Abstract
Abstract
Background:
Microbiologic studies suggest that complicated (CAA) and uncomplicated (UAA) acute appendicitis are different entities. Routine peritoneal fluid cultures continue to be controversially related to a low positive rate, found mainly in UAA; to isolation of typical micro-organisms with expected susceptibilities; and to a community-acquired intra-abdominal infection. The aim of this study was to describe microbiologic isolates in CAA and UAA and the usefulness of peritoneal fluid cultures to determine the susceptibilities to our antibiotic therapy guidelines.
Methods:
This study was a retrospective review of a prospective database collected at University San Juan Hospital (Spain) between June 2014 and June 2017. Complicated acute appendicitis was defined as gangrenous or perforated, whereas UAA was defined as phegmonous or suppurative. Our antibiotic recommendations are amoxicillin-clavulanic acid and in patients with β-lactam allergies, metronidazole plus aztreonam, and an aminoglycoside (gentamicin or tobramycin). Microbiologic cultures were performed in 264 patients, 157 with a CAA and 107 with a UAA.
Results:
The positive culture rate was significantly higher in CAA (59%) than in UAA (24.3%). Gram-positive cocci (51.6% CAA; 23.1% UAA), including Streptococcus constellatus (29% CAA; 3.8% UAA), and anaerobes (67.7% CAA; 42.3% UAA) were significantly more common in CAA. The rates of antibiotic resistance were amoxicillin-clavulanic acid 14% (17.2% CAA; 3.8% UAA), gentamicin or tobramycin 8.4% (9.7% CAA; 3.8% UAA), ciprofloxacin 5.9% (6.5% CAA; 3.8% UAA), and ertapenem 10.9% (14% CAA; 0 UAA).
Conclusions:
The culture-positive rate was higher in CAA, with different isolates and susceptibilities than in UAA, identifying a higher frequency of gram-positive cocci (including S. constellatus) and anaerobes. We recommend obtaining peritoneal fluid cultures in CAA, which frequently will lead to a change in the antimicrobial drug therapy guidelines, creating specific recommendations in AA.
A
Microbiologic studies question the classical explanation of AA: An obstruction of the lumen and subsequent accumulation of secretions, rising intraluminal pressure, layer necrosis, and invasion by micro-organisms, instead assigning a central role in pathogenesis to a change in the microbiome of the appendix. Differences between complicated (CAA) and uncomplicated acute appendicitis (UAA) are apparent, suggesting they are different entities [2–4].
Routine peritoneal fluid cultures yielding a low positive rate, mainly in UAA, continue to be controversial, with isolation of typical micro-organisms with an expected resistance rate according to CA-IAI; and the results are not received in as many as 50% of patients until after hospital discharge [5–8]. The Surgical Infection Society (SIS) recommends obtaining cultures of peritoneal fluid in higher-risk patients with CA-IAI and in patients with hospital-acquired IAI to identify potential resistant or opportunistic pathogens and, if adequate resources are available, to analyze epidemiologic data. They do not, however, recommend routine peritoneal fluid cultures in lower-risk patients with CA-IAI to guide antimicrobial agent therapy [9].
The aim of this study was to describe the microbiologic isolates in CAA and UAA and to determine the utility of peritoneal fluid cultures to determine the susceptibilities in relation to our antibotic therapy guidelines.
Patients and Methods
Study design
This study was a retrospective review of a prospective database collected at San Juan University Hospital (Spain) between June 2014 and June 2017. The inclusion criterion was a diagnosis of AA with free peritoneal fluid culture. The exclusion criteria were patients aged below 7 years or weight lower than 20 kg, AA without free peritoneal fluid, and non-appendiceal inflammatory pathology and a normal appendix. The study was approved by the Ethics Committee of our institution.
Definitions
The categorization of AA was performed according to surgery reports and histopathologic examination. Complicated appendicitis was defined as gangrenous (layer necrosis) or perforated, whereas UAA was defined as phlegmonous or suppurative. In the event of any discrepancy, the surgical findings were decisive.
