INTRODUCTION
Acute appendicitis is a common disease among patients who come to the emergency room and requires surgery. The lifetime risk of appendicitis is 8.6% in males and 6.7% in females [
1]. The standard treatment for acute appendicitis is appendectomy with appropriate antibiotic treatment, and the choice of appropriate antibiotics is especially important in treatment of complicated appendicitis. Antimicrobial treatment involves a delicate balance of optimizing empirical therapy, which has been shown to improve clinical outcomes, while simultaneously reducing unnecessary antimicrobial use [
2]. Intraoperative culture has been commonly performed from both the appendix and peritoneal fluid or pus, when present. The spectrum of isolated organisms and their antibiotic susceptibility profiles could help physicians choose appropriate antibiotics and treat intra-abdominal infection more effectively.
Two treatment guidelines about intra-abdominal infection have been published. The first was published by the Surgical Infection Society and the Infectious Diseases Society of America (SISIDSA) to diagnose and manage intra-abdominal infections [
3]. The second was published by the World Society of Emergency Surgery (WSES) in 2017 [
4]. In addition, the WSES published their guidelines for diagnosis and management of acute appendicitis in 2016 [
5]. According to the SIS-IDSA guidelines, the antibiotics used for empirical treatment of community-acquired intraabdominal infection should be active against enteric gram-negative aerobic and facultative bacilli and enteric gram-positive streptococci. Depending on the severity of the infection, the recommended antibiotics regimens are ticarcillin-clavulanate, cefoxitin, ertapenem, moxifloxacin, or tigecycline as single-agent therapy or a combination of metronidazole with cefazolin, cefuroxime, ceftriaxone, cefotaxime, levofloxacin, or ciprofloxacin [
3]. In noncritically ill patients with community-acquired intra-abdominal infection, the WSES guidelines recommend a similar antibiotic regimen including amoxicillin/clavulanate; ceftriazone with metronidazole; cefotaxime with metronidazole; or ciprofloxacin with metronidazole, moxifloxacin, ertapenem, or tigecycline depending on risk of ESBL-producing bacteria [
4,
5].
Multidrug resistance (MDR) caused by worldwide overuse of antibiotics is an ongoing issue. Especially, organisms with resistance to broad and extended-spectrum antibiotics can lead to complications in treatment. The most relevant MDR pathogens found in acute appendicitis are Enterobacteriaceae, Staphylococcus aureus,
Klebsiella pneumoniae, Acinetobacter baumannii,
Pseudomonas aeruginosa (
P. aeruginosa), and Enterococci bacteria [
6]. Patients with MDR bacteria in acute appendicitis suffered more infectious complications and needed longer hospitalizations than patients with antibiotic susceptible bacteria [
6]. Furthermore, the prevalence of antibiotic-resistant
E. coli isolated in intra-abdominal infection is especially high in the Asia-Pacific region [
7].
Therefore, this study was designed to confirm isolated bacteria and their susceptibilities to antibiotics to treat complicated appendicitis in Korea and to reveal bacteria and their susceptibilities affecting the clinical course of patients. Furthermore, we confirmed the effectiveness of current empiric antibiotics and the usefulness of bacterial culture in patients with complicated appendicitis.
RESULTS
General patient characteristics are shown in
Table 1. In the study period, 187 patients had undergone bacterial culture after being diagnosed complicated appendicitis, and the culture results were negative in 14 patients (7.5%). The positive rate of bacterial culture in complicated appendicitis was 92.5%, and a total of 173 patients was included in this study. The mean age of the enrolled patients was 44.8±21.3 years, and mean length of hospital stay was 8.9±7.0 days. Male patients (n=110, 63.6%) were more common than female patients (n=63, 36.4%). Included in this study were 135 cases of complicated appendicitis (78.0%) without abscess and 38 cases of appendicitis (22.0%) with abscess. All operations were conducted by laparoscopic surgery. In 42 cases (18%), resection of the cecum was needed because of severe inflammation in the appendiceal stump.
The mean length of intravenous antibiotic treatment was 8.8±4.9 days. And 71.7% (124 of 173) of patients prescribed oral antibiotics when they discharged. In the result, mean length of total antibiotic treatment was mean 12.7±7.4 days.
