Station, TX, USA); p \ 0.05 was considered to indicate statistical significance. We examined the difference in risk factors and laboratory data such as WBC and CRP between complicated and uncomplicated AP by chi-square and the Student t test. In addition, we investigated the relationship between initial antibiotic treatment failure and risk factors and between the duration of the given antibiotics and the kind of antibiotics used by univariate logistic regression analyses.
Results
Susceptibility testing
Patient characteristics
Susceptibility testing for each antibiotic was performed according to Clinical Laboratory Standards Institute (CLSI) guideline M100-S18, 2008, using the agar plate dilution method. The isolates were inoculated with 104 CFU/ml and the antibiotic susceptibilities were tested. E. coli ATCC 25922 was used as a quality control.
Patient characteristics and symptoms on admission are shown in Tables 1 and 2. The median age of the patients was 71 (range 20–89) years old, among 61 females (83.6%) and 12 males (16.4%). There were 49 patients (67.1%) with underlying urinary tract disease (complicated AP), 18 patients who had previously used antibiotics, and 3 patients with DM (6.6%) (Table 1). There were more female patients (p = 0.05) and patients who had previously used antibiotics (p = 0.05) in complicated AP cases than in uncomplicated AP cases. There was no significant difference in positive blood culture (p = 0.27) or septic shock (p = 0.16) between the complicated AP and uncomplicated AP cases (Table 3).
Risk factors for antibiotic treatment We set risk factors for antibiotic treatment failure (age, gender, DM, the presence of indwelling urinary catheter, the previous use of antibiotics, vital shock signs such as an altered level of consciousness, and positive blood culture)
123
J Infect Chemother (2011) 17:231–237 Table 1 Patient characteristics and underlying urinary tract diseases
Urine culture was performed in 50 cases; blood culture was performed in 30 cases DM diabetes mellitus, WBC white blood cell, CRP C-reactive protein, VUR vesicoureteral reflux, BPH benign prostatic hyperplasia
Table 2 Symptoms of the patients in this study
Characteristics (n = 73)
No. (%)
Underlying urinary tract disease
No. (%)
Gender (female)
61 (83.6)
Hydronephrosis
22 (30.1)
Positive urine culture
34/50 (68.0)
Ureteral stenosis
14 (19.2)
Underlying urinary tract disease (complicated UTI)
49 (67.1)
Ureteral stone
13 (17.8)
Positive blood culture
9/30 (29.0)
Renal stone
9 (12.3)
Indwelling catheter
19 (26.0)
Neurogenic bladder
4 (5.5)
Previous use of antibiotics
18 (24.7)
Bladder tumor
4 (5.5)
Septic shock Patients with DM
5 (6.8) 3 (4.1)
VUR BPH
4 (5.5) 1 (1.4)
Median (range)
Overactive bladder
1 (1.4)
Age (years old)
71 (20–89)
Ureteral tumor
1 (1.4)
Period of treatment with i.v. antibiotics (days)
4 (1–9)
Neobladder (rectal carcinoma)
1 (1.4)
Body temperature on admission (°C)
38.6 (37.0–41.8)
Megaureter
1 (1.4)
Body temperature after treatment (°C)
36.5 (35.7–36.9)
Ureterocutaneostomy
1 (1.4)
3
Peripheral WBC before treatment (/mm )
11100 (2100–30600)
Serum CRP before treatment (mg/dl)
10.8 (0–34.8)
Symptoms
No. (%)
OR
p value
Complicated AP (n = 49)
Uncomplicated AP (n = 24)
23 (46.9)
14 (58.3)
0.63
0.36
16 (32.7)
11 (45.8)
0.57
0.27
6 (12.2)
2 (8.3)
0.72
0.73
Flank pain Appetite loss
3 (6.1) 6 (12.2)
2 (8.3) 4 (16.7)
1.53 0.7
0.62 0.61
General fatigue
6 (12.2)
4 (16.7)
0.72
0.73
Shivering
4 (8.2)
2 (8.3)
0.7
0.61
Vomiting
3 (6.1)
2 (8.3)
0.46
0.35
Macrohematuria
3 (6.1)
3 (12.5)
N/A
0.32
Bladder tamponade
2 (4.1)
0 (0.0)
0.98
0.98
Pain Lumbago Abdominal pain
AP acute pyelonephritis, OR odds ratio, N/A not applicable
233
Altered level of consciousness
2 (4.1)
0 (0.0)
N/A
0.32
Pollakisuria
2 (4.1)
3 (12.5)
0.3
0.18
Feeling of residual urine
2 (4.1)
3 (12.5)
0.3
0.18
Nausea
1 (2.0)
0 (0.0)
N/A
0.48
Miction pain
0 (0.0)
1 (4.2)
0
0.15
First-line and second-line antibiotics used CAZ (30.6%) was most frequently used as the first-line antibiotic in complicated AP and CZOP (25.0%) in uncomplicated AP. Carbapenems such as DRPM and MEPM were less frequently used as first-line antibiotics in both complicated AP (6.2%) and uncomplicated AP (4.2%) (Table 4). On the other hand, carbapenems were used in 5 out of 11 cases as the second-line treatment for complicated AP and both cases of uncomplicated AP. The first-
line antibiotic therapy failed in 11 out of the 49 cases (22.5%) of complicated AP and 2 out of the 24 cases (8.3%) of uncomplicated AP (Table 5). Causative bacteria isolated from urine culture Regarding the causative bacteria isolated from urine, 9 out of the 28 strains (32.1%) isolated in complicated AP cases and 9 out of the 19 strains (47.4%) isolated in uncomplicated AP cases were E. coli. There were 3 strains of
123
234
J Infect Chemother (2011) 17:231–237
Table 3 Risk factors in complicated and uncomplicated UTI cases
Urine culture was performed in 32 cases in complicated AP and 18 cases in uncomplicated AP; blood culture was performed in 19 cases in complicated AP and 11 cases in uncomplicated AP AP acute pyelonephritis, OR odds ratio The p value (and OR) with a significant difference in statistical analysis was expressed as bold values
Risk factor
Complicated AP (n = 49) Underlying urinary tract disease
49 (100.0)
Indwelling catheter
19 (38.8)
Cephems CAZ CZOP CTM
OR
38 (77.6)
23 (95.8)
Previous use of antibiotics
17 (34.7)
1 (4.2)
0.15 0.05 12.22 0.01
Patients with DM
1 (2.0)
2 (8.3)
0.23 0.2
Septic shock
4 (8.2)
1 (4.2)
2.04 0.53
Positive urine culture Positive blood culture
20/32 (62.5) 4/19 (21.1)
14/18 (77.8) 5/11 (45.5)
0.48 0.27 0.32 0.16
Median (range) Age (years old)
70 (20–89)
71 (23–77)
Body temperature on admission (°C)
38.6 (37.0–41.8)
38.5 (37.1–40.7)
0.28
Body temperature after treatment (°C)
36.5 (5.6%)
36.5 (35.7–36.9)
0.41
Period of treatment (days)
4 (1–9)
5 (2–9)
0.01
0.04
Peripheral WBC before treatment (/mm3) 11200 (2100–30600) 11400 (3400–28400)
0.38
Serum CRP before treatment (mg/dl)
0.58
10.7 (0–34.8)
10.8 (0.2–28.5)
Table 5 First- and second-line antibiotics used in the 13 cases that required a change in antibiotic
No. (%)
Case no. First-line antibiotics Days Second-line antibiotics Days
Complicated AP (n = 49)
Uncomplicated AP (n = 24)
40 (81.6)
17 (70.8)
1
CTM
2
CZOP
5 (20.8)
2
CTM
1
TAZ/PIPC
6
6 (25.0) 4 (16.6)
3
CTM
6
MEPM
5
4
CAZ
1
CZOP
5
CAZ
5
MEPM
2 7 5
15 (30.6) 3 (6.1) 10 (20.4)
Complicated AP (n = 11)
CTRX
9 (18.3)
1 (4.2)
5
FMOX
3 (4.1)
1 (4.2)
SBT/CPZ
1 (2.1)
0 (0)
6 7
CAZ CTRX
3 4
IPM DRPM
6 (12.2)
5 (20.8)
8
TAZ/PIPC
7
MEPM
TAZ/PIPC
4
CAZ
Penicillins
p value
Uncomplicated AP (n = 24)
Gender (female)
Table 4 First-line antibiotics used Antibiotics
No. (%)
7
7
TAZ/PIPC
6 (12.2)
3 (12.4)
9
SBT/ABPC
0 (0)
1 (4.2)
10
DRPM
1
TAZ/PIPC
3
0 (0)
1 (4.2)
11
DRPM
5
VCM
4
3 (6.2)
1 (4.2)
2 (4.1)
1 (4.2)
1
CAZ
2
MEPM
5
1 (2.1)
0 (0)
2
PIPC
3
MEPM
7
0 (0)
1 (4.2)
0 (0)
1 (4.2)
49 (100)
24 (100)
PIPC Carbapenems DRPM MEPM Fluoroquinolones CPFX Total
AP acute pyelonephritis
Citrobacter freundii and 2 strains of methicillin-resistant Staphlococcus aureus (MRSA) in complicated AP cases; however, there was no significant difference in causative bacteria between complicated and uncomplicated AP cases (Table 6). E. coli showed no significant difference in the rate of antibiotic resistance between complicated and uncomplicated AP cases (Table 7).
123
10
Uncomplicated AP (n = 2)
AP acute pyelonephritis
Correlation of initial antibiotic treatment failure with risk factors and with each antibiotic Initial antibiotic treatment failure was significantly related to CRP [odds ratio (OR) = 1.11, p = 0.02] in complicated AP and to positive blood culture [OR = not applicable (N/A), p \ 0.01] in uncomplicated AP (Table 8). In addition, there was no significant correlation of initial treatment failure with a particular antibiotic.
J Infect Chemother (2011) 17:231–237
235
Table 6 Causative bacteria isolated from urine cultures Bacteria
No. (%)
Table 7 Antibiotic resistance of E. coli isolated from urine cultures in complicated and uncomplicated AP cases Antibiotics
Resistant strains; no. (%)
Complicated AP (n = 28a)
Uncomplicated AP (n = 19a)
Escherichia coli
9 (32.1)
9 (47.4)
Citrobacter freundii
3 (10.7)
0 (0.0)
ABPC
3 (33.3) 3 (33.3)
5 (50.0)
CEZ
1 (11.1)
1 (10.0)
Complicated AP (n = 9)
Uncomplicated AP (n = 10) 5 (50.0)
Staphlococcus aureus MRSA
2 (7.1) 2 (7.1)
1 (5.3) 0 (0.0)
PIPC
Klebsiella pneumoniae
1 (3.6)
2 (10.5)
CTM
0 (0)
0 (0)
0 (0)
0 (0)
Proteus mirabilis
1 (3.6)
1 (3.6)
CAZ
Enterococcus faecalis
1 (3.6)
0 (0.0)
CPR
0 (0)
0 (0)
0 (0.0)
CMZ
0 (0)
0 (0)
0 (0.0)
CCL
0 (0)
0 (0)
a
Enterobacter aerogenes Acinetobacter baumanii
1 (3.6) 1 (3.6)
Candida species
1 (3.6)
0 (0.0)
CPDX
0/5 (0)
0 (0)
Gram-positive rods
1 (3.6)
0 (0.0)
CFPN
0/5a (0)
0/6b (0)
Prevotella melaninogenica
1 (3.6)
0 (0.0)
FMOX
0 (0)
0 (0)
4 (21.1)
IPM
0 (0)
0 (0)
AZT
0 (0)
0 (0)
GM AMK
2 (22.2) 0 (0)
0 (0) 0 (0)
MINO
0 (0)
2 (20)
FOM
0 (0)
0 (0)
LVFX
3 (33.3)
3 (30)
ST
0 (0)
3 (30)
C/A
a
0/5 (0)
0/6b (0)
S/C
0 (0)
0 (0)
No growth
9 (32.1)
Twenty-eight strains were isolated from urine cultures in complicated AP cases and 19 strains were isolated in uncomplicated AP cases AP acute pyelonephritis, MRSA methicillin-resistant Staphlococcus aureus
Correlation of the duration of the given initial antibiotic with risk factors and with the each antibiotic We revealed significant correlations between longer duration of antibiotics and DM (OR = 3.55, p = 0.01) and positive blood culture (OR = N/A, p \ 0.01) in cases of uncomplicated AP by logistic regression analyses. We also demonstrated that age (OR = 0.05, p = 0.02) was to related to shorter duration of the given initial antibiotics in complicated AP cases. In addition, TAZ/PIPC (OR = 3.46, p = 0.01) was significantly correlated with longer duration of antibiotic treatment in complicated AP (Table 9).
Discussion UTIs are major infectious diseases, often requiring intravenous antibiotics and hospitalization for treatment. Previous reports showed that 29.5% of the patients who visited the emergency room had UTIs, and among them, the most frequent diagnosis was acute uncomplicated UTI (24.2%) [11]. UTIs are common and potentially severe infections which emergency doctors must manage effectively; however, the best therapy remains controversial, as which intravenous antibiotics should be selected as the first-line treatment has not been established [2]. This study investigated the efficacy and risk factors of antibiotic treatments in febrile AP cases.
AP acute pyelonephritis a
Susceptibility to CPDX, CFPN, and C/A was tested for 5 strains from complicated AP cases
b
Susceptibility to CFPN and C/A was tested for 6 strains from uncomplicated AP cases
Risk factors such as recent hospitalization, previous use of antibiotics, and immunosuppression (for example due to steroid use) were reported to relate to resistant pathogens, indicating that clinical factors available at presentation can be used for the risk stratification of patients with AP [12]. Our study revealed that positive blood culture related to initial treatment failure and the duration of the given initial antibiotics in uncomplicated AP cases. Higher CRP led to initial treatment failure, and TAZ/PIPC was related to longer duration of initial antibiotic treatment in complicated AP cases, suggesting that DM is the major risk factor and penicillins may not suitable for the treatment of febrile AP. Previous reports demonstrated that AP is often treated with i.v. fluoroquinolones or extended-spectrum cephalosporins and that emergent admission is often required in severe cases [6]. Cephems play a valuable role in initial treatment of patients with acute urogenital infections [5, 7– 9]. In our study, cephems such as CAZ and CZOP were the most frequently used antibiotics, and they showed tolerable
123
236
J Infect Chemother (2011) 17:231–237
Table 8 Correlation of risk factors with initial treatment failure, as assessed by univariate analysis
Complicated AP
Uncomplicated AP
OR (95% CI)
p value
OR (95% CI)
p value
Age
1.05 (1.00–1.11)
0.07
1.92 (0.68–5.40)
0.22
Gender (female)
1.40 (0.25–7.68)
0.7
N/A
N/A
Patients with DM
N/A
N/A
21 (0.69–642.98)
0.08
Risk factor
AP acute pyelonephritis, OR odds ratio, CI confidence interval, N/A not applicable, DM diabetes mellitus, CRP C-reactive protein The p value (and OR) with a significant difference in statistical analysis was expressed as bold values
Indwelling catheter
1.43 (0.37–5.55)
0.61
N/A
N/A
Previous use of antibiotics
2.95 (0.74–11.69)
0.13
N/A
N/A
Septic shock
1.17 (0.11–12.48)
0.9
N/A
N/A
Positive blood culture
0.67 (0.54–8.16)
0.75
N/A
<0.01
Peripheral WBC before treatment
1.00 (1.00–1.00)
0.44
1.00 (1.00–1.00)
0.87
Serum CRP before treatment
1.11 (1.02–1.21)
0.02
0.97 (0.80–1.19)
0.81
Laboratory data
Table 9 Correlation of risk factors with duration of initial antibiotic by univariate analyses
Complicated AP
Uncomplicated AP
b coeff. (95% CI)
p value
b coeff. (95% CI)
p value
0.46
Risk factor
AP acute pyelonephritis, b coeff. b coefficient, CI confidence interval, DM diabetes mellitus, N/A not applicable, CRP C-reactive protein The p value (and b coefficient), with a significant difference in statistical analysis was expressed as bold values
Age
0.05 (0.01–0.09)
0.02
0.01 (-0.02 to 0.05)
Gender (female)
0.09 (-1.77 to 1.95)
0.92
1.30 (-2.60 to 5.21)
0.50
Patients with DM
N/A
N/A
3.55 (1.16–5.93)
0.01
Indwelling catheter
0.10 (-1.69 to 1.50)
0.90
N/A
N/A
Previous use of antibiotics
0.76 (-0.85 to 2.38)
0.35
0.26 (-4.21 to 3.69)
0.89
Septic shock
1.04 (-1.78 to 3.86)
0.46
0.78 (-3.15 to 4.71)
0.46
Positive blood culture
0.35 (-2.53 to 3.23)
0.80
2.53 (0.70–4.37)
0.01
0.01 (-0.01 to 0.01) 0.08 (-0.01 to 0.17)
0.78 0.09
0.01 (–0.01 to 0.01) 0.08 (-0.02 to 0.17)
0.63 0.11
TAZ/PIPC
3.46 (1.32–5.60)
0.01
1.62 (0.66–3.90)
0.63
PIPC
N/A
N/A
N/A
N/A
Laboratory data Peripheral WBC before treatment Serum CRP before treatment Antibiotic
efficacy when used as the first-line antibiotics in empirical treatments, as noted in previous reports. Previous reports demonstrated that the causative bacteria in serious UTI cases were E. coli \60%, other Enterobacteriaceae such as Klebsiella pneumoniae, C. freundii, and E. faecalis, Gram-negative bacilli such as Pseudomonas aeruginosa, and Gram-positive bacteria such as Staphylococcus aureus, and they showed the efficacy of advancedgeneration cephalosporins and fluoroquinolones [3]. The most common bacteria other than E. coli in this study were S. aureus, C. freundii, and K. pneumonia, as also noted in previous reports. Therefore, we considered there was good indication for the use of cephems—especially advancedgeneration cephalosporins—as the first-line antibiotics for these isolates of bacteria, as shown in previous reports [6, 8]. Regarding bacterial antibiotic resistance, previous reports revealed that female AP caused by ESBL-producing
123
bacteria could not be treated by CTRX [13]. Among the uropathogenic strains isolated from UTI patients in Japan, the proportion of E. coli strains that were less sensitive to fluoroquinolone was reported to be over 20%, and isolates of ESBL-producing E. coli were found to be increasing [14, 15]. In this study, E. coli isolated from urine cultures showed[30% resistance to ABPC, PIPC and LVFX in both complicated and uncomplicated AP cases, and there was no ESBL-producing E. coli. Multidrug-resistant P. aeruginosa and MRSA were also difficult to treat, and the rates of susceptibility of Enterococcus faecium to beta-lactams and fluoroquinolones were 0 and 6%, respectively [15]. Among our cases, there were 2 cases of MRSA in complicated AP, but no multidrug-resistant P. aeruginosa that was difficult to treat. In conclusion, we have demonstrated that CRP is related to initial treatment failure and TAZ/PIPC is related to
J Infect Chemother (2011) 17:231–237
longer duration of antibiotic treatment in complicated AP cases, while positive blood culture was significantly correlated with initial antibiotic treatment failure and longer duration of antibiotic treatment in uncomplicated AP cases. It is necessary to assess risk factors and assign the appropriate treatment strategy immediately for the patients with febrile AP. We plan to conduct a prospective randomized study of antibiotic efficacy to establish an effective standard of treatment for AP. Acknowledgments We thank Drs. Masaru Lee and Yoshihisa Kinoshita for taking care of patients, Dr. Daisuke Sugiyama for statistical analyses, and Gary Mawyer, M.F.A., for editing the English of this article.
References 1. Barbara PG, Elena C, Francesca R, Valerio V, Mario SR, Mauro F. The clinical spectrum of acute ‘‘uncomplicated’’ pyelonephritis from an emergency medicine perspective. Int J Antimicrob Agents. 2008;31:46–53. 2. Miller O 2nd, Hemphill RR. Urinary tract infection and pyelonephritis. Emerg Med Clin N Am. 2001;19:655–74. 3. Carson C, Naber KG. Role of fluoroquinolones in the treatment of serious bacterial urinary tract infections. Drugs. 2004;64:1359–73. 4. Kyriakidou KG, Rafailidis P, Matthaiou DK, Athanasiou S, Falagas ME. Short- versus long-course antibiotic therapy for acute pyelonephritis in adolescents and adults: a meta-analysis of randomized controlled trials. Clin Ther. 2008;30:1859–68. 5. Rollino C. Acute pyelonephritis in adults. G Ital Nefrol. 2007; 24:121–31.
237 6. Molina Caban˜ero JC, Ochoa Sangrador C. Grupo Investigador del Proyecto Hospitalization criteria in urinary tract infections. An Pediatr (Barc). 2007;67:469–77. 7. Bruye`re F, Cariou G, Boiteux JP, Hoznek A, Mignard JP, Escaravage L, et al. Acute prostatitis. Prog Urol. 2008;18:19–23. 8. Arakawa S, Kamidono S. Clinical evaluation of drug efficacy in UTI in men: the significance of bacterial elimination on day one. Infection. 1994;22:23–6. 9. Bruye`re F, Cariou G, Boiteux JP, Hoznek A, Mignard JP, Escaravage L, et al. Acute pyelonephritis. Prog Urol. 2008;18:14–8. 10. Arslan H, Azap OK, Ergonul O, Timurkaynak F, Urinary Tract Infection Study Group. Risk factors for ciprofloxacin resistance among Escherichia coli strains isolated from communityacquired urinary tract infections in Turkey. J Antimicrob Chemother. 2005;56:914–8. 11. den Engelsen C, van der Werf C, Matute AJ, Delgado E, Schurink CA, Hoepelman AI. Infectious diseases and the use of antibiotics in outpatients at the emergency department of the University Hospital of Leon, Nicaragua. Int J Infect Dis. 2009;13:349–54. 12. Efstathiou SP, Pefanis AV, Tsioulos DI, Zacharos ID, Tsiakou AG, Mitromaras AG, et al. Acute pyelonephritis in adults: prediction of mortality and failure of treatment. Arch Intern Med. 2003;163:1206–12. 13. Suankratay C, Jutivorakool K, Jirajariyavej S. A prospective study ceftriaxone treatment in acute pyelonephritis caused by extended-spectrum beta-lactamase-producing bacteria. J Med Assoc Thail. 2008;91:1172–81. 14. Shigemura K, Arakawa S, Miura T, Nakano Y, Tanaka K, Fujisawa M. Significance of fluoroquinolone-resistant Escherichia coli in urinary tract infections. Jpn J Infect Dis. 2008;61:226–8. 15. Ishikawa K, Hayakawa S, Miyakawa S, Kusaka M, Shiroki R, Hoshinaga K. Survey of the susceptibility of urinary isolates to antibacterial agents in 2003. J Infect Chemother. 2003;11:44–7.
123
ORIGINAL ARTICLE
Selection of first-line i.v. antibiotics for acute pyelonephritis in patients requiring emergency hospital admission Tomihiko Yasufuku • Katsumi Shigemura Masuo Yamashita • Soichi Arakawa • Masato Fujisawa
•
Received: 14 December 2009 / Accepted: 27 July 2010 / Published online: 14 September 2010 Ó Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases 2010
Abstract Febrile urinary tract infections (UTIs) often require the intravenous infusion of antibiotics and/or hospitalization. Acute pyelonephritis (AP) is one of the most severe forms of UTI, and the antibiotics we should use as the first line and the risk factors for treatment failure remain controversial. The objective of this study was to investigate the efficacy of i.v. antibiotics selected for the treatment of febrile AP and to examine the risk factors for antibiotic resistance. We set risk factors for antibiotic treatment failure such as age, sex, and the presence of underlying urinary tract disease. We classified all cases into 49 cases of complicated AP and 24 cases of uncomplicated AP according to the presence of underlying urinary tract diseases, and examined the characteristics of the patients and the efficacy of the antibiotics used in this study. We investigated risk factors which relate to initial treatment failure and the duration of antibiotic treatment. Initial antibiotic treatment failure was significantly correlated to C-reactive protein in complicated AP and to positive blood culture in uncomplicated AP. We revealed a significant correlation between the duration of the given antibiotics and diabetes mellitus or positive blood culture in uncomplicated AP, and tazobactam/piperacillin was significantly related to prolongation of antibiotic treatment in complicated AP. In conclusion, in this study, a positive blood
T. Yasufuku K. Shigemura (&) S. Arakawa M. Fujisawa Division of Urology, Department of Surgery, Kobe University Graduate School of Medicine, Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan e-mail: [email protected] K. Shigemura M. Yamashita Department of Urology, Akashi Municipal Hospital, Akashi, Japan
culture was the representative risk factor that related to both initial treatment failure and longer duration of the given antibiotics in uncomplicated AP. Keywords
Pyelonephritis Emergency Empiric therapy
Introduction Acute pyelonephritis (AP) [1–3] is a common urinary tract infection (UTI) which often requires emergency hospitalization. Infections are mostly caused by bacteria in the urinary tract, and the most common pathogens belong to the Enterobacteriaceae family, especially Escherichia coli (E. coli) [4]. AP is a frequent and possibly severe infection, and accounts for more than 100,000 hospitalizations per year in the United States [5]. The severity of AP varies from relatively mild to life-threatening [4]. It is necessary to diagnose and evaluate the severity of the disease correctly, and the decision to hospitalize should be based an the evaluation of the patient’s clinical status and the estimated risk factors, such as diabetes mellitus (DM) [6]. Treatment strategies in AP include the use of antibiotics with or without surgical procedures such as ureteral catheterization, percutaneous drainage, or nephrectomy, especially for life-threatening renal abscess [1, 5, 7]. It is important to select the proper antibiotics to achieve treatment success and to avoid the development of antimicrobial resistance [4]. The first-line antibiotic that should be used and how long it should be administered remain controversial issues [2, 4]. Previous reports demonstrated the efficacy of cephems in UTI cases [6, 8] and recommended that severe cases should be treated with fluoroquinolones or extended-spectrum cephalosporin with or without associated with aminoglycoside [5, 9].
123
232
J Infect Chemother (2011) 17:231–237
In this study, we investigated the statistical correlation of the antibiotics used or the risk factors with treatment success or duration of antibiotic treatment in both complicated and uncomplicated febrile AP cases which required hospitalization with intravenous antimicrobial agents.
and examined the difference in risk factors for complicated and uncomplicated AP. Previous use of an antibiotic was defined as the administration of an antibiotic for more than 2 days during the previous 1 year [10].
Patients and methods
There were patients who were cured by the initial intravenous antibiotic and cases in which the initial antibiotics used were less effective at improving fever and pyuria, so a change to another antibiotic was required. The cases that needed a change to a second-line antibiotic were defined as ‘‘initial treatment failure.’’ In addition, we evaluated the duration of the intravenous antibiotic used initially. We investigated the difference between complicated and uncomplicated AP regarding the rate of decrease of white blood cells (WBC), C-reactive protein (CRP), and causative bacteria.
Patient selection We examined 73 adult patients diagnosed with febrile AP who were hospitalized in Akashi Municipal Hospital in Japan from August 2002 to November 2008. AP was defined on the basis of clinical criteria (fever, costovertebral pain, and dysuric symptoms) and microbiologic criteria (pyuria and/or positive urine culture with C104 cfu/ ml) [4]. Three cases needing drainage were excluded from our study, and 70 cases of AP treated with antibiotics only were analyzed retrospectively. All cases were febrile (C37.0°C) and treated with i.v. antibiotics only, such as cephems, penicillins, carbapenems, or fluoroquinolones. Our cases were classified as 49 cases of complicated AP and 24 cases of uncomplicated AP, and risk factors and the difference between each group were investigated. Antibiotics The antibiotics used in this study were ceftazidime (CAZ), cefotiam (CTM), ceftriaxone (CTRX), cefozopran (CZOP), sulbactam/cefoperazone (SBT/CPZ), flomoxef (FMOX), piperacillin (PIPC), tazobactam/piperacillin (TAZ/PIPC), sulbactam/ampicillin (SBT/ABPC), meropenem (MEPM), doripenem (DRPM), imipenem/cilastatin (IPM/CS), ciprofloxacin (CPFX), and vancomycin (VCM).
Evaluation of antibiotic efficacy of the initial antibiotic treatment and duration of the treatment
Statistical analysis Statistical analyses were performed using STATA (StataCorp LP, Lakeway Drive, College Station, TX, USA); p \ 0.05 was considered to indicate statistical significance. We examined the difference in risk factors and laboratory data such as WBC and CRP between complicated and uncomplicated AP by chi-square and the Student t test. In addition, we investigated the relationship between initial antibiotic treatment failure and risk factors and between the duration of the given antibiotics and the kind of antibiotics used by univariate logistic regression analyses.
Results
Susceptibility testing
Patient characteristics
Susceptibility testing for each antibiotic was performed according to Clinical Laboratory Standards Institute (CLSI) guideline M100-S18, 2008, using the agar plate dilution method. The isolates were inoculated with 104 CFU/ml and the antibiotic susceptibilities were tested. E. coli ATCC 25922 was used as a quality control.
Patient characteristics and symptoms on admission are shown in Tables 1 and 2. The median age of the patients was 71 (range 20–89) years old, among 61 females (83.6%) and 12 males (16.4%). There were 49 patients (67.1%) with underlying urinary tract disease (complicated AP), 18 patients who had previously used antibiotics, and 3 patients with DM (6.6%) (Table 1). There were more female patients (p = 0.05) and patients who had previously used antibiotics (p = 0.05) in complicated AP cases than in uncomplicated AP cases. There was no significant difference in positive blood culture (p = 0.27) or septic shock (p = 0.16) between the complicated AP and uncomplicated AP cases (Table 3).
Risk factors for antibiotic treatment We set risk factors for antibiotic treatment failure (age, gender, DM, the presence of indwelling urinary catheter, the previous use of antibiotics, vital shock signs such as an altered level of consciousness, and positive blood culture)
123
J Infect Chemother (2011) 17:231–237 Table 1 Patient characteristics and underlying urinary tract diseases
Urine culture was performed in 50 cases; blood culture was performed in 30 cases DM diabetes mellitus, WBC white blood cell, CRP C-reactive protein, VUR vesicoureteral reflux, BPH benign prostatic hyperplasia
Table 2 Symptoms of the patients in this study
Characteristics (n = 73)
No. (%)
Underlying urinary tract disease
No. (%)
Gender (female)
61 (83.6)
Hydronephrosis
22 (30.1)
Positive urine culture
34/50 (68.0)
Ureteral stenosis
14 (19.2)
Underlying urinary tract disease (complicated UTI)
49 (67.1)
Ureteral stone
13 (17.8)
Positive blood culture
9/30 (29.0)
Renal stone
9 (12.3)
Indwelling catheter
19 (26.0)
Neurogenic bladder
4 (5.5)
Previous use of antibiotics
18 (24.7)
Bladder tumor
4 (5.5)
Septic shock Patients with DM