Antibiotic protocol
Our antibiotic protocol included amoxicillin-clavulanic acid and in patients with β-lactam allergies, metronidazole plus aztreonam, along with an aminoglycoside (gentamicin or tobramycin). Ertapenem was preferred if we suspected extended-spectrum β-lactamase–producing bacteria. If required by disease severity, imipenem, meropenem, or piperacillin-tazobactam were preferred.
Statistical analysis
Statistical analysis was performed with IBM SPSS® version 21.0 (Armonk, NY). Qualitative variables were defined by frequency and percentage and quantitative variables by the median (25th; 75th percentile) using the χ2 test for qualitative data comparison and the Mann-Whitney test for quantitative results, considering a p value <0.05 to be statistically significant.
Results
A total of 394 patients were operated on during the period of the study. We included 264 patients (157 with a CAA and 107 with an UAA) after excluding 130 because of the lack of peritoneal fluid culture. Demographic data collected included gender (98 [62.5%] males and 59 [37.6%] females in CAA, 63 [58.9%] males and 44 [41.1%] females in UAA; p = NS); age (37 years [95% confidence interval [CI] 19–58] in CAA and 25 [16–43]; p = 0.002), chronic heart disease (6 [3.8%] in CAA and 3 [2.8%] in UAA; p = NS), diabetes mellitus (8 [5.1%] in CAA and 6 [5.6%] in UAA; p = NS).
The positive microbiological culture rate was significantly higher in CAA (93; 59.0%) than in UAA (26; 24.3%) (Tables 1 and 2). The rates of antibiotic resistance are shown in Table 3. According to the most frequently isolated micro-organisms, regardless of the type of AA, the numbers (percentages) of resistant organisms were Escherichia coli (2 [2.4] to amoxicillin-clavulanic acid, 5 [6] to aminoglycosides, and 7 [8.4] to ciprofloxacin); Streptococcus constellatus (0 to amoxicillin-clavulanic acid and 2 [7.1] to aminoglycosides); Pseudomonas aeruginosa (0 to both ciprofloxacin and aminoglycosides), Bacteroides fragilis (0 to amoxicillin-clavulanic acid and 4 [30.8] to clindamycin), and Prevotella spp. (1 [2.6] to amoxicillin-clavulanic acid and 15 [39.4] to clindamycin).
CAA = complicated acute appendicitis; NS = not significant; UAA = uncomplicated acute appendicitis.
CAA = complicated acute appendicitis; NS = not significant; UAA = uncomplicated acute appendicitis.
CAA = complicated acute appendicitis; UAA = uncomplicated acute appendicitis.
Post-operative complications occurred in 61 (38.9%) of CAA and 12 (11.2%) of UAA; p < 0.001. The number of surgical site infections was 25 (16%) in CAA and 5 (4.7%) in UAA. The number of intra-abdominal abscesses was 22 (14%) in CAA and 3 (2.8%) in UAA. The hospital stay was significantly longer in CAA (6 d [4–8]) than in UAA (3 [2–5]); p < 0.001.
Discussion
In the 1990s, Baron et al. performed a comparative microbiologic culture study of CAA and UAA and reported a similar frequency of aerobic bacteria but a lower frequency of Enterococcus spp., Enterobacter cloacae, and Lactobacillus spp. and a higher frequency of Pseudomonas aeruginosa, Fusobacterium spp., and Clostridium perfringens in CAA, which created a hypothesis about the role of the absence or excess of some micro-organisms in the pathogenesis of CAA [10]. Jackson et al. did a study based on the appendiceal lumen after appendectomy and found differences between normal appendix and AA, with a significant increase in bacterial genera Peptostreptococcus, Bilophila, Bulleidia, Fusobacterium, Mogibacterium, Aminobacterium, Proteus, Actinomyces, Aaerovorax, Anaerofilum, and Porphyromonas in AA. Furthermore, there were differences between CAA and UAA, with a significant increase in Bulleida, Fusibacterium, Prevotella, Prophyromonas, and Dialister in CAA [2]. In our series, there was a significant increase in gram-positive cocci including S. constellatus and anaerobic organisms as well as a non-significant increase in P. aeruginosa, Streptococcus spp., Bacteroides fragilis and non-fragilis, and Prevotella spp. in CAA, which is the same as in the two previous studies and in line with their conclusions about the differences between CAA and UAA, in that they now are considered different entities.
