Issue: Newsletter 31 | November 1, 2026
Citation of Articles |
PICO |
Main Results |
Risk of Bias |
|---|---|---|---|
P: 60 very preterm or very low-birth-weight neonates requiring central venous access in a single level III neonatal intensive care unit of a public tertiary hospital I: Split septum needleless connector (n=30) C: Mechanical valve needleless connector (n=30) O: CLABSI incidence rate and hazard ratio of CLABSI (followed up to 48 hours after central line removal); CLABSI-related mortality, length of stay and hospital cost |
The CLABSI incidence rate was lower with split septum connectors than mechanical valve connectors (3.0 vs 16.2 per 1000 catheter days), but the difference was not statistically significant (adjusted hazard ratio 0.153, 95% CI 0.02–1.52; P=0.109). The mechanical valve group also had numerically higher CLABSI-related mortality (7% vs 0%), longer length of stay (49.8 vs 42.3 days) and higher hospital cost, none of which were statistically significant. Overall, split septum connectors showed a trend towards fewer CLABSIs in very preterm or very low-birth-weight neonates, which should be weighed carefully when choosing a connector type. |
Moderate to high risk: The open-label, single-centre design could introduce performance and detection bias, and the small sample of 60 neonates gives very wide confidence intervals and limited power to detect a difference. Groups differed in mean gestational age (27.4 vs 30.3 weeks), which may influence infection risk despite adjustment, and the single-centre setting limits generalisability. |
|
P: 241 patients aged ≥12 years with a long-term central venous catheter (in place ≥5 days) and central line-associated bloodstream infection (CLABSI) at 34 centres in the United States, Puerto Rico and India (228 received lock therapy) I: Systemic antibiotics plus minocycline-EDTA-ethanol lock therapy (MLT) for 7–14 days C: Systemic antibiotics plus standard-of-care antibiotic lock therapy (SOC-ALT) for 7–14 days O: Median time to catheter failure (clinical failure: death, infection-related catheter removal or worsening symptoms; or microbiological failure: persistent/recurrent infection or new deep-seated infection); catheter retention; safety |
Catheter retention was higher with MLT than SOC-ALT (57% vs 38%; P=0.0025). Median time to catheter failure was longer with MLT in the intention-to-treat population (not estimable vs 33 days; HR 0.53, P=0.0006) and the modified intention-to-treat population (not estimable vs 35 days; HR 0.50, P=0.0005). Clinical catheter failure (13% vs 28%; P=0.006) and microbiological failure (5% vs 15%; P=0.012) were both lower with MLT. Adverse events, serious adverse events and mortality were similar between groups, with no drug-related serious adverse events. Overall, MLT was superior to standard antibiotic lock therapy for salvaging long-term central venous catheters in CLABSI and was well tolerated. |
Moderate risk: Blinding of participants and clinicians to the lock solution is not described, which could introduce performance bias and influence decisions about catheter removal, a component of the primary outcome. Thirteen randomised patients did not receive lock therapy, and recruitment over six years across three countries with varied standard-of-care locks may limit consistency of the comparator. |
|
P: Six primary care centres in Catalonia, Spain, all implementing a standard antimicrobial stewardship (AMS) programme; prescribing assessed in patients aged over 14 years with upper respiratory tract infection (URTI) or urinary tract infection (UTI) I: Advanced AMS with infectious disease expert support via a telephone helpline and biweekly meetings (3 centres) C: Standard AMS programme alone (3 centres) O: Total antibiotic consumption (defined daily doses per 1000 inhabitants per day, DID); unnecessary and inadequate prescriptions; 30-day reconsultations or hospitalisations |
Between October 2021 and March 2023, overall antibiotic consumption rose from 7.08 to 7.79 DID. The rise was smaller with advanced AMS (+0.40 DID, 95% CI −0.40 to 1.20) than with standard AMS (+1.01 DID, 95% CI 0.15 to 1.88), but there was no significant difference between groups in monthly consumption trends (incidence rate ratio 0.95, 95% CI 0.19–4.74). Across URTI/UTI episodes, 25.0% of prescriptions were unnecessary and 50.3% of indicated treatments were inadequate. By the end of the intervention, advanced AMS reduced unnecessary prescriptions (21.2% vs 28.3%; OR 0.68, 95% CI 0.51–0.91) and inadequate prescriptions (38.8% vs 52.9%; OR 0.54, 95% CI 0.36–0.81), with no difference in reconsultations or hospitalisations. Overall, adding infectious disease experts did not significantly reduce total antibiotic use but improved prescribing quality without compromising patient safety. |
Moderate risk: With only six clusters (three per arm), baseline differences between centres may not be balanced by randomisation and the estimate of between-group difference in consumption is very imprecise. Clinicians could not be blinded to the intervention, although outcome assessors were blinded, and prescribing quality was measured through point-prevalence surveys that may not capture all episodes. |
|
P: 20,023 household members aged ≥18 years in community tuberculosis case-finding in Butha-Buthe District, Lesotho, and uMgungundlovu District, South Africa (22.6% living with HIV) I: Computer-aided detection on digital chest x-ray followed by point-of-care C-reactive protein testing when the CAD score fell within a set range (CAD4TBv7-CRP approach) C: Computer-aided detection on digital chest x-ray alone (CAD4TBv7 approach), each with confirmatory Xpert MTB/RIF Ultra testing as required (paired, within-person comparison) O: Tuberculosis case yield (non-inferiority margin −10%) and cost per detected case |