5 (6.8) 3 (4.1)
VUR BPH
4 (5.5) 1 (1.4)
Median (range)
Overactive bladder
1 (1.4)
Age (years old)
71 (20–89)
Ureteral tumor
1 (1.4)
Period of treatment with i.v. antibiotics (days)
4 (1–9)
Neobladder (rectal carcinoma)
1 (1.4)
Body temperature on admission (°C)
38.6 (37.0–41.8)
Megaureter
1 (1.4)
Body temperature after treatment (°C)
36.5 (35.7–36.9)
Ureterocutaneostomy
1 (1.4)
3
Peripheral WBC before treatment (/mm )
11100 (2100–30600)
Serum CRP before treatment (mg/dl)
10.8 (0–34.8)
Symptoms
No. (%)
OR
p value
Complicated AP (n = 49)
Uncomplicated AP (n = 24)
23 (46.9)
14 (58.3)
0.63
0.36
16 (32.7)
11 (45.8)
0.57
0.27
6 (12.2)
2 (8.3)
0.72
0.73
Flank pain Appetite loss
3 (6.1) 6 (12.2)
2 (8.3) 4 (16.7)
1.53 0.7
0.62 0.61
General fatigue
6 (12.2)
4 (16.7)
0.72
0.73
Shivering
4 (8.2)
2 (8.3)
0.7
0.61
Vomiting
3 (6.1)
2 (8.3)
0.46
0.35
Macrohematuria
3 (6.1)
3 (12.5)
N/A
0.32
Bladder tamponade
2 (4.1)
0 (0.0)
0.98
0.98
Pain Lumbago Abdominal pain
AP acute pyelonephritis, OR odds ratio, N/A not applicable
233
Altered level of consciousness
2 (4.1)
0 (0.0)
N/A
0.32
Pollakisuria
2 (4.1)
3 (12.5)
0.3
0.18
Feeling of residual urine
2 (4.1)
3 (12.5)
0.3
0.18
Nausea
1 (2.0)
0 (0.0)
N/A
0.48
Miction pain
0 (0.0)
1 (4.2)
0
0.15
First-line and second-line antibiotics used CAZ (30.6%) was most frequently used as the first-line antibiotic in complicated AP and CZOP (25.0%) in uncomplicated AP. Carbapenems such as DRPM and MEPM were less frequently used as first-line antibiotics in both complicated AP (6.2%) and uncomplicated AP (4.2%) (Table 4). On the other hand, carbapenems were used in 5 out of 11 cases as the second-line treatment for complicated AP and both cases of uncomplicated AP. The first-
line antibiotic therapy failed in 11 out of the 49 cases (22.5%) of complicated AP and 2 out of the 24 cases (8.3%) of uncomplicated AP (Table 5). Causative bacteria isolated from urine culture Regarding the causative bacteria isolated from urine, 9 out of the 28 strains (32.1%) isolated in complicated AP cases and 9 out of the 19 strains (47.4%) isolated in uncomplicated AP cases were E. coli. There were 3 strains of
123
234
J Infect Chemother (2011) 17:231–237
Table 3 Risk factors in complicated and uncomplicated UTI cases
Urine culture was performed in 32 cases in complicated AP and 18 cases in uncomplicated AP; blood culture was performed in 19 cases in complicated AP and 11 cases in uncomplicated AP AP acute pyelonephritis, OR odds ratio The p value (and OR) with a significant difference in statistical analysis was expressed as bold values
Risk factor
Complicated AP (n = 49) Underlying urinary tract disease
49 (100.0)
Indwelling catheter
19 (38.8)
Cephems CAZ CZOP CTM
OR
38 (77.6)
23 (95.8)
Previous use of antibiotics
17 (34.7)
1 (4.2)
0.15 0.05 12.22 0.01
Patients with DM
1 (2.0)
2 (8.3)
0.23 0.2
Septic shock
4 (8.2)
1 (4.2)
2.04 0.53
Positive urine culture Positive blood culture
20/32 (62.5) 4/19 (21.1)
14/18 (77.8) 5/11 (45.5)
0.48 0.27 0.32 0.16
Median (range) Age (years old)
70 (20–89)
71 (23–77)
Body temperature on admission (°C)
38.6 (37.0–41.8)
38.5 (37.1–40.7)
0.28
Body temperature after treatment (°C)
36.5 (5.6%)
36.5 (35.7–36.9)
0.41
Period of treatment (days)
4 (1–9)
5 (2–9)
0.01
0.04
Peripheral WBC before treatment (/mm3) 11200 (2100–30600) 11400 (3400–28400)
0.38
Serum CRP before treatment (mg/dl)
0.58
10.7 (0–34.8)
10.8 (0.2–28.5)
Table 5 First- and second-line antibiotics used in the 13 cases that required a change in antibiotic
No. (%)
Case no. First-line antibiotics Days Second-line antibiotics Days
Complicated AP (n = 49)
Uncomplicated AP (n = 24)
40 (81.6)
17 (70.8)
1
CTM
2
CZOP
5 (20.8)
2
CTM
1
TAZ/PIPC
6
6 (25.0) 4 (16.6)
3
CTM
6
MEPM
5
4
CAZ
1
CZOP
5
CAZ
5
MEPM
2 7 5
15 (30.6) 3 (6.1) 10 (20.4)
Complicated AP (n = 11)
CTRX
9 (18.3)
1 (4.2)
5
FMOX
3 (4.1)
1 (4.2)
SBT/CPZ
1 (2.1)
0 (0)
6 7
CAZ CTRX
3 4
IPM DRPM
6 (12.2)
5 (20.8)
8
TAZ/PIPC
7
MEPM
TAZ/PIPC
4
CAZ
Penicillins
p value
Uncomplicated AP (n = 24)
Gender (female)
Table 4 First-line antibiotics used Antibiotics
No. (%)
7
7
TAZ/PIPC
6 (12.2)
3 (12.4)
9
SBT/ABPC
0 (0)
1 (4.2)
10
DRPM
1
TAZ/PIPC
3
0 (0)
1 (4.2)
11
DRPM
5
VCM
4
3 (6.2)
1 (4.2)
2 (4.1)
1 (4.2)
1
CAZ
2
MEPM
5
1 (2.1)
0 (0)
2
PIPC
3
MEPM
7
0 (0)
1 (4.2)
0 (0)
1 (4.2)
49 (100)
24 (100)
PIPC Carbapenems DRPM MEPM Fluoroquinolones CPFX Total
AP acute pyelonephritis
Citrobacter freundii and 2 strains of methicillin-resistant Staphlococcus aureus (MRSA) in complicated AP cases; however, there was no significant difference in causative bacteria between complicated and uncomplicated AP cases (Table 6). E. coli showed no significant difference in the rate of antibiotic resistance between complicated and uncomplicated AP cases (Table 7).
123
10
Uncomplicated AP (n = 2)
AP acute pyelonephritis
Correlation of initial antibiotic treatment failure with risk factors and with each antibiotic Initial antibiotic treatment failure was significantly related to CRP [odds ratio (OR) = 1.11, p = 0.02] in complicated AP and to positive blood culture [OR = not applicable (N/A), p \ 0.01] in uncomplicated AP (Table 8). In addition, there was no significant correlation of initial treatment failure with a particular antibiotic.
