Abstract
Introduction:
Antimicrobial resistance (AMR) is a growing threat to global public health. In hospitals, infant mortality due to bacterial sepsis is associated with AMR. The epidemiology of AMR in Bolivia (a lower-middle-income country) has not been sufficiently discussed. The aim of this study was to analyze AMR patterns over 8 years at a pediatric hospital in patients with hospital-acquired bloodstream infections.
Methods:
This is a retrospective and analytical revision of AMR in bacteria isolated from blood cultures, performed between 2012 and 2019, in a pediatric hospital in Bolivia. Data analysis was conducted with Stata v14.0, and Fisher’s exact tests were used to determine statistical significance.
Results:
Four hundred and fifty-five (7.2%) positive blood cultures were identified from 6315 blood culture reports between 2012 and 2019. Klebsiella pneumoniae was the most frequently isolated bacteria and showed a significant change in its AMR profile over the 8 years of the study. Gram-negative bacteria and Staphylococcus aureus were also frequently isolated, and all demonstrated high levels of resistance to commonly used antibiotics. Overall, most of the clinically important microorganisms had high rates of AMR.
Conclusions:
In the present study, we report that isolated bacteria showed significant resistance to multiple drugs, and most demonstrated increased resistance over time. Worryingly, K. pneumoniae showed an increasing resistance to commonly used antibiotics. Overall, despite the limitations, our study, which is one of the first of its kind in Bolivia, demonstrates the need for stricter policies of antibiotic stewardship in similar settings due to the global threat of AMR.
Keywords: Bacteria, blood culture, drug resistance, epidemiology, pediatrics, sepsis
INTRODUCTION
Antimicrobial resistance (AMR) is a growing and important threat to global health that increases morbidity and mortality and negatively impacts worldwide economy.[1,2,3] AMR, defined as the ability of a microorganism to grow in the presence of antimicrobial agents, is attributed to the improper and excessive use of antibiotics that trigger microbial evolution.[4,5,6] In upper-income countries, 30%–50% of antibiotics used within hospitals are considered unnecessary.[7] In low-to-middle-income countries (LMICs), AMR is especially problematic due to unregulated sales, inappropriate applications, and lax prescription of antibiotics.[8]
Pediatric sepsis is a major cause of childhood deaths in LMICs where mortality can be up to 40%.[9] AMR decreases therapeutic options, thus increasing the probability of treatment failure and mortality.[10] Over 70% of hospital-acquired bloodstream infections (HA-BSIs) are now resistant to one or more commonly used antibiotics,[7] and HA-BSIs are more likely to be caused by resistant organisms than community-acquired infections.[11] Accurate data describing the AMR profile and its evolution at a pediatric hospital in an LMIC could lead to more appropriate medical treatment and improved antibiotic stewardship, reducing current and future mortality and morbidity in LMICs.
Disappointingly, the overall progression of AMR in pediatric HA-BSIs is poorly reported in LMICs and has not been previously discussed in a Bolivian pediatric hospital.[12] Hospital de Niños Dr. Mario Ortiz Suárez is a tertiary care pediatric hospital for children between infancy and age 14. Located in Santa Cruz de la Sierra, one of the most populous cities in Bolivia, it is the largest pediatric hospital in the city, seeing approximately 20,000 patients annually. This study analyzes AMR and AMR changes over 8 years, divided into two periods of 4 years each, at this pediatric hospital in patients with HA-BSI.
METHODS
Study design
A retrospective, analytical chart review was carried out in a pediatric hospital in Santa Cruz de la Sierra, Bolivia. Records of blood cultures performed by the Clinical Laboratory Department for patients treated between the years 2012 and 2019 were identified and reviewed. The inclusion criteria were patients aged 14 years or younger, based on the hospital’s admission criteria, and blood cultures performed 48–72 h after hospital admission, based on the generally accepted definition for HA-BSI.[13] We did not include any blood cultures drawn within the patient’s first 48 h of hospitalization as this BSI would not be considered hospital acquired. Records reported as contaminated and records with incomplete data were excluded from the study.
Standard procedures of the hospital
This study is a chart review, and we did not have access to the patients’ clinical data beyond the blood culture results.
In general, blood culture samples were obtained by peripheral venipuncture, taking approximately 1 ml of blood, and inoculating it into neonatal blood culture bottles (Britania Lab., Caba, Argentina) with a sterile syringe. Prior to inoculation, the blood culture bottle caps are cleaned with alcohol to avoid any contamination. After inoculation, they are incubated at 37°C for 24 h. The culture bottles are observed for signs of bacterial growth, such as turbidity, hemolysis, or bubble formation. Bottles not suggestive of bacterial growth are placed back in the incubator at 37°C until the 7th day of incubation, with daily review for signs of bacterial growth. If there is no sign of bacterial growth at day 7, they are noted as negative.
Culture bottles with signs of bacterial growth are grown in blood-based agar, chocolate agar, MacConkey agar, mannitol salt agar, or Sabouraud agar, depending on the type of bacterial colonies observed. For blood agar and mannitol salt agar, the catalase, DNAse, coagulase identification test, and novobiocin test are performed. Growth on MacConkey agar is followed up with a biochemical gallery and oxidase test[14] [Supplementary Tables 1-3].
Blood cultures are reported as positive if bacterial growth is isolated and identified. The laboratory determines the antimicrobial sensitivity by the Kirby-Bauer diffusion method,[15] categorizing the results of each antibiotic as sensitive, intermediate, or resistant according to the cutoff points in force and recommended by the Clinical and Laboratory Standards Institute (CLSI) in each year.[16] Ten antibiotics are tested for Gram-positive cocci and 17 for Gram-negative bacilli.
Patients are treated according to physician discretion and with recommendations from the infectious disease team. Generally, blood cultures are drawn prior to the initiation of antibiotics when clinically possible.
Data collection
Paper charts were manually entered by two independent investigators into Microsoft Excel, which was used for data organization. 6395 blood culture records were collected and reviewed. Of those, 5860 blood cultures were negative for microbial growth and were excluded. Of the blood cultures positive for microbial growth, 4 records contained unidentified bacteria, 11 records were not confirmed in the logbooks, and 65 blood culture records isolated Candida spp., so these 80 records were excluded. Therefore, 455 records were analyzed. Subsequently, the antibiotics tested for each bacterium by the CLSI in each year were identified; antibiotics which were not confirmed, as well as those which no longer had a cutoff point, were eliminated.
Statistical analysis
The data were analyzed according to the most frequently isolated bacteria in each age group. Bacteria were grouped according to the bacterial classification based on bacteria morphology and Gram staining. General AMR was ascertained for Gram-positive cocci and representative Gram-negative bacilli, considering the laboratory’s categorical antimicrobial sensitivity results. In addition, resistance progression was determined over two time periods of 4 years each, which was an 8-year chart review. The statistical program Stata v. 14.0 (College Station, Texas) was used for this progression utilizing Fisher’s exact test.
Ethics
The project was deemed exempt from review by the Institutional Review Board at Hospital de Niños Dr. Mario Ortiz Suárez (IRB00011318) as it is a retrospective chart review with no personal identifiers.
RESULTS
Six thousand three hundred and ninety-five blood culture reports, drawn 48 h after patient admission, were identified in the hospital’s clinical laboratory between 2012 and 2019; 80 reports were eliminated as discussed in methods. Of the 6315 reports remaining, 5860 resulted in no growth and were considered negative blood cultures. Therefore, a total of 455 (7.2%) positive blood cultures with isolated bacteria were identified for analysis. Data were grouped according to patient age. There were 168 patients (36.8%) <1 month old (neonates), 188 patients (41.1%) between 1 month and 12 months old, and 101 patients (22.1%) >1 year and ≤14 years old. The 14 most common bacteria causing HA-BSI were identified.
The most frequently isolated bacteria at all ages were Klebsiella pneumoniae (20.7%). In neonates the second and third most frequent were Coagulase-negative staphylococci (19.6%) and Enterobacter spp. (15.5%), respectively. Pseudomonas spp. was the second most frequent in children aged 1 to 12 months (15%), and in children older than 1 year and ≤14 years (14%) [Table 1].
Table 1.
Bacterial agents causing nosocomial sepsis according to age (n=455)
| Bacteria | Age | Total, n (%) | ||
|---|---|---|---|---|
|
| ||||
| <1 month old, n (%) | 1–12 months old, n (%) | >12 and <180 months old, n (%) | ||
| K. pneumoniae | 37 (22.0) | 40 (21.4) | 17 (17.0) | 94 (20.7) |
| Coagulase-negative staphylococci | 33 (19.6) | 27 (14.4) | 12 (12.0) | 72 (15.8) |
| Enterobacter spp. | 26 (15.5) | 18 (9.6) | 9 (9.0) | 53 (11.7) |
| Pseudomonas spp. | 6 (3.6) | 28 (15.0) | 14 (14.0) | 48 (10.6) |
| S. epidermidis | 19 (11.3) | 20 (10.7) | 6 (6.0) | 45 (9.9) |
| Acinetobacter spp. | 16 (9.5) | 12 (6.4) | 10 (10.0) | 38 (8.4) |
| S. aureus | 8 (4.8) | 12 (6.4) | 12 (12.0) | 32 (7.0) |
| E. coli | 1 (0.6) | 11 (5.9) | 10 (10.0) | 22 (4.8) |
| Serratia spp. | 8 (4.8) | 3 (1.6) | 0 | 11 (2.4) |
| K. oxytoca | 4 (2.4) | 3 (1.6) | 2 (2.0) | 9 (2.0) |
| Burkholderia spp. | 2 (1.2) | 1 (0.5) | 6 (6.0) | 9 (2.0) |
| Enterococcus spp. | 3 (1.8) | 4 (2.1) | 2 (2.0) | 9 (2.0) |
| Streptococcus spp. | 5 (3.0) | 4 (2.1) | 0 | 9 (2.0) |
| Salmonella spp. | 0 | 4 (2.1) | 0 | 4 (0.9) |
<1 month old (n=168), 1–12 months old (n=187), >12 and <180 months old (n=100). K. oxytoca: Klebsiella oxytoca, S. aureus: Staphylococcus aureus, E. coli: Escherichia coli, S. epidermidis: Staphylococcus epidermidis, K. pneumoniae: Klebsiella pneumoniae
Staphylococcus aureus showed a statistically significant change in resistance over the 8 years to amikacin (0%–28.6%, P = 0.048) and clindamycin (8.7%–55.6%, P = 0.010). Coagulase-negative staphylococci showed reductions in antibiotic resistance over the 8 years to chloramphenicol (36.7%–9.1%; P = 0.009), gentamicin (78.6%–64.1%; P = 0.011), and trimethoprim-sulfamethoxazole (80.6%–50%; P = 0.005). Staphylococcus epidermidis showed an increase in resistance to clindamycin over the 8 years (60.7%–94.1%; P = 0.044) and was resistant to oxacillin, gentamicin, erythromycin, clindamycin, ciprofloxacin, and trimethoprim-sulfamethoxazole at rates >70% in the final 4 years [Table 2].
Table 2.
Progression of antibiotic resistance of the most frequent Gram-positive cocci in blood cultures
| Antibiotic resistance | S. aureus (n=32) | P | Coagulase-negative Staphylococcus (n=72) | P | S. epidermidis (n=45) | P | |||
|---|---|---|---|---|---|---|---|---|---|
|
|
|
|
|||||||
| 2012–2015, n (%) | 2016–2019, n (%) | 2012–2015, n (%) | 2016–2019, n (%) | 2012–2015, n (%) | 2016–2019, n (%) | ||||
| Oxacillin | 14/23 (60.9) | 8/9 (88.9) | 0.210 | 31/31 (100.0) | 37/41 (90.2) | 0.129 | 26/28 (92.9) | 16/16 (100.0) | 1.000 |
| Chloramphenicol | 3/22 (13.6) | 0/5 (0.0) | 1.000 | 11/30 (36.7) | 3/33 (9.1) | 0.009 | 9/28 (32.1) | 2/12 (16.7) | 0.451 |
| Amikacin | 0/23 (0.0) | 2/7 (28.6) | 0.048 | 10/28 (35.7) | 8/39 (20.5) | 0.410 | 13/26 (50.0) | 7/16 (43.8) | 0.846 |
| Gentamicin | 3/19 (15.9) | 2/7 (28.6) | 0.588 | 22/28 (78.6) | 25/39 (64.1) | 0.011 | 23/27 (85.2) | 13/17 (76.5) | 0.690 |
| Erythromycin | 5/22 (22.7) | 5/9 (55.6) | 0.105 | 31/31 (100.0) | 37/41 (90.2) | 0.129 | 26/28 (92.9) | 16/17 (94.1) | 1.000 |
| Clindamycin | 2/23 (8.7) | 5/9 (55.6) | 0.010 | 23/31 (74.2) | 31/41 (75.6) | 0.559 | 17/28 (60.7) | 16/17 (94.1) | 0.044 |
| Ciprofloxacin | 2/23 (8.7) | 3/9 (33.3) | 0.258 | 27/31 (87.1) | 30/40 (75.0) | 0.437 | 18/28 (64.3) | 13/17 (76.5) | 0.110 |
| Tetracycline | 2/22 (9.1) | 0/4 (0.0) | 1.000 | 2/31 (6.4) | 1/29 (3.4) | 1.000 | 3/26 (11.5) | 0/10 (0.0) | 0.545 |
| Minocycline | 0/15 (0.0) | 0/9 (0.0) | - | 0/18 (0.0) | 0/40 (0.0) | - | 0/23 (0.0) | 0/17 (0.0) | - |
| TMP-SMX | 2/23 (8.7) | 1/8 (12.5) | 1.000 | 25/31 (80.6) | 19/38 (50.0) | 0.005 | 24/28 (85.7) | 13/17 (76.5) | 0.452 |
S. aureus: 2012–2015 (n=23); 2016–2019 (n=9), Coagulase negative Staphylococcus: 2012–2015 (n=31); 2016–2019 (n=41), S. epidermidis: 2012–2015 (n=28); 2016–2019 (n=17). All values in bold have P<0.05. TMP-SMX: Trimethoprim-sulfamethoxazole, S. aureus: Staphylococcus aureus, S. epidermidis: Staphylococcus epidermidis
In Table 3, the three most frequent Gram-negative bacilli that cause HA-BSIs and their resistance patterns can be seen. K. pneumoniae showed increased resistance over the 8 years to cefepime (37.5%–82.2%; P < 0.001), gentamicin (71.7%–91.1%; P = 0.030), and trimethoprim-sulfamethoxazole (72.3%–91.3%, P = 0.030) with a reduction in resistance to nalidixic acid (42.2%–13.3%; P < 0.001). K. pneumoniae demonstrated an average antibiotic resistance >50% between 2016 and 2019 amoxicillin-clavulanic acid, cefotaxime, ceftazidime, cefepime, gentamicin, tetracycline, and trimethoprim-sulfamethoxazole. Enterobacter spp. showed decreased resistance over the 8 years to nalidixic acid (50%–26.7%, P = 0.003). On average from 2016 to 2019, Enterobacter spp. showed resistance of >50% to amoxicillin-clavulanic acid, cefoxitin, cefotaxime, ceftazidime, and cefepime, gentamicin, and trimethoprim-sulfamethoxazole. Escherichia coli showed no significant changes in resistance patterns, however, in 2016-2019 it had a resistance rate of ≥ 50% in amoxicillin-clavulanic gentamicin, nalidixic acid, ciprofloxacin, tetracycline, and trimethoprim-sulfamethoxazole [Table 3].
Table 3.
Progression of antibiotic resistance of the most frequent Enterobacteriaceae in blood cultures
| Antibiotic resistance | K. pneumoniae (n=94) | Enterobacter spp. (n=53) | E. coli (n=22) | ||||||
|---|---|---|---|---|---|---|---|---|---|
|
|
|
|
|||||||
| 2012–2015, n (%) | 2016–2019, n (%) | P | 2012–2015, n (%) | 2016–2019, n (%) | P | 2012-2015, n (%) | 2016-2019, n (%) | P | |
| AMC | 31/47 (66.0) | 29/41 (70.7) | 0.802 | 27/28 (96.4) | 17/18 (94.4) | 1.000 | 9/12 (75.0) | 5/7 (71.4) | 0.583 |
| Cefoxitin | 5/42 (11.9) | 1/27 (3.7) | 0.499 | 21/27 (77.8) | 12/12 (100.0) | 0.151 | 3/11 (27.3) | 0/3 (0) | 0.629 |
| Cefotaxime | 34/47 (72.3) | 40/46 (87.0) | 0.144 | 24/28 (85.7) | 21/24 (87.5) | 1.000 | 6/12 (50.0) | 5/10 (50) | 1.000 |
| Ceftazidime | 22/42 (52.4) | 18/30 (60.0) | 0.174 | 11/25 (44.0) | 8/15 (53.3) | 0.312 | 3/12 (25.0) | 2/4 (50) | 0.547 |
| Cefepime | 12/32 (37.5) | 37/45 (82.2) | <0.001 | 7/20 (35.0) | 14/24 (58.3) | 0.129 | 3/7 (42.9) | 5/10 (50) | 1.000 |
| Imipenem | 0/47 (0.0) | 0/46 (0.0) | - | 0/28 (0.0) | 1/24 (4.2) | 0.462 | 0/12 (0.0) | 0/10 (0) | - |
| Chloramphenicol | 15/47 (31.9) | 9/30 (30.0) | 0.451 | 12/27 (44.4) | 6/15 (40.0) | 0.667 | 2/11 (18.2) | 1/4 (25) | 1.000 |
| Amikacin | 7/45 (15.6) | 8/44 (18.2) | 0.791 | 2/27 (7.4) | 1/24 (4.2) | 1.000 | 1/11 (9.1) | 4/10 (40) | 0.149 |
| Gentamicin | 33/46 (71.7) | 41/45 (91.1) | 0.030 | 11/28 (39.3) | 12/23 (52.2) | 0.407 | 6/12 (50.0) | 6/10 (60) | 0.691 |
| Nalidixic acid | 19/45 (42.2) | 4/30 (13.3) | <0.001 | 13/26 (50.0) | 4/15 (26.7) | 0.003 | 5/11 (45.5) | 3/4 (75) | 0.569 |
| Ciprofloxacin | 14/47 (29.8) | 16/46 (34.8) | 0.095 | 6/28 (21.4) | 1/23 (4.4) | 0.194 | 4/12 (33.3) | 6/9 (66.7) | 0.091 |
| Tetracycline | 28/37 (75.7 | 22/26 (84.6) | 0.531 | 2/20 (10.0) | 5/15 (33.3) | 0.288 | 7/7 (100.0) | 3/4 (75) | 0.364 |
| TMP-SMX | 34/47 (72.3) | 42/46 (91.3) | 0.030 | 10/25 (40.0) | 11/22 (50.0) | 0.564 | 7/12 (58.3) | 8/8 (100) | 0.055 |
K. pneumoniae: 2012–2015 (n=48); 2016–2019 (n=46), Enterobacter spp.: 2012–2015 (n=29); 2016–2019 (n=24), Escherichia coli: 2012–2015 (n=12); 2016–2019 (n=10). All values in bold have P<0.05. AMC: Amoxicillin-clavulanic acid, TMP-SMX: Trimethoprim-sulfamethoxazole, K. pneumoniae: Klebsiella pneumoniae, E. coli: Escherichia coli
Pseudomonas spp. demonstrated resistance of 50% or greater in the 2016–2019 time period to ceftazidime, cefepime, imipenem, meropenem, and gentamicin. Acinetobacter spp. showed an increase in resistance over the 8 years to ceftazidime (41.7%–92.3%; P = 0.002), cefepime (54.6%–92.3; P = 0.016), imipenem (50%–88.5%; P = 0.016), trimethoprim-sulfamethoxazole (50%–95.6%; P = 0.003), and ciprofloxacin (45.5%–88.5%; P = 0.011) [Table 4].
Table 4.
Progression of antibiotic resistance of the most frequent non-fermenting Gram-negative bacilli in blood cultures
| Antibiotic resistance | Pseudomonas spp. (n=48) | Acinetobacter spp. (n=38) | ||||
|---|---|---|---|---|---|---|
|
|
|
|||||
| 2012–2015, n (%) | 2016–2019, n (%) | P | 2012–2015, n (%) | 2016–2019, n (%) | P | |
| SAM | - | - | 5/9 (55.6) | 22/25 (88.0) | 0.106 | |
| Ceftazidime | 13/24 (54.2) | 12/22 (54.6) | 0.878 | 5/12 (41.7) | 24/26 (92.3) | 0.002 |
| Cefepime | 20/26 (76.9) | 11/22 (50.0) | 0.164 | 6/11 (54.6) | 24/26 (92.31) | 0.016 |
| Imipenem | 13/26 (50.0) | 13/22 (59.1) | 0.551 | 6/12 (50.0) | 23/26 (88.5) | 0.016 |
| Meropenem | 15/25 (60.0) | 9/17 (52.9) | 0.433 | 7/12 (58.3) | 20/23 (87.0) | 0.091 |
| Aztreonam | 13/26 (50.0) | 4/13 (30.8) | 0.408 | - | - | - |
| Amikacin | 11/23 (47.8) | 9/21 (42.9) | 0.903 | 5/12 (41.7) | 17/24 (70.8) | 0.216 |
| Gentamicin | 18/26 (69.2) | 14/20 (70.0) | 1.000 | 8/12 (66.7) | 21/23 (91.3) | 0.151 |
| Minocycline | - | - | - | 1/10 (10.0) | 4/23 (17.4) | 0.581 |
| TMP-SMX | - | - | - | 6/12 (50.0) | 22/23 (95.6) | 0.003 |
| Ciprofloxacin | 8/25 (32.0) | 8/22 (36.4) | 0.768 | 5/11 (45.5) | 23/26 (88.5) | 0.011 |
Pseudomonas spp.: 2012–2015 (n=26); 2016–2019 (n=22), Acinetobacter spp.: 2012–2015 (n=12); 2016–2019 (n=26). All values in bold have P<0.05. SAM: Ampicillin-sulbactam, TMP-SMX: Trimethoprim-sulfamethoxazole
DISCUSSION
In this study, of the 6315 reports remaining, we examined 455 positive blood culture reports meeting criteria for representing HA-BSIs over a period of 8 years from 2012 to 2019, and we describe the evolution in AMR between 2012 and 2015 and 2016 and 2019.
