1. Introduction
Acute cholangitis (AC), a bacterial infection characterized by inflammation of the bile ducts, is a critical gastrointestinal condition with a noteworthy morbidity and mortality rate [
1]. The mortality rate for acute cholangitis can vary based on a number of factors, including the severity of the condition, the timeliness of treatment, and underlying patient health factors, ranging between 1 and 10% with prompt treatment [
2,
3]. The higher severity end can typically reach mortality rates as high as 50%, being associated with multiple comorbidities or failure to perform biliary drainage [
4,
5,
6,
7]. Globally, acute cholangitis represents a major challenge, primarily precipitated by biliary obstruction from gallstones or malignancy. This condition necessitates a sizable number of emergency surgical interventions [
8,
9]. Distinct from primary biliary cholangitis, the global incidence of AC stands at approximately 8–12 cases per 100,000 individuals annually [
10,
11,
12]. The wide range of AC patients encountered in healthcare systems underscores the importance of discerning the variables affecting disease progression and outcomes. With increasing cases, healthcare settings witness a diverse array of patients, which makes it essential to understand the differentiating factors that influence the disease progression and outcomes, namely, the underlying nature of the condition being malignant or benign, the intervention techniques employed, and existing comorbidities [
13,
14,
15,
16,
17].
Increasing antibiotic resistance and the emergence of multidrug resistance patterns in bacterial pathogens that are responsible for different infections, including acute cholangitis, have exponentially escalated the complexity of managing this condition [
18,
19]. In the last decade, the alarming rise in antibiotic resistance has been corroborated by numerous studies, highlighting increased patient morbidity, prolonged hospital stays, and heightened financial strains on the healthcare system [
20,
21,
22]. This crisis, further fueled by the inappropriate use of antimicrobial agents and natural selection pressure, calls for an in-depth investigation to streamline treatment protocols, particularly focusing on groups with varying underlying pathology and treatment interventions.
Stent placement and anterior sphincterotomy, routinely employed in the management of acute cholangitis, have demonstrated varying degrees of success and complications [
23,
24]. The utilization of these interventions, frequently determined by the malignancy status as a source of obstruction, has been noted to potentially influence the patient’s susceptibility to antibiotic resistance [
25]. Consequently, analyzing the role these interventions play in dictating the cultures and antibiotic resistance patterns becomes pivotal.
Moreover, contemporary literature indicates a possible correlation between the nature of biliary obstruction (malignant vs. benign) and the bacterial flora present, which could subsequently influence antibiotic resistance patterns [
26,
27]. While malignancies involve complex pathological processes and potential immunosuppression, benign obstructions might present different bacterial profiles. By understanding these distinctions, we aim to provide a more comprehensive perspective on antibiotic resistance in acute cholangitis patients, irrespective of the underlying etiology [
28].
In light of escalating antibiotic resistance concerns, there is a pressing need for a study offering a current perspective and establishing a foundation for ensuing research. Hence, this study endeavors to discern potential variances in multidrug resistance patterns among patients subjected to diverse intervention techniques. The study’s primary objectives encompass detailing culture and resistance outcomes among AC patients post sphincterotomy or stent placement; and meticulously examining antibiotic resistance patterns, spotlighting multidrug resistance variances as influenced by the stent material.
3. Results
From our patient cohort of 488 individuals diagnosed with acute cholangitis, we observed a mean age of 69.3 years (SD ± 10.9). Males constituted 46.5% (n = 227) of the cohort. Broken down by age categories, the majority of our patients were older adults (>65 years), comprising 67.4% (n = 329) of the total. Middle-aged adults (40–65 years) represented 28.3% (n = 138), while young adults (18–39 years) were the least prevalent, making up only 4.3% (n = 21). In terms of clinical presentation, the most common symptom was jaundice, experienced by 89.5% (n = 437) of the patients. Abdominal pain was also a frequent complaint, reported by 73.3% (n = 358). Fever and chills were noted in 31.5% (n = 154) of the individuals.