Of 73 participants identified with tuberculosis by at least one approach, CAD alone identified 69 (94.5%) and CAD followed by CRP identified 60 (82.2%), a difference of −12.3% (95% CI −23.0 to −1.6), so non-inferiority was not met. The cost per detected case was US$5454 for CAD alone and US$7486 for CAD plus CRP, making the combined approach 37.3% more expensive. Overall, adding CRP testing after CAD offered no advantage because of its lower case yield and higher cost, whereas CAD alone appears to be an effective and economically viable screening strategy in high-burden areas. |
Moderate risk: This paired screen-positive design compares two algorithms within the same individuals rather than randomising participants, which removes between-group confounding but means only participants positive on at least one approach contribute to the primary outcome. The small number of tuberculosis cases (73) limits precision, and cases missed by both approaches would not be detected, so absolute sensitivity cannot be estimated. |
|
P: 399 men who have sex with men aged ≥16 years with at least one sexually transmitted infection (STI) risk behaviour in the past 12 months, attending community-based organisations in seven provinces in China I: Co-creation plus pay-it-forward (free doxy-PEP supported by peer donations) (CC + PIF); or co-creation plus pay-it-forward with social network distribution of doxy-PEP to peers (CC + PIF + SND) C: Co-creation alone (educational materials introducing doxy-PEP co-created with local community members) O: Doxy-PEP use within 72 hours of condomless anal or oral sex with a male partner on at least one occasion at 6 months |
At 6 months, doxy-PEP use after condomless sex was reported by 41% (48/117) of participants in the co-creation group, 76% (89/117) in the CC + PIF group (risk difference 31.3, 95% CI 19.1–43.6; p<0.0001) and 73% (80/109) in the CC + PIF + SND group (risk difference 34.3, 95% CI 22.5–46.2; p<0.0001). Among peers recruited by participants, doxy-PEP use was higher with social network distribution than with referral to community-based organisations (68% vs 40%; risk difference 28.3, 95% CI 11.8–44.8). Overall, community-engaged strategies substantially increased doxy-PEP uptake in a setting with low baseline use, with co-creation and pay-it-forward working synergistically. |
Moderate risk: The open-label design and reliance on self-reported doxy-PEP use could introduce performance and reporting bias, and 14% of participants were lost to follow-up at 6 months, with missing data handled by multiple imputation. Findings among men attending community-based organisations in China may not generalise to other populations or health systems. |
|
P: 108 healthy adults aged 18–49 years at two sites in the USA (73 underwent oral challenge with Shigella sonnei 53G) I: Two doses of WRSs2 live-attenuated S. sonnei vaccine (10⁶ or 5 × 10⁵ CFU, 28 days apart), or one dose of placebo followed by one dose of WRSs2 C: Placebo O: Endpoint review committee-adjudicated shigellosis after challenge; safety |
Adjudicated shigellosis occurred in 9% (3/34) of challenged participants given pooled two-dose vaccine and 81% (21/26) of placebo recipients, giving a vaccine efficacy of 89% (95% CI 71–96; p<0.0001). Six participants had grade 3 post-vaccination adverse events, prompting a protocol amendment to reduce the vaccine dose and revise eligibility criteria; there were no vaccine-related serious adverse events or deaths. Overall, WRSs2 provided high-level protection against S. sonnei shigellosis in a controlled human infection model, although the grade 3 adverse events indicate that further optimisation of the safety-efficacy balance is needed. |
Low to moderate risk: The double-blind, placebo-controlled design with adjudicated outcomes minimises performance and detection bias. However, the mid-trial change in dose and allocation ratio, pooling of vaccine groups and small numbers challenged limit precision, and efficacy in a controlled challenge model of healthy US adults may not reflect protection in endemic settings or in children. |
|
P: 17,701 adults aged ≥60 years in stable health with a history of urinary tract infection within the previous 2 years, at 332 sites in 20 countries I: Single intramuscular dose of ExPEC9V, a nine-valent E. coli O-serotype conjugate vaccine (n=8858) C: Placebo (n=8843) O: First invasive E. coli disease (IED) in blood or sterile sites caused by ExPEC9V O-serotypes; immunogenicity; safety |
A prespecified futility analysis indicated lack of efficacy and the trial was stopped early. There were 24 IED cases caused by vaccine O-serotypes (11 with ExPEC9V vs 13 with placebo), giving a vaccine efficacy of 14.63% (95% CI −106.57 to 65.38), indicating no protective effect. ExPEC9V had an acceptable safety profile, with more frequent low-grade adverse events than placebo, balanced serious adverse events and no vaccine-related deaths. Overall, ExPEC9V did not prevent invasive E. coli disease in older adults despite robust and sustained antibody responses, and its development has been discontinued. |
Low risk: The double-blind, placebo-controlled design with central randomisation minimises selection, performance and detection bias. Early termination for futility with only 24 events means the efficacy estimate is very imprecise, and the restriction to adults with a recent urinary tract infection may limit generalisability, although the null result is consistent and unlikely to be explained by bias. |
|
P: 366 healthy children aged 6–17 years primed with acellular pertussis vaccines at 16 sites in the UK, Australia and Costa Rica I: Intranasal BPZE1 with intramuscular placebo (n=123); or intranasal BPZE1 with intramuscular Tdap (n=121) C: Intramuscular Tdap with intranