A prospective observational study of community-acquired bacterial bloodstream infections in Metro Manila, the Philippines

Community-acquired bacterial bloodstream infections are caused by diverse pathogens with changing antimicrobial-resistance patterns. In low-middle income countries in Southeast Asia, where dengue fever is endemic and a leading cause of fever, limited information is available about bacterial bloodstream infections due to challenges of implementing a blood culture service. This study describes bacterial bloodstream pathogens and antimicrobial-resistance patterns in Metro Manila, the Philippines. We aimed to identify the proportion of patients with a positive blood culture, the bacteria isolated and their antimicrobial resistance patterns, and the clinical characteristics of these patients, in this dengue endemic area. We conducted a prospective observational study in a single hospital enrolling febrile patients clinically suspected of having a community-acquired bacterial bloodstream infection between 1st July 2015 and 30th June 2019. Each patient had a blood culture and additional diagnostic tests according to their clinical presentation. We enrolled 1315 patients and a significant positive blood culture was found in 77 (5.9%) including Staphylococcus aureus (n = 20), Salmonella enterica Typhi (n = 18), Escherichia coli (n = 16), Streptococcus pneumoniae (n = 3) and Burkholderia pseudomallei (n = 2). Thirty-four patients had meningococcal disease diagnosed by culture (n = 8) or blood PCR (n = 26). Additional confirmed diagnoses included leptospirosis (n = 177), dengue virus infection (n = 159) and respiratory diphtheria (n = 50). There were 79 (6.0%, 95%CI 4.8%−7.4%) patients who died within 28 days of enrollment. Patients with a positive blood culture were significantly more likely to die than patients with negative culture (15.2% vs 4.4%, P<0.01). Among S. aureus isolates, 11/20 (55%) were methicillin-resistant (MRSA) and ST30: USA1100 was dominant sequence type (88.9%). Antimicrobial-susceptibility was well preserved in S. enterica Typhi. Among hospitalized patients with clinically suspected community-acquired bacterial bloodstream infection in Metro Manila, the Philippines, 5.9% had a blood culture confirmed infection of whom 15.6% died. S. aureus, including a significant number of MRSA (USA1100 clones), S. enterica Typhi, E.coli and Neisseria meningitidis were frequently identified pathogens.

. Final laboratory and clinical confirmed diagnosis when more than two pathogens were identified by study tests Appendix 3. Table S2. Characteristics of patients enrolled as the dengue control (Clinical diagnosed with dengue fever) Appendix 4. Table S3. Physical and clinical signs of enrolled patients and the association with positive results of blood culture Appendix 5. Sample collection and diagnostic tests to be perform per protocol

Samples collected by the study
We obtained blood samples (Acute and Convalescent) from enrolled patients during the enrollment.
The blood volume is shown below according to the age.

Acute phase sample
Adult and children over 6 years: Total 10−19 ml (6-14 ml for bacterial culture divided equally between two blood culture bottles; 1 ml for complete blood count (CBC), 2 mL for renal function and liver enzymes; 2 mL EDTA research sample for polymerase chain reaction (PCR) and other tests.) For children aged between 3 and 6 years: Total 8−15 ml (4−10 ml for bacterial culture divided equally between two blood culture bottles (or in one blood culture bottle if volume < 6 mL); 1 ml for CBC, 2 mL for renal function and liver enzymes; 2 mL EDTA research sample for PCR and other tests.
For children aged between 1 and 2 years: Total 6−11 ml: 2−6 ml for bacterial culture in one blood culture bottle, 1 ml for CBC, 2 mL for renal function and liver enzymes; 2 mL EDTA research sample for PCR and other tests.

Convalescent samples
We obtained an additional 2 mL EDTA research blood sample drawn for convalescent serology between 7 and 10 days of after enrollment or on the day of discharge if sooner.
The research blood sample were taken into an EDTA tube and centrifuged within 3 hours at1000 g for 10 minutes. Then it was divided into buffy coat and plasma in separate tubes. Plasma was divided into two tubes. These were stored at -80°C freezer.
A urine sample was also collected from each patient for microscopy and culture when relevant; an assay of antimicrobial activity; and storage at -80°C DNA from blood samples One plasma tube was centrifuged at 21,000 g for 10 min. Then DNA extracted from the plasma pellet using QIAamp DNA Mini Kit (QIAGEN Inc., Valencia, CA, USA). DNA was also extracted from buffy coat samples using the QIAamp DNA Mini Kit (QIAGEN Inc., Valencia, CA, USA).

Laboratory procedure of each test
Blood culture Trained nurse collected blood from a single peripheral site. We performed blood culture processing, isolate identification, and antimicrobial susceptibility testing (AST) at the SLH-Nagasaki collaborative laboratory in SLH. Blood was inoculated into two aerobic blood culture bottles of automated blood culture systems. Only aerobic culture was performed but not anaerobic culture. We

Dengue virus Conventional RT-PCR [5]
RNAs were extracted from 100 μL plasm portion by using ISOGEN II (Wako, Japan) accor ding to the manufacturer's protocol. The generic pan-dengue primers used, which targeted th e 3′ noncoding region of dengue viruses, were pan-dengue forward (5′-TCAATATGCTGAA ACGCGCGAGAAACCG-3′) and pan-dengue reverse (5′-GAAAACTTTTCTTCGTACCACGG GACTAA -3′). Conventional RT-PCR reactions were performed on the Veriti Thermal Cycler isolate. Conventional serogrouping PCR assay for genotyping for serogroup was performed using extracted DNA from blood buffy coat samples or isolates when blood PCR or blood culture was positive following the WHO manuals [10]. Because preliminary analysis showed serogroup B and Y were common in Manila, we used the primers of serogroup B (synD) and Y(synF).

Diagnostic tests of X-ray
If pneumonia was suspected, Chest X-ray was ordered by an attending physician. The X-ray was assessed by both radiologist and the physician by blind method. They were requested to choose three categories namely, "not pneumonia", "possible pneumonia", and "highly suspected pneumonia". If both the radiologist and the physician chose "highly suspected pneumonia", we defined the patients as X-ray confirmed pneumonia.
The detection of PCT was based on immunochromatography technique using Quantitative Immunoassay Analyzer (Sekisui Medical, Japan). The Alere Afinion CRP kit with Alere Afinion AS100 Analyzer (Alere Medical, Japan) was used to measure CRP. Tests require 120 µl and 1.5 µl of either serum or whole blood and have a detection range of 0.2-10 ng/ml and 5-160 mg/L, respectively.

Analysis of Staphylococcus aureus
Laboratory test for MRSA was done by the cefoxitin disk diffusion test. Detection of inducible clindamycin resistance was by the disc approximation D-zone test. In addition, PCR and multiplex

Community-acquired bacterial bloodstream infections in Manila Supplementary appendix
Appendix 5. Table S4. The blood culture volume and blood culture positivity