After surgery, 30 cases of complications were observed. The most common complication was ileus (11, 5.2%). Superficial SSI and organ/space SSI were observed in 8 (4.6%) and 9 cases (5.2%), respectively. Gastroenteritis was relatively rare (n=2, 1.2%), and 11 patients required readmission for any related reasons (6.4%). Perioperative mortality was not observed within the first 30 days after surgery.
Data of cultured organisms are summarized at
Table 2. The number of cultured organisms was 231 in 173 patients. Two or more organisms were cultured from 45 patients (19.5%). Gramnegative bacteria (n=190, 82.3%) were more common than gram-positive bacteria (n=41, 17.7%). The most commonly cultured organism was
Escherichia coli (n=113, 48.9%), while the second most common organism was
P. aeruginosa (n=21, 9.1%). Next were
Streptococcus anginosus (n=11, 4.8%) and
Streptococcus constellatus (n=10, 4.3%) followed by
Klebsiella pneumonia (n=9, 3.9%) and
Bacteriodes fragilis (n=9, 3.9%). When classified by species, the most common bacterial species was also
E. coli (n=113, 48.9%).
Streptococcus spp. (n=29, 12.6%),
Pseudomonas spp. (n=23, 10.0%),
Bacteriodes spp. (n=22, 9.5%),
Klebsiella spp. (n=11, 4.8%), and
Enterococcus spp. (n=8, 3.5%) were also common species in complicated appendicitis.
Antibiotics resistance of isolated bacteria is summarized at
Table 3. There was only one case of healthcare-associated infection. All other cases were community-acquired infections. Ampicillin resistance (n=132, 57.1%) was most commonly observed. Resistance to trimethoprim/sulfamethoxazole (TMP/SMX) (n=69, 30.0%), cefotaxime (n=51, 22.1%), and cefazolin (n=47, 20.3%) were also relatively common; 21 cases (9.1%) of ESBL positive, 4 cases (1.7%) of carbapenem resistance and 1 case (0.4%) of vancomycin resistance were observed. ESBL-producing bacteria are mostly found in
E. coli (n=19, 16.8%) and
K. pneumonia (n=2, 22.2%). Vancomycin resistance was found in
Enterococcus faecium, which was the only healthcare-associated infection.
E. coli reveal a variety of antibiotics resistance. Resistance to ampicillin and TMP/SMX was reported in 77 cases (68.1%) and 47 cases (41.6%). Resistance to cefazolin (n=30, 26.5%), cefotaxime (n=25, 22.1%), ceftazidime (n=22, 19.5%), cefepime (n=21, 18.6%), and gentamycin (n=22, 19.5%) were also not rare. The positive rate of ESBL was 16.8% (19 of 113). The antibiotics with the least resistance were piperacillin/tazobactam (n=3, 2.7%) and imipenem (n=2, 1.8%).
P. aeruginosa revealed a high resistance rate to ampicillin (n=21, 100%) and cefotaxime (n=20, 95.2%). However, resistance rates to piperacillin/tazobactam (n=1, 4.8%), ceftazidime (n=0, 0%), cefepime (n=0, 0%), ciprofloxacin (n=1, 11.1%), and imipenem (n=0, 0%) were low.
Gram-positive bacteria reveal relatively low antibiotic resistance. Only 1 Streptococcus pneumoniae and 1 E. faecium revealed resistance to antibiotics.
Bacteria isolated from patients who had complications are shown in
Table 4. The results of statistical analyses for the relationships between bacterial cultures and complications are summarized in
Table 5. Detection of
Pseudomonas or ESBL positive cultures was not related to any complication. However, detection of
Pseudomonas did affect length of hospital stay (10.5 days, P=0.042). Organisms with resistance to gentamycin (P=0.026) and TMP/SMX (P=0.002) were significantly correlated with organ/space SSI. Organisms with resistance to gentamycin were related to longer length of hospital stay (11.5 days, P=0.049). No specific organism or antibiotic resistance was related to ileus or gastroenteritis. And, patients’ characteristics including sex, age, type of surgery (appendectomy or cecectomy), or ASA physical status classification were not related with any complications.
Empirical antibiotic treatment with ceftriaxone and metronidazole showed susceptibility in 174 of 231 organisms (75.3%). Conversely, 56 of 231 (24.2%) organisms showed resistance to empirical antibiotic treatment. Because duplicated resistant organisms were found in 1 patient, 55 patients were identified as cases of IEAT (55 of 173, 31.8%). These cases of IEAT revealed a significant relationship with the organ/space SSI (7 of 55, 12.7%; P=0.005). However, IEAT also did not affect the rate of superficial SSI (2 of 55, 3.6%; P=1.000), ileus (2 of 55, 3.6%; P=0.506), or gastroenteritis (0 of 55, 0%; P=1.000). The mean length of hospital stay for patients with IEAT was similar to others (9.4 days vs. 8.7 days, P=0.571).