A previous study in a pediatric hospital in South Africa showed that the most frequent microorganisms associated with hospital-acquired infections were K. pneumoniae and Acinetobacter baumannii; we partially replicated these results as K. pneumoniae was our most frequent isolate.[17] In the Cortes et al. study, in 33 Colombian hospitals, it was observed that the most frequently isolated microorganisms in the intensive care unit were staphylococci coagulase-negative, S. aureus, K. pneumoniae, E. coli, and A. baumannii (39.6%, 12, 3%, 8.2%, 5.7%, and 4%, respectively).[18] Furthermore, an international multicenter study noted that HA-BSI in intensive care unit patients was isolated Acinetobacter spp. 12.2% and Klebsiella spp. 11.9% as the two most frequent microorganisms out of a total of 1317 isolated microorganisms.[19] In Taiwan, the four bacteria most frequently isolated in adults due to a nosocomial bloodstream infection are coagulase-negative staphylococci (16%), S. aureus (13%), A. baumannii (8%), and E. coli (8%).[20] The difference in the frequency of the bacteria isolated in our study may be attributed to differences in the study population and geographic region.
Other studies in pediatric patients report S. aureus and coagulase-negative staphylococci as the most frequent causes of hospital-acquired sepsis.[21,22] For example, from 2009 to 2011, a public pediatric hospital in La Paz, Bolivia, obtained 318 positive blood cultures, and the most frequent isolate was S. aureus (21%). Among Gram-negative bacilli, the most commonly identified bacteria were E. coli (11.6%).[23]
We identified coagulase-negative staphylococci (15.8%) as the second most frequent cause of positive blood cultures. Unfortunately, we did not have access to the medical records to determine correlation with clinical sepsis. However, studies of pediatric patients that verified blood culture results with clinical evaluation found that 67.9%–82.2% of the resulting coagulase-negative staphylococci blood cultures were not clinically significant and were considered contaminants.[24,25] Therefore, we considered the majority of our positive blood cultures resulting from coagulase-negative staphylococci to be contaminants.
K. pneumoniae, an important cause of HA-BSI, is difficult to eradicate in hospital settings and shows a high frequency of natural and acquired resistance to a wide variety of antibiotics.[26,27] In our study, it showed higher resistance to cefotaxime, cefepime, gentamicin, tetracycline, and trimethoprim-sulfamethoxazole than the study carried out on pediatric patients in 16 hospitals in Brazil, where the resistance to the mentioned antibiotics was <40%.[28] A study of HA-BSI in children in England reports <30% resistance to ceftazidime, gentamicin, and cefotaxime/ceftriaxone in Klebsiella spp.[29] Among the main factors that make the patterns of antibiotic resistance differ from one study to another is the great diversity that exists in the management of bloodstream infections acquired in the hospital (HA-BSI), as well as the intrinsic characteristics of pathogens isolated. Both factors almost always depend on the local context.
Our study showed Pseudomonas spp. had >50% resistance to ceftazidime, imipenem, meropenem, and gentamicin. In contrast, studies in Mexico, the United States, Malaysia, and Brazil showed a relatively low AMR to the same antimicrobials with resistance rates lower than 40%.[28,30,31,32]
Finally, our study demonstrated that Acinetobacter has a resistance of 60%–70% to commonly used antibiotics. A study in pediatric hospitals also demonstrated a high resistance in Acinetobacter >45%.[26] However, a study at the Materno Infantil Hospital in Cochabamba, Bolivia, analyzed 36 isolates of A. baumannii. Of the 31.3% that came from blood cultures, the study reported 80.6% to be extreme drug resistant and 8.3% to be multidrug resistant.[33,34] This increased resistance in three clinically significant bacteria is concerning both for the treatment of patients at our hospital and for the trend of AMR globally in LMIC pediatric hospitals. Differences in age distribution, study population, local medical practice, antibiotic stewardship, and geographic differences may account for the differences in resistance seen between the Brazilian study and our own. It may also be that in the 9 years between this study and our own, AMR has increased significantly. This is certainly a worst-case scenario, as increasing AMR will likely lead to increased morbidity and mortality.
In the present study, we could suggest two possible resistance mechanisms: methicillin resistance and methylase. The first seems likely given that oxacillin resistance persists in the three analyzed microorganisms and the second mechanism seems likely as the three Staphylococcus analyzed are resistant to clindamycin and erythromycin.
Increasing AMR is one of the most serious global threats to public health,[35,36,37] with estimates of >$500 million USD in expenses in a single country.[3,36] Worldwide, physicians are seeing an increase in multidrug-resistant organisms, which are associated with increased morbidity and mortality.[38] In the US alone, 2.8 million infections are caused by antibiotic-resistant organisms and lead to 35,000 deaths annually.[39]
As discussed previously, the causes of AMR include poor antibiotic stewardship, overuse of last-resort antibiotics, evolution of microbials, and interaction between animal and agricultural antibiotic use.[40] Management of AMR includes accurate surveillance systems and development of novel antimicrobial agents which are effective despite AMR.[38] In Latin America, 67.4% of pediatric patients with HA-BSIs receive pediatric conserve (last-resort) antibiotics.[41] This is an astoundingly high number and concerning for the future treatment of infections in this region as multiple studies have found a relationship between antimicrobial overuse and AMR. Antimicrobial stewardship policies decrease AMR and improve patient outcomes.[42] In at least one pediatric hospital, the implementation of a stewardship program kept rates of AMR from increasing.[43] However, the development of this plus the antibiotic selection should be based on patient characteristics and local contexts, considering that the pattern of antibiotic sensitivity will differ across the settings. As in a study carried out on adult patients in Indonesia where isolated pathogens differ from our study as well as the pattern of AMR.[44] There is an urgent need for strict antimicrobial stewardship policies, pharmaceutical development, and continued surveillance of AMR in pediatric hospitals in LMIC.
Limitations
There are several limitations to our study. First, the low number of positive blood cultures could be the lack of hospital protocols and criteria for the correct use of blood cultures as diagnostic tools. Second, the small sample sizes, particularly with strains that had n < 30, could have influenced the statistical power when calculating the progression of the AMR. Third, it was not possible to obtain data on changes in empirical therapy antibiotic therapy and how it may have changed from year to year. Fourth, there was a variation in the number of antibiotics tested because the same antibiotics were not tested in all bacteria. Fifth, we did not have access to patients’ clinical data to compare to blood culture findings. Sixth, the hospital-generated susceptibility reports were often incomplete with limited adherence to CLSI M39 guidelines. Finally, an important limitation was the loss of a considerable amount of antibiotic data because many lacked cutoff points and some were no longer in the CLSI for the corresponding years.
CONCLUSIONS
This study shows a high resistance to antibiotics by bacteria isolated in the blood cultures of pediatric patients with HA-BSIs in a tertiary care pediatric hospital in Santa Cruz, Bolivia. We demonstrated significant changes in AMR in clinically important bacteria over just 8 years, showing a need to implement strong antimicrobial stewardship programs and continue to support pharmaceutical research. AMR is a dangerous and growing threat to global public health and decisive action should be taken to combat its effects.
Research quality and ethics statement
This study was approved by the Institutional Review Board “Hospital de Niños Dr. Mario Ortiz Suárez” (IRB00011318). The authors followed applicable EQUATOR Network guidelines during the conduct of this research project.
Conflicts of interest
There are no conflicts of interest.
Acknowledgment
The authors thank all the staff of Hospital de Niños Dr. Mario Ortiz Suárez in Bolivia for their support and collaboration throughout the data collection and analysis process to improve local public health. We would also like to give a special thanks to the staff of the Clinical Laboratory Department.
SUPPLEMENTARY TABLES
Supplementary Table 1.
Microbiological identification of Enterobacteriaceae in blood cultures
| Bacteria | Selective medium MacConkey agar | Identification test; battery of biochemical tests | |||||
|---|---|---|---|---|---|---|---|
|
| |||||||
| TSI | LIA | Citrate | SIM | Urea | Ornithine | ||
| K. pneumoniae | Mucous pink colonies (lactose fermentation) | Acid/acid with gas production | Alkaline/alkaline | Positive | Hydrogen sulfide: Negative Indole: Negative Motility: Negative | Positive | Negative |
| Enterobacter spp. | Mucous pink colonies (lactose fermentation) | Acid/acid with gas production | Alkaline/alkaline or alkaline/acid | Positive | Hydrogen sulfide: Negative Indole: Negative Motility: Positive | Positive | Positive or negative |
| E. coli | Dry pink colonies (lactose fermentation) | Acid/acid with gas production | Alkaline/alkaline | Negative | Hydrogen sulfide: Negative Indole: Positive Motility: Positive | Negative | Positive |
K. pneumoniae: Klebsiella pneumoniae, E. coli: Escherichia coli, TSI: Triple sugar iron, LIA: Lysine iron agar, SIM: Sulfite indole motility
Supplementary Table 2.
Microbiological identification of nonfermenting Gram-negative in blood cultures
| Bacteria | Selective medium MacConkey agar | Identification test | ||
|---|---|---|---|---|
|
| ||||
| TSI | Oxidase test | Gram staining | ||
| Pseudomonas spp. | Flat transparent colonies with pigmentation, lactose negative | Alkaline/alkaline | Positive | Gram-negative bacilli well defined |
| Acinetobacter spp. | Transparent colonies, lactose negative | Alkaline/alkaline | Negative | Gram-negative bacilli or coccobacilli, pleomorphic |
TSI: Triple sugar iron
Supplementary Table 3.
Microbiological identification of Gram-positive cocci in blood cultures
| Bacteria | Selective medium | Preliminary information Catalase test | Identification test | ||||
|---|---|---|---|---|---|---|---|
|
|
|
||||||
| Blood agar plate | Chocolate agar | Mannitol salt agar | DNAse test | Coagulase test | Novobiocin test | ||
| S. aureus | High and shiny colonies, creamy consistency. Complete hemolysis halo around the colonies (beta-hemolysis) | Small whitish colonies, circular in shape and rounded edges | Hydrolysis of mannitol acidification of the medium | Bubble release | Appearance of transparent halos around growing area | Serum coagulation | Resistant |
| Staphylococcus coagulase negative | Smooth gray colonies, mucoid appearance. No causes hemolysis (γ-hemolysis) | Greyish white colonies | Small colonies of color ranging from white to pink | Bubble release | There is not presence of characteristic halos | There is not serum clotting | Resistant |
| S. epidermidis | Smooth gray colonies, mucoid appearance. No causes hemolysis (γ-hemolysis) | Greyish white colonies | There is no change in the color of the medium | Bubble release | There is not presence of characteristic halos | There is not serum clotting | Susceptible |
S. aureus: Staphylococcus aureus, S. epidermidis: Staphylococcus epidermidis
Funding Statement
Nil.
REFERENCES
- 1.Calderon G, Aguilar L. Antimicrobial resistance: More resistant microorganisms and antibiotics resistant microorganisms and with less activity. Medical Journal of Costa Rica and Central America. 2016;73:757–63. [Google Scholar]
- 2.CDC. Antibiotic Resistance Threats in the United States, 2019 Atlanta, Georgia. 2019. [[Last accessed on 2020 Mar 05]]. Available from: https://stacks.cdc.gov/view/cdc/82532 .
- 3.Ahmad M, Khan AU. Global economic impact of antibiotic resistance: A review. J Glob Antimicrob Resist. 2019;19:313–6. doi: 10.1016/j.jgar.2019.05.024. [DOI] [PubMed] [Google Scholar]
- 4.World Health Organization. Antimicrobial resistance. [[Last accessed on 2020 Mar 07]]. Available from: https://www.who.int/news-room/fact-sheets/detail/antimicrobial-resistance .
- 5.Martinez-Martinez L, Calvo J. Development of antibiotic resistance:causes, consequences and their importance for public health. Journal Enfermedades Infecciosas y Microbiología Clínica. 2010;28(Suppl 4):49. doi: 10.1016/S0213-005X(10)70035-5. [DOI] [PubMed] [Google Scholar]
- 6.World Health Organization. WHO Publishes List of Bacteria for which New Antibiotics are Urgently Needed. 2017. [[Last accessed on 2020 Mar 10]]. Available from: https://www.who.int/en/news-room/detail/27-02-2017-who-publishes-list-of-bacteria -for-which-new-antibiotics-are-urgently-needed .
- 7.Centers for Disease Control. Antibiotic Prescribing and Use in Hospitals and Long-Term Care. 2017 [Google Scholar]
- 8.Carrillo R, Peña C, Sosa J. Sepsis from molecular basis to the campaign to increase survival. Acad Nac Med/Mexico. 2015:6980. [Google Scholar]
- 9.de Souza DC, Barreira ER, Faria LS. The epidemiology of sepsis in childhood. Shock. 2017;47:2–5. doi: 10.1097/SHK.0000000000000699. [DOI] [PubMed] [Google Scholar]
- 10.Plunkett A, Tong J. Sepsis in children. BMJ. 2015;350:h3017. doi: 10.1136/bmj.h3017. [DOI] [PubMed] [Google Scholar]
- 11.Thaden JT, Li Y, Ruffin F, Maskarinec SA, Hill-Rorie JM, Wanda LC, et al. Increased costs associated with bloodstream infections caused by multidrug-resistant gram-negative bacteria are due primarily to patients with hospital-acquired infections. Antimicrob Agents Chemother. 2017;61:e01709–16. doi: 10.1128/AAC.01709-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Morkel G, Bekker A, Marais BJ, Kirsten G, van Wyk J, Dramowski A. Bloodstream infections and antimicrobial resistance patterns in a South African neonatal intensive care unit. Paediatr Int Child Health. 2014;34:108–14. doi: 10.1179/2046905513Y.0000000082. [DOI] [PubMed] [Google Scholar]
- 13.Kouchak F, Askarian M. Nosocomial infections: The definition criteria. Iran J Med Sci. 2012;37:72–3. [PMC free article] [PubMed] [Google Scholar]
- 14.Bou G, Fernandez-Olmos A, Garcia C, Saez-Nieto JA, Valdezate S. Methods of bacterial identification in the microbiology laboratory. Journal Enfermedades Infecciosas y Microbiología Clínica. 2011;29:6018. doi: 10.1016/j.eimc.2011.03.012. [DOI] [PubMed] [Google Scholar]
- 15.Malbran C. Method for determination of antimicrobialsensitivity by diffusion. Serv Antimicrob. 2012;27:912. [Google Scholar]
- 16.CLSI. CLSI Supplement M100. 27th ed. Wayne, PA: Clinical and Laboratory Standards Institute; 2017. Performance standards for antimicrobial susceptibility testing. [Google Scholar]
- 17.Lochan H, Pillay V, Bamford C, Nuttall J, Eley B. Bloodstream infections at a tertiary level paediatric hospital in South Africa. BMC Infect Dis. 2017;17:750. doi: 10.1186/s12879-017-2862-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cortes JA, Leal AL, Montañez AM, Buitrago G, Castillo JS, Guzman L, et al. Frequency of microorganisms isolated in patients with bacteremia in intensive care units in Colombia and their resistance profiles. Braz J Infect Dis. 2013;17:346–52. doi: 10.1016/j.bjid.2012.10.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tabah A, Koulenti D, Laupland K, Misset B, Valles J, de Carvalho Bruzzi F, et al. Characteristics and determinants of outcome of hospital-acquired bloodstream infections in intensive care units: The EUROBACT international cohort study. Intensive Care Med. 2012;38:1930–45. doi: 10.1007/s00134-012-2695-9. [DOI] [PubMed] [Google Scholar]
- 20.Wu CJ, Lee HC, Lee NY, Shih HI, Ko NY, Wang LR, et al. Predominance of gram-negative bacilli and increasing antimicrobial resistance in nosocomial bloodstream infections at a university hospital in southern Taiwan, 1996-2003. J Microbiol Immunol Infect. 2006;39:135–43. [PubMed] [Google Scholar]
- 21.Ramirez A, Moreno L, Moreno L, Nuñez M, Cebada R, Aguirre C. Frequency and susceptibility profile of isolates obtained from blood cultures in a tertiary hospital center. Journal Anales Medicos Mexico. 2015;60:27. [Google Scholar]
- 22.Larru B, Gong W, Vendetti N, Sullivan KV, Localio R, Zaoutis TE, et al. Bloodstream infections in hospitalized children: Epidemiology and antimicrobial susceptibilities. Pediatr Infect Dis J. 2016;35:507–10. doi: 10.1097/INF.0000000000001057. [DOI] [PubMed] [Google Scholar]
- 23.Rodriguez Auad JP, Arias LL, Johnson KM, Caniza M. BloodCulture Procedures and Results in a Pediatric Hospital in La Paz, Bolivia: Opportunities for Improving Efficiency and Decreasing Cost. American Journal of Infection Control. 2012;40:e137. [Google Scholar]
- 24.Okay S, Gülhan B, Kanık S, Özkaya A, İclal G. Comparison of contamination and infection of CoNS bacteremia in hospitalized children's blood cultures, 18 August 2023, PREPRINT (Version 1) available at Research Square. [https://doi.org/10.21203/rs.3.rs-3271774/v1] [Google Scholar]
- 25.El-Naggari MA, Al-Mulaabed SW, Al-Muharrmi Z, Mani R, Abdelrahim R, Abdwani R. Blood culture contaminants in a paediatric population retrospective study from a tertiary hospital in Oman. Sultan Qaboos Univ Med J. 2017;17:e202–8. doi: 10.18295/squmj.2016.17.02.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Santistevan Y, Carmona Y, Perez F, Yoima D, Diaz N, Garcia G, et al. Klebsiella and Acinetobacter infections and antibiotic resistance in Cuban pediatric hospitals. Rev Cubana Med Trop. 2014;66:40014. [Google Scholar]
- 27.Moremi N, Claus H, Mshana SE. Antimicrobial resistance pattern: A report of microbiological cultures at a tertiary hospital in Tanzania. BMC Infect Dis. 2016;16:756. doi: 10.1186/s12879-016-2082-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Pereira CA, Marra AR, Camargo LF, Pignatari AC, Sukiennik T, Behar PR, et al. Nosocomial bloodstream infections in Brazilian pediatric patients: Microbiology, epidemiology, and clinical features. PLoS One. 2013;8:e68144. doi: 10.1371/journal.pone.0068144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Blackburn RM, Henderson KL, Minaji M, Muller-Pebody B, Johnson AP, Sharland M. Exploring the epidemiology of hospital-acquired bloodstream infections in children in England (January 2009-March 2010) by linkage of national hospital admissions and microbiological databases. J Pediatric Infect Dis Soc. 2012;1:284–92. doi: 10.1093/jpids/pis084. [DOI] [PubMed] [Google Scholar]
- 30.Ares MÁ, Alcántar-Curiel MD, Jiménez-Galicia C, Rios-Sarabia N, Pacheco S, De la Cruz MÁ. Antibiotic resistance of gram-negative bacilli isolated from pediatric patients with nosocomial bloodstream infections in a Mexican tertiary care hospital. Chemotherapy. 2013;59:361–8. doi: 10.1159/000362085. [DOI] [PubMed] [Google Scholar]
- 31.Logan LK, Gandra S, Mandal S, Klein EY, Levinson J, Weinstein RA, et al. Multidrug- and carbapenem-resistant Pseudomonas aeruginosa in children, United States, 1999-2012. J Pediatric Infect Dis Soc. 2017;6:352–9. doi: 10.1093/jpids/piw064. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Ariffin N, Hasan H, Ramli N, Ibrahim NR, Taib F, Rahman AA, et al. Comparison of antimicrobial resistance in neonatal and adult intensive care units in a tertiary teaching hospital. Am J Infect Control. 2012;40:572–5. doi: 10.1016/j.ajic.2012.02.032. [DOI] [PubMed] [Google Scholar]
- 33.Cerezales M, Ocampo-Sosa AA, Álvarez Montes L, Díaz Ríos C, Bustamante Z, Santos J, et al. High prevalence of extensively drug-resistant Acinetobacter baumannii at a children hospital in Bolivia. Pediatr Infect Dis J. 2018;37:1118–23. doi: 10.1097/INF.0000000000001962. [DOI] [PubMed] [Google Scholar]
- 34.Magiorakos AP, Srinivasan A, Carey RB, Carmeli Y, Falagas ME, Giske CG, et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: An international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect. 2012;18:268–81. doi: 10.1111/j.1469-0691.2011.03570.x. [DOI] [PubMed] [Google Scholar]
- 35.Ferri M, Ranucci E, Romagnoli P, Giaccone V. Antimicrobial resistance: A global emerging threat to public health systems. Crit Rev Food Sci Nutr. 2017;57:2857–76. doi: 10.1080/10408398.2015.1077192. [DOI] [PubMed] [Google Scholar]
- 36.Septimus EJ. Antimicrobial resistance: An antimicrobial/diagnostic stewardship and infection prevention approach. Med Clin North Am. 2018;102:819–29. doi: 10.1016/j.mcna.2018.04.005. [DOI] [PubMed] [Google Scholar]
- 37.Nolte O. Antimicrobial resistance in the 21st century: A multifaceted challenge. Protein Pept Lett. 2014;21:330–5. doi: 10.2174/09298665113206660106. [DOI] [PubMed] [Google Scholar]
- 38.Davido B, Bouchand F, Dinh A, Perronne C, Villart M, Senard O, et al. Reinforcement of an antimicrobial stewardship task force aims at a better use of antibiotics of last resort: The COLITIFOS study. Int J Antimicrob Agents. 2017;50:142–7. doi: 10.1016/j.ijantimicag.2017.03.030. [DOI] [PubMed] [Google Scholar]
- 39.Shrestha P, Cooper BS, Coast J, Oppong R, Do Thi Thuy N, Phodha T, et al. Enumerating the economic cost of antimicrobial resistance per antibiotic consumed to inform the evaluation of interventions affecting their use. Antimicrob Resist Infect Control. 2018;7:98. doi: 10.1186/s13756-018-0384-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.McEwen SA, Collignon PJ. Antimicrobial resistance: A one health perspective. Microbiol Spectr. 2018;6 doi: 10.1128/microbiolspec.arba-0009-2017. [doi:10.1128/microbiolspec. ARBA-0009-2017] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Brinkac L, Voorhies A, Gomez A, Nelson KE. The threat of antimicrobial resistance on the human microbiome. Microb Ecol. 2017;74:1001–8. doi: 10.1007/s00248-017-0985-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Bielicki JA, Sharland M, Versporten A, Goossens H, Cromwell DA. Using risk adjustment to improve the interpretation of global inpatient pediatric antibiotic prescribing. PLoS One. 2018;13:e0199878. doi: 10.1371/journal.pone.0199878. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Liu XL, Yang J, Chen XH, Hua ZY. Effects of antibiotic stewardship on neonatal bloodstream infections. Zhongguo Dang Dai Er Ke Za Zhi. 2016;18:796–801. doi: 10.7499/j.issn.1008-8830.2016.09.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Pradipta IS, Sandiana AT, Halimah E, Diantini A, Lestari K, Abdulah R. Microbial and resistance profile in isolate from adult sepsis patients: An observational study at an Indonesian private hospital during 2009-2012. Int J Pharm Sci Rev Res. 2013;19:24–9. [Google Scholar]