Concerning the timing of ERCP intervention, 65.1% (n = 318) underwent the procedure emergently within 48 h, while 14.9% (n = 73) received it urgently between 48 and 72 h. A total of 19.9% (n = 97) had the procedure performed later than 72 h post presentation. The median duration of hospitalization for these patients was 8.6 days, with an interquartile range (IQR) of 5.3 days. Hospital admissions during weekends accounted for 27.0% (n = 132) of the cases. In our cohort, 20.9% (n = 102) underwent cholecystectomy. As per the Tokyo severity score, a significant majority had a Grade III severity at 83.6% (n = 408), followed by Grade I at 10.5% (n = 51), and Grade II at 5.9% (n = 29).
Regarding prior interventions, 73.8% (n = 361) were ERCP-naïve. A total of 6.7% (n = 33) had a previous sphincterotomy, while 19.3% (n = 94) had undergone both a sphincterotomy and stent placement previously. Lastly, when examining the etiology of the obstruction, the causes were nearly evenly split between malignant (50.4%, n = 246) and benign (49.6%, n = 242) factors, as detailed in
Table 1.
The laboratory data were compared across the three distinct patient groups: those who previously underwent sphincterotomy (n = 33), those with both sphincterotomy and stent (n = 94), and the ERCP-naïve group (n = 361). For white blood cell (WBC) counts, patients with both a sphincterotomy and stent exhibited the highest mean value at 12.8 thousands/mm3, followed closely by the sphincterotomy-only group at 11.5 thousands/mm3. The ERCP-naïve group had a slightly lower mean count of 10.6 thousands/mm3. The observed differences were statistically significant, with a p-value of 0.003.
Levels of C-reactive protein (CRP) were also disparate. The group with both sphincterotomy and stent manifested the highest mean CRP levels at 116.9 mg/L. This was contrasted by the ERCP-naïve patients whose mean CRP was at 91.5 mg/L, while the sphincterotomy-only group hovered in-between at 108.3 mg/L. These variations were substantiated, with a p-value of 0.014. Total bilirubin levels seemed relatively consistent across all groups, with the differences not reaching statistical significance (p-value of 0.383). Platelet (PLT) counts were lowest in the ERCP-naïve group at a mean of 241.8 thousands/mm3, compared to the sphincterotomy-only group at 272.5 thousands/mm3, and 266.1 thousands/mm3 for those with both interventions. This discrepancy was statistically significant, with a p-value of 0.041.
A pronounced difference was observed in the international normalized ratio (INR). The ERCP-naïve group had a considerably lower mean INR value of 1.3, in stark contrast to the substantially elevated values of 3.0 and 3.3 in the sphincterotomy-only and sphincterotomy with stent groups, respectively. The divergence here was starkly significant, with a p-value of <0.001.
Differences in bacterial presence in bile cultures across groups were discernible. The ERCP-naïve patients had a significantly higher incidence of sterile bile (38.2%) compared to the other groups, while those with both a sphincterotomy and stent exhibited a pronounced presence of two bacteria (50.0%). Blood cultures displayed a similar trend; a significant proportion of ERCP-naïve patients had sterile blood results (74.8%), while the incidence of two bacteria was more pronounced in the group with both interventions at 11.3%, a significant difference with a
p-value of 0.001, as presented in
Table 2.
Table 3 describes the interventions performed based on the etiology of obstruction. Regarding benign etiologies in a total of 242 patients, choledocholithiasis was the predominant cause and showed a statistically significant difference across the groups. A total of 66.7% of the patients with a previous sphincterotomy, 5.0% of those with both a sphincterotomy and stent placement, and 29.8% of the ERCP-naïve patients had choledocholithiasis, with a
p-value of <0.001. This indicates that the presence of choledocholithiasis varied significantly between these intervention groups. Other benign etiologies such as vaterian ampulloma, benign choledochal stenosis, Mirizzi syndrome, and liver abscess showed no significant differences across the groups, with
p-values of 0.491, 0.342, 0.239, and 0.411, respectively.
For malignant etiologies, which included 246 patients, there were statistically significant variations in the prevalence of pancreatic cancer, cholangiocarcinoma, and malignant vaterian ampulloma across the groups. Pancreatic cancer was present in 15.2% of patients with a previous sphincterotomy, 41.5% of those with a sphincterotomy and stent, and 30.7% of ERCP-naïve patients, with a significant p-value of 0.002. Cholangiocarcinoma was diagnosed in 15.2% of the previous sphincterotomy group, 9.3% of the sphincterotomy and stent group, and 6.1% of the ERCP-naïve group, with a highly significant p-value of <0.001. Malignant vaterian ampulloma was noted in 0.0% of patients with only a previous sphincterotomy, 16.0% of those with both interventions, and 4.2% of ERCP-naïve patients, with another significant p-value of <0.001. Other malignancies like malignant extrinsic compression and gallbladder cancer did not differ significantly across the groups, with p-values of 0.501 and 0.588, respectively.