In 56 cultured organisms that reveal resistance to empiric antibiotics, the rate of cross resistance to piperacillin/tazobactam, cefepime, ciprofloxacin, imipenem, and TMP/SMX was 7.1% (4 of 56), 42.9% (24 of 56), 25.0% (14 of 56), 7.1% (4 of 56), and 67.9% (38 of 56), respectively. And ESBL-producing bacteria were identified in 21 organisms (21 of 56, 37.5%).
In 7 cases with organ/space SSI with IEAT diagnoses, the antibiotic regimen was changed based on the culture results. These 7 cases included 4 cases of MDR-P. aeruginosa, 1 case of ESBL-producing E. coli, 1 case of MDR-E. coli, and 1 case of vancomycinresistance E. faecium. A vancomycin-resistant E. faecium was isolated from an immunosuppressed patient, previously mentioned as healthcare-associated infection, who had been diagnosed and treated for acute myeloid leukemia, and antibiotic treatment was performed with the recommendation and approval from an tients with community-acquired infection, the aninfectious disease (ID) specialist. In all the other patibiotics were changed to piperacillin/tazobactam with approval by an ID specialist.
DISCUSSION
In the treatment of complicated appendicitis, the 2 main components are source control and antibiotic therapy. Source control generally includes draining the abscess or surgical removal of the appendix. However, recently, these classical ideas have been changed. The new WSES Jerusalem guidelines suggest nonoperative treatment with antibiotics could be effective in selected patients with uncomplicated appendicitis [
5]. In the same guideline, nonoperative management is also suggested as a reasonable first treatment for appendicitis with phlegmon or abscess [
5]. In this era of changing the concept of traditional treatment, effective antimicrobial treatment is becoming more important in the treatment of appendicitis.
However, the initial choice of antibiotics is always empiric because patients need immediate treatment, while bacterial growth to obtain results of culture and antibiotic susceptibility needs about 24–72 hours. Therefore, recent bacteriologic data are needed clinically to choose the appropriate antibiotics. Moreover, emerging MDR bacteria makes selection of empiric antibiotics more difficult.
Previous studies have reported the most common isolated bacteria in appendicitis to be
E. coli.
Streptococcus spp.,
Enterococcus spp., and
P. aeruginosa are also commonly observed. However, the proportion of each bacterial type and resistance rate vary [
10-
13]. The study for Monitoring Antimicrobial Resistance Trends (SMART) trial is the only large study that includes data from 122 hospitals in 39 countries focusing on epidemiology and antibiotic susceptibility in acute appendicitis. This study revealed the highest prevalence of ESBL-producing
E. coli (16.6%) in the Asia-Pacific region. All species studied for the test were quite susceptible to carbapenems [
7].
In this study,
E. coli (n=113, 48.9%),
Streptococcus spp. (n=29, 12.6%),
Pseudomonas spp. (n=23, 10.0%),
Bacteriodes spp. (n=22, 9.5%),
Klebsiella spp. (n=11, 4.8%), and
Enterococcus spp. (n=8, 3.5%) are common. This result shows relatively lower rate of
E. coli and higher rate of
Pseudomonas spp. than previous studies [
11,
12]. The rate of ESBL-producing
E. coli was revealed as 16.8%. The antibiotic resistance rate throughout all cultured organisms to cefoxitin, cefotaxime, ceftazidime, and ciprofloxacin were 10.4%, 22.1%, 10.8%, and 12.6%, respectively.
E. coli revealed respective resistance rate to these antibiotics as 9.7%, 22.1%, 19.5%, and 21.2%. In SMART trial, resistance rate of
E. coli to cefoxitin, ceftriaxone, ceftazidime, and ciprofloxacin were 9%, 10.1%, 10.1%, and 22.0%, respectively. And ESBL-producing
E. coli was detected in 8.9% worldwide. The result of this study reveals that antibiotic resistance rate in Koreans is higher than other countries. And the rate of ESBL-producing
E. coli in this study is similar to the data of Asia-Pacific region in SMART trial (16.6%) [
7].
Pseudomonas aeruginosa revealed very high resistance to cefotaxime, 95.2%. Moreover, 230 of 231 isolated bacteria were community-acquired infections. These results show that the isolated bacteria in complicated appendicitis have a notably high rate of MDR, even in patients with community-acquired infection.
Current guidelines from SIS-IDSA and WSES recommend a several antibiotics regimen for community-acquired complicated appendicitis [
3-
5]. Among these regimens, the combination of ceftriaxone and metronidazole has been reported as appropriate in comparison with the past regimen, with a rate of IEAT reported at 4.1% (2 of 49) [
14,
15]. In this study, the rate of IEAT was 31.2% (54 of 173) using the same antibiotic regimen. This quite high rate of IEAT suggests that some countries have a higher risk of IEAT, like Korea, meet the moment to consider changing the empiric antibiotic regimen for specific high-risk patients. For this, well-designed, multicenter studies are needed.
In this study, detection of Pseudomonas and resistance to gentamycin were related with longer length of hospital stay (P=0.042 and P=0.049, respectively). Resistance to gentamycin and TMPSMX were related with a higher rate of organ/space SSI (P=0.026 and P=0.002, respectively). The mechanism of this relationship was not proven in this study; additional study is needed to prove this clearly. Length of hospital stay in IEAT patients was not statistically different from other, non-IEAT patients, most likely because there are so many other factors that can lead to longer hospital stay, such as ileus.
Recently, several studies have claimed that intraoperative culture is not needed in acute appendicitis. These studies argue that intraoperative culture did not influence clinical outcome in patients undergoing appendectomy. The reason for their argument is that the rate of positive culture is relatively low in most cases and cases needed to change antibiotics are rare because cultured organisms are mostly susceptible to empiric antibiotics [
10,
16-
19]. However, all studies questioning the use of intraperitoneal swabs were open, nonrandomized, or retrospective studies with incompletely matched control groups and nonstandardized swab collection techniques; consequently, they lacked power to inform surgical practice [
20]. Our study is also nonrandomized and retrospective; therefore, we also cannot definitively show evidence that a specific organism affected morbidity or length of hospital stay of patients. However, our study reveals that IEAT increases organ/space SSI (7 of 54). Seven patients with IEAT needed to change antibiotics according to the results of culture, and they all recovered without any additional problems.
Bacterial culture is an inexpensive, simple test that does not harm the patient. Therefore, routine intraoperative culture is necessary, at least in complicated appendicitis, although the majority of the patients may not need it.
In the present study, the second choice of antibiotics in all community-acquired infection was piperacillin/tazobactam according to the antibiotic susceptibility tests. Even in patients with ESBLproducing bacteria, if piperacillin/tazobactam was revealed as effective in vitro, it was used. Although carbapenem has been considered the drug of choice for treating ESBL-producing bacteria, some authors report piperacillin/tazobactam is a useful alternative and not inferior to carbapenem [
21-
23]. Since piperacillin/tazobactam and carbapenem reveal high susceptibilities (97.4% and 98.3%, respectively) in this study, these 2 antibiotics could be selected in suspecting resistance of empirical antibiotics, even if the result of culture is not available.
This report includes relatively the small number of patients, because the bacterial culture was not routinely performed intraoperatively. In our institution, bacterial culture was selectively performed by the discretion of each surgeon and only 1 surgeon routinely performed intraoperative bacterial culture in most cases with complicated appendicitis.
The methods of culture were different depending on the type of complicated appendicitis in this study. When suppurative fluid or abscess is present, these are collected in culture tube. In cases of gangrenous appendicitis without free perforation, swabbing of appendiceal lumen was performed for culture. Up to now, there has not been a standard method of culture for complicated appendicitis. Especially, the culture method of appendiceal swab is not standardized [
20]. Therefore, further studies are needed to develop an approved standard method of routine intraoperative culture in patients with complicated appendicitis.
The rate of IEAT was 31.8% in complicated appendicitis. IEAT did not lead to adverse clinical outcomes in the majority of patients. This means that proper source control still plays a more important role in treatment of complicated appendicitis. Nevertheless, IEAT for complicated appendicitis may lead to increased rates of organ/space SSI.
In conclusion, the rate of antibiotic resistance was high in community-acquired complicated appendicitis in Koreans. Routine intraoperative culture in patients with complicated appendicitis may provide information about an appropriate antibiotic regimen when current empiric antibiotics treatment fails.