Escherichia coli was the most commonly isolated bacterium (60%–80%) and had 80%–85% susceptibility to amoxicillin-clavulanic acid, 80% to the aminoglycosides, and 70%–85% to ciprofloxacin, which is the main reason this drug has been excluded from most guidelines for the empiric treatment of intra-abdominal infection (IAI) unless local cultures show otherwise [11–14]. Authors believe the high susceptibility to ciprofloxacin is seen mostly in young patients with AA without co-morbidities, and specific guidelines have been developed that are different from most of those for IAI, which include all kinds of patients and pathologies.
Pseudomonas aeruginosa is a frequent isolate in different series (15%–35%), but was slightly lower in ours (10.9%), and all of these isolates are related to CAA. This should be kept in mind when we establish empiric antibiotic guidelines because P. aeruginosa is inherently resistant to amoxicillin-clavulanic acid and ertapenem. Obinwa et al. recommend the addition of an antibiotic to treat Pseudomonas in appendiceal peritonitis when the locally recommended empiric antibiotic does not have any anti-pseudomonal activity [11,14,15].
Anaerobic bacteria were isolated frequently in our series in contrast to the results reported by other authors (10%–15%). This could be because either anaerobic cultures were not performed according to hospital protocol in those studies if the local antibiotic guidelines show a high susceptibility rate or there are failures in obtaining peritoneal fluid, providing appropriate transport methods, etc. Empiric antibiotic guidelines recommend the use of amoxicillin-clavulanic acid, piperacillin-tazobactam, carbapenems, metronidazole plus aztreonam, cephalosporin, or aminoglycosides, whereas clindamycin is not recommended because of the low susceptibility rate, which also was the case in our study [9,11,16].
The SIS recommends obtaining cultures of peritoneal fluid in higher-risk patients (age ≥70 years; malignant disease; significant cardiovascular, hepatic, or renal disease; hypoalbuminemia; generalized peritonitis; delayed initial source control; inability to achieve adequate source control; or suspected infection with resistant pathogens) with CA-IAI and in patients with hospital-acquired IAI to identify any resistant or opportunistic pathogens; and, if adequate resources are available, to analyze epidemiologic data. However, the SIS does not recommend routine peritoneal fluid culture in lower-risk patients with CA-IAI to guide antibiotic therapy [9]. Earlier authors recommend obtaining cultures according to our results in CAA because of different isolates and susceptibilities in comparison with UAA.
For initial empiric antibiotic therapy, American guidelines recommend drugs with susceptibility rates higher than 90% to treat the expected micro-organisms only. The SIS recommends treatment of CAA with the same antibiotics recommended for other lower-risk patients with CA-IAI unless they meet the criteria identifying them as higher risk, in which case, the preferred therapy would be ertapenem, moxifloxacin (for patients with significant reactions to β-lactam drugs), or combination regimens such as cefotaxime, ceftriaxone, or ciprofloxacin plus metronidazole [9]. According to our results, we do not recommend either amoxicillin-clavulanic acid or ertapenem in CAA, mainly because of our frequent isolation of P. aeruginosa. A specific guideline for empiric antibiotic therapy in AA at our institution was used in suspected UAA, amoxicillin-clavulanic acid and in patients with β-lactam allergy, metronidazole plus aztreonam plus an aminoglycoside (gentamicin or tobramycin) or ciprofloxacin; whereas in suspected CAA, metronidazole plus aztreonam, an aminoglycoside, or ciprofloxacin were preferred, avoiding amoxicillin-clavulanic acid and ertapenem.
The main limitation of our study is the lack of data from patients younger than 7 years or weight lower than 20 kg. They are referred to the Pediatric Surgery Department of the referral hospital.
Conclusions
The culture-positive rate was higher in CAA, with different isolates and susceptibilities, than in UAA, identifying a higher frequency of gram-positive cocci (including S. constellatus) and anaerobic bacteria. According to this finding, we recommend obtaining peritoneal fluid cultures in CAA, which can lead to a change in the antibiotic therapy guidelines, creating specific recommendations in AA.
Footnotes
Author Disclosure Statement
No competing financial interests exist for any of the authors.