Across the 488 bile samples analyzed, the study discerned a range of multidrug-resistant microorganisms, with a total incidence of multidrug resistance (MDR) of 19.9%. ESBL was found in 12.3% of the total samples, illustrating notable variations among the groups (p-value < 0.001). Of significant mention, the group that underwent both sphincterotomy and stent placement displayed a pronouncedly elevated prevalence of ESBL at 28.7%. In contrast, the ERCP-naïve group registered an incidence of 8.9%, with the lowest prevalence of 3.0% observed in the sphincterotomy-only cohort.
With MRSA, its overall prevalence was relatively low at 0.4%. Both the combined sphincterotomy and stent group and the ERCP-naïve group showed minimal detection of MRSA at 1.1% and 0.3%, respectively. Remarkably, the sole sphincterotomy group recorded no instances of MRSA. VRE was discerned in 2.7% of the bile samples. The group with both sphincterotomy and stent placement revealed a higher incidence of VRE at 8.5%, in comparison to the ERCP-naïve group and sphincterotomy-only group at 1.1% and 3.0%, respectively, with a significant p-value of <0.001.
CRE was identified in 4.5% of the total samples. The group with both sphincterotomy and stent insertion recorded a relatively higher incidence at 11.7%, contrasting with the 2.8% in the ERCP-naïve group and 3.0% in the sphincterotomy-only group. The
p-value was notably significant at <0.001. In the context of the overall MDR organisms, there was a marked difference between the groups, manifested by a
p-value of <0.001. The group with both sphincterotomy and stent placement demonstrated a considerably elevated MDR prevalence at 50.0%. This was distinctively higher than the 13.0% in the ERCP-naïve cohort and the 9.1% in the group with only a sphincterotomy, as described in
Table 4.
Table 5 delineates a comprehensive assessment of microbial characteristics identified in bile samples that were stratified based on distinct therapeutic strategies encompassing previous sphincterotomy, previous sphincterotomy accompanied by prior stent placement, and ERCP-naïve patients. The analysis revealed significant discrepancies in the distribution of specific microorganisms across various groups. Predominantly,
Escherichia coli dominated with a prevalence of 30.7% across all samples. A conspicuous difference was noted across the cohorts, with the highest occurrence in the ERCP-naïve group at 36.8%, closely followed by the sphincterotomy-only group at 30.3%. Conversely, the sphincterotomy and stent group demonstrated a starkly lower rate of 7.4% (
p-value < 0.001).
Klebsiella spp. was detected in 18.4% of samples, most prevalent in the combined sphincterotomy and stent group at 37.2%, while the sphincterotomy-only and ERCP-naïve groups revealed incidences of 24.2% and 13.0%, respectively (
p-value < 0.001). The less common
Pseudomonas spp. (6.7% overall) exhibited a non-significant variance among the cohorts with a
p-value of 0.080, though the combined intervention group had a slightly elevated rate at 10.6%. Other Gram-negative strains such as
Enterobacter spp.,
Acinetobacter spp.,
and Citrobacter spp. had their own unique distribution patterns across the patient cohorts, with significant variances observed in
Acinetobacter spp. and
Citrobacter spp. at
p-values of 0.006 and <0.001, respectively.
Regarding the Gram-positive bacteria, Enterococcus spp. was observed in 23.6% of all bile samples, markedly concentrated in the sphincterotomy and stent group at a high rate of 43.6%. In comparison, the ERCP-naïve and sphincterotomy-only groups demonstrated incidences of 18.0% and 27.3%, respectively (p-value < 0.001). Meanwhile, Streptococcus spp. and Staphylococcus spp. revealed overall prevalences of 3.9% and 4.1%, respectively. Notably, their distributions across the stratified patient groups were not statistically significant.
Table 6 provides an in-depth analysis into antibiotic resistance patterns among bile cultures differentiated by their respective interventions. In the context of ampicillin/sulbactam, resistance was evident in 33.3% of the samples, sourced from a subset of 66 cases. The previous sphincterotomy group exhibited the lowest resistance rate at 6.1%. In contrast, a slightly elevated resistance was discerned in the group treated with both sphincterotomy and stent placement, at 9.6%. The
p-value, at 0.036, suggests a statistically significant difference in resistance across these groups.
Piperacillin/tazobactam and fluoroquinolone (ciprofloxacin/levofloxacin) resistance patterns illustrated pronounced disparities with p-values of 0.002 and 0.177, respectively. Specifically, the combination of previous sphincterotomy and stent placement reflected a heightened resistance to fluoroquinolones at 19.1%. This suggests that this group may be particularly prone to resist this antibiotic class. Moreover, resistance to penems, encompassing meropenem and imipenem, spanned 17.7% of the samples from 339 cases. The ERCP-naïve group, or the non-intervention group, revealed a resistance rate of 13.0%, in sharp contrast to an absolute non-resistance in the sphincterotomy-only group, implying the potential potency of penems in treating acute cholangitis in patients managed by sphincterotomy.
Delving into resistance patterns concerning different cephalosporin generations, a multifaceted picture emerges. Although resistance to the 2nd and 4th generation cephalosporins did not reveal significant inter-group variations, with p-values of 0.441 and 0.246, respectively, the 3rd generation cephalosporin resistance showcased statistically significant differences, with a p-value of 0.014. Notably, the sphincterotomy group encountered a heightened resistance rate of 9.1% for these cephalosporins, raising concerns regarding their efficacy for this cohort.
Lastly, the evaluation highlighted pronounced resistance to piperacillin and ticarcillin/clavulanic acid in 55.8% and 32.2% of samples, respectively. These resistance patterns showed substantial variations, reinforced by p-values of 0.084 and 0.004. Most strikingly, the combination group of sphincterotomy and stent placement exhibited a marked piperacillin resistance rate of 30.9%.
Table 7 presents an in-depth examination of bile culture outcomes and multidrug resistance patterns between patients with previously placed metal stents and those with plastic stents. For Gram-negative bacteria,
Escherichia coli was detected in 57.6% of the metal stent group and 52.5% of the plastic stent group (
p = 0.634). Similarly,
Klebsiella spp. was found in 57.6% and 44.3% of the metal and plastic stent cohorts, respectively, with an insignificant difference.
Pseudomonas spp. demonstrated a presence of 18.2% in the metal stent group and 9.8% in the plastic stent group, with a
p-value of 0.247.
Enterobacter spp. was less prevalent in both groups, with occurrences of 3.0% and 6.6% for metal and plastic stent patients, respectively (
p = 0.467).
Acinetobacter spp. was not detected in either group.
Citrobacter spp. was found in 15.2% of the metal stent patients and 9.8% of the plastic stent group (with a
p-value of 0.444).
Regarding Gram-positive bacteria, Enterococcus spp. was observed in 36.4% of metal stent patients and 44.3% of plastic stent patients (p = 0.458). Streptococcus spp. was minimally detected, with a 3.3% occurrence in the metal stent cohort and 4.9% in the plastic stent group, with an insignificant difference in proportions. Staphylococcus spp. was involved in 6.1% of the metal stent patients and 1.6% of those with plastic stents (p-value = 0.244).
Regarding multidrug resistance, the study observed extended-spectrum beta-lactamases (ESBL) in 21.2% of metal stent patients and 9.8% of plastic stent patients. Methicillin-resistant Staphylococcus aureus (MRSA) was detected in 1.6% of the plastic stent group and was absent in the metal stent group. Vancomycin-resistant enterococci (VRE) were present in 9.1% of the metal stent group, compared to a mere 1.6% in the plastic stent cohort. Carbapenem-resistant Enterobacteriaceae (CRE) were found in 9.1% of metal stent patients and 3.2% of those with plastic stents.
Assessing the bacterial presence in bile samples, a sterile culture was identified in 9.1% of metal stent patients and 3.3% of plastic stent patients. Samples with a single bacterium were noted in 21.2% of the metal stent cohort and 32.8% of the plastic stent group. Bile samples with two bacterial species were identified in 57.6% of metal stent patients and 55.7% of the plastic stent group. Lastly, samples with three or more bacterial species were recorded in 12.1% of metal stent patients and 8.2% of plastic stent patients, without expressing statistical significance.