J Infect Chemother (2011) 17:231–237
235
Table 6 Causative bacteria isolated from urine cultures Bacteria
No. (%)
Table 7 Antibiotic resistance of E. coli isolated from urine cultures in complicated and uncomplicated AP cases Antibiotics
Resistant strains; no. (%)
Complicated AP (n = 28a)
Uncomplicated AP (n = 19a)
Escherichia coli
9 (32.1)
9 (47.4)
Citrobacter freundii
3 (10.7)
0 (0.0)
ABPC
3 (33.3) 3 (33.3)
5 (50.0)
CEZ
1 (11.1)
1 (10.0)
Complicated AP (n = 9)
Uncomplicated AP (n = 10) 5 (50.0)
Staphlococcus aureus MRSA
2 (7.1) 2 (7.1)
1 (5.3) 0 (0.0)
PIPC
Klebsiella pneumoniae
1 (3.6)
2 (10.5)
CTM
0 (0)
0 (0)
0 (0)
0 (0)
Proteus mirabilis
1 (3.6)
1 (3.6)
CAZ
Enterococcus faecalis
1 (3.6)
0 (0.0)
CPR
0 (0)
0 (0)
0 (0.0)
CMZ
0 (0)
0 (0)
0 (0.0)
CCL
0 (0)
0 (0)
a
Enterobacter aerogenes Acinetobacter baumanii
1 (3.6) 1 (3.6)
Candida species
1 (3.6)
0 (0.0)
CPDX
0/5 (0)
0 (0)
Gram-positive rods
1 (3.6)
0 (0.0)
CFPN
0/5a (0)
0/6b (0)
Prevotella melaninogenica
1 (3.6)
0 (0.0)
FMOX
0 (0)
0 (0)
4 (21.1)
IPM
0 (0)
0 (0)
AZT
0 (0)
0 (0)
GM AMK
2 (22.2) 0 (0)
0 (0) 0 (0)
MINO
0 (0)
2 (20)
FOM
0 (0)
0 (0)
LVFX
3 (33.3)
3 (30)
ST
0 (0)
3 (30)
C/A
a
0/5 (0)
0/6b (0)
S/C
0 (0)
0 (0)
No growth
9 (32.1)
Twenty-eight strains were isolated from urine cultures in complicated AP cases and 19 strains were isolated in uncomplicated AP cases AP acute pyelonephritis, MRSA methicillin-resistant Staphlococcus aureus
Correlation of the duration of the given initial antibiotic with risk factors and with the each antibiotic We revealed significant correlations between longer duration of antibiotics and DM (OR = 3.55, p = 0.01) and positive blood culture (OR = N/A, p \ 0.01) in cases of uncomplicated AP by logistic regression analyses. We also demonstrated that age (OR = 0.05, p = 0.02) was to related to shorter duration of the given initial antibiotics in complicated AP cases. In addition, TAZ/PIPC (OR = 3.46, p = 0.01) was significantly correlated with longer duration of antibiotic treatment in complicated AP (Table 9).
Discussion UTIs are major infectious diseases, often requiring intravenous antibiotics and hospitalization for treatment. Previous reports showed that 29.5% of the patients who visited the emergency room had UTIs, and among them, the most frequent diagnosis was acute uncomplicated UTI (24.2%) [11]. UTIs are common and potentially severe infections which emergency doctors must manage effectively; however, the best therapy remains controversial, as which intravenous antibiotics should be selected as the first-line treatment has not been established [2]. This study investigated the efficacy and risk factors of antibiotic treatments in febrile AP cases.
AP acute pyelonephritis a
Susceptibility to CPDX, CFPN, and C/A was tested for 5 strains from complicated AP cases
b
Susceptibility to CFPN and C/A was tested for 6 strains from uncomplicated AP cases
Risk factors such as recent hospitalization, previous use of antibiotics, and immunosuppression (for example due to steroid use) were reported to relate to resistant pathogens, indicating that clinical factors available at presentation can be used for the risk stratification of patients with AP [12]. Our study revealed that positive blood culture related to initial treatment failure and the duration of the given initial antibiotics in uncomplicated AP cases. Higher CRP led to initial treatment failure, and TAZ/PIPC was related to longer duration of initial antibiotic treatment in complicated AP cases, suggesting that DM is the major risk factor and penicillins may not suitable for the treatment of febrile AP. Previous reports demonstrated that AP is often treated with i.v. fluoroquinolones or extended-spectrum cephalosporins and that emergent admission is often required in severe cases [6]. Cephems play a valuable role in initial treatment of patients with acute urogenital infections [5, 7– 9]. In our study, cephems such as CAZ and CZOP were the most frequently used antibiotics, and they showed tolerable
123
236
J Infect Chemother (2011) 17:231–237
Table 8 Correlation of risk factors with initial treatment failure, as assessed by univariate analysis
Complicated AP
Uncomplicated AP
OR (95% CI)
p value
OR (95% CI)
p value
Age
1.05 (1.00–1.11)
0.07
1.92 (0.68–5.40)
0.22
Gender (female)
1.40 (0.25–7.68)
0.7
N/A
N/A
Patients with DM
N/A
N/A
21 (0.69–642.98)
0.08
Risk factor
AP acute pyelonephritis, OR odds ratio, CI confidence interval, N/A not applicable, DM diabetes mellitus, CRP C-reactive protein The p value (and OR) with a significant difference in statistical analysis was expressed as bold values
Indwelling catheter
1.43 (0.37–5.55)
0.61
N/A
N/A
Previous use of antibiotics
2.95 (0.74–11.69)
0.13
N/A
N/A
Septic shock
1.17 (0.11–12.48)
0.9
N/A
N/A
Positive blood culture
0.67 (0.54–8.16)
0.75
N/A
<0.01
Peripheral WBC before treatment
1.00 (1.00–1.00)
0.44
1.00 (1.00–1.00)
0.87
Serum CRP before treatment
1.11 (1.02–1.21)
0.02
0.97 (0.80–1.19)
0.81
Laboratory data
Table 9 Correlation of risk factors with duration of initial antibiotic by univariate analyses
Complicated AP
Uncomplicated AP
b coeff. (95% CI)
p value
b coeff. (95% CI)
p value
0.46
Risk factor
AP acute pyelonephritis, b coeff. b coefficient, CI confidence interval, DM diabetes mellitus, N/A not applicable, CRP C-reactive protein The p value (and b coefficient), with a significant difference in statistical analysis was expressed as bold values
Age
0.05 (0.01–0.09)
0.02
0.01 (-0.02 to 0.05)
Gender (female)
0.09 (-1.77 to 1.95)
0.92
1.30 (-2.60 to 5.21)
0.50
Patients with DM
N/A
N/A
3.55 (1.16–5.93)
0.01
Indwelling catheter
0.10 (-1.69 to 1.50)
0.90
N/A
N/A
Previous use of antibiotics
0.76 (-0.85 to 2.38)
0.35
0.26 (-4.21 to 3.69)
0.89
Septic shock
1.04 (-1.78 to 3.86)
0.46
0.78 (-3.15 to 4.71)
0.46
Positive blood culture
0.35 (-2.53 to 3.23)
0.80
2.53 (0.70–4.37)
0.01
0.01 (-0.01 to 0.01) 0.08 (-0.01 to 0.17)
0.78 0.09
0.01 (–0.01 to 0.01) 0.08 (-0.02 to 0.17)
0.63 0.11
TAZ/PIPC
3.46 (1.32–5.60)
0.01
1.62 (0.66–3.90)
0.63
PIPC
N/A
N/A
N/A
N/A
Laboratory data Peripheral WBC before treatment Serum CRP before treatment Antibiotic
efficacy when used as the first-line antibiotics in empirical treatments, as noted in previous reports. Previous reports demonstrated that the causative bacteria in serious UTI cases were E. coli \60%, other Enterobacteriaceae such as Klebsiella pneumoniae, C. freundii, and E. faecalis, Gram-negative bacilli such as Pseudomonas aeruginosa, and Gram-positive bacteria such as Staphylococcus aureus, and they showed the efficacy of advancedgeneration cephalosporins and fluoroquinolones [3]. The most common bacteria other than E. coli in this study were S. aureus, C. freundii, and K. pneumonia, as also noted in previous reports. Therefore, we considered there was good indication for the use of cephems—especially advancedgeneration cephalosporins—as the first-line antibiotics for these isolates of bacteria, as shown in previous reports [6, 8]. Regarding bacterial antibiotic resistance, previous reports revealed that female AP caused by ESBL-producing
123
bacteria could not be treated by CTRX [13]. Among the uropathogenic strains isolated from UTI patients in Japan, the proportion of E. coli strains that were less sensitive to fluoroquinolone was reported to be over 20%, and isolates of ESBL-producing E. coli were found to be increasing [14, 15]. In this study, E. coli isolated from urine cultures showed[30% resistance to ABPC, PIPC and LVFX in both complicated and uncomplicated AP cases, and there was no ESBL-producing E. coli. Multidrug-resistant P. aeruginosa and MRSA were also difficult to treat, and the rates of susceptibility of Enterococcus faecium to beta-lactams and fluoroquinolones were 0 and 6%, respectively [15]. Among our cases, there were 2 cases of MRSA in complicated AP, but no multidrug-resistant P. aeruginosa that was difficult to treat. In conclusion, we have demonstrated that CRP is related to initial treatment failure and TAZ/PIPC is related to
J Infect Chemother (2011) 17:231–237
longer duration of antibiotic treatment in complicated AP cases, while positive blood culture was significantly correlated with initial antibiotic treatment failure and longer duration of antibiotic treatment in uncomplicated AP cases. It is necessary to assess risk factors and assign the appropriate treatment strategy immediately for the patients with febrile AP. We plan to conduct a prospective randomized study of antibiotic efficacy to establish an effective standard of treatment for AP. Acknowledgments We thank Drs. Masaru Lee and Yoshihisa Kinoshita for taking care of patients, Dr. Daisuke Sugiyama for statistical analyses, and Gary Mawyer, M.F.A., for editing the English of this article.
References 1. Barbara PG, Elena C, Francesca R, Valerio V, Mario SR, Mauro F. The clinical spectrum of acute ‘‘uncomplicated’’ pyelonephritis from an emergency medicine perspective. Int J Antimicrob Agents. 2008;31:46–53. 2. Miller O 2nd, Hemphill RR. Urinary tract infection and pyelonephritis. Emerg Med Clin N Am. 2001;19:655–74. 3. Carson C, Naber KG. Role of fluoroquinolones in the treatment of serious bacterial urinary tract infections. Drugs. 2004;64:1359–73. 4. Kyriakidou KG, Rafailidis P, Matthaiou DK, Athanasiou S, Falagas ME. Short- versus long-course antibiotic therapy for acute pyelonephritis in adolescents and adults: a meta-analysis of randomized controlled trials. Clin Ther. 2008;30:1859–68. 5. Rollino C. Acute pyelonephritis in adults. G Ital Nefrol. 2007; 24:121–31.
237 6. Molina Caban˜ero JC, Ochoa Sangrador C. Grupo Investigador del Proyecto Hospitalization criteria in urinary tract infections. An Pediatr (Barc). 2007;67:469–77. 7. Bruye`re F, Cariou G, Boiteux JP, Hoznek A, Mignard JP, Escaravage L, et al. Acute prostatitis. Prog Urol. 2008;18:19–23. 8. Arakawa S, Kamidono S. Clinical evaluation of drug efficacy in UTI in men: the significance of bacterial elimination on day one. Infection. 1994;22:23–6. 9. Bruye`re F, Cariou G, Boiteux JP, Hoznek A, Mignard JP, Escaravage L, et al. Acute pyelonephritis. Prog Urol. 2008;18:14–8. 10. Arslan H, Azap OK, Ergonul O, Timurkaynak F, Urinary Tract Infection Study Group. Risk factors for ciprofloxacin resistance among Escherichia coli strains isolated from communityacquired urinary tract infections in Turkey. J Antimicrob Chemother. 2005;56:914–8. 11. den Engelsen C, van der Werf C, Matute AJ, Delgado E, Schurink CA, Hoepelman AI. Infectious diseases and the use of antibiotics in outpatients at the emergency department of the University Hospital of Leon, Nicaragua. Int J Infect Dis. 2009;13:349–54. 12. Efstathiou SP, Pefanis AV, Tsioulos DI, Zacharos ID, Tsiakou AG, Mitromaras AG, et al. Acute pyelonephritis in adults: prediction of mortality and failure of treatment. Arch Intern Med. 2003;163:1206–12. 13. Suankratay C, Jutivorakool K, Jirajariyavej S. A prospective study ceftriaxone treatment in acute pyelonephritis caused by extended-spectrum beta-lactamase-producing bacteria. J Med Assoc Thail. 2008;91:1172–81. 14. Shigemura K, Arakawa S, Miura T, Nakano Y, Tanaka K, Fujisawa M. Significance of fluoroquinolone-resistant Escherichia coli in urinary tract infections. Jpn J Infect Dis. 2008;61:226–8. 15. Ishikawa K, Hayakawa S, Miyakawa S, Kusaka M, Shiroki R, Hoshinaga K. Survey of the susceptibility of urinary isolates to antibacterial agents in 2003. J Infect Chemother. 2003;11:44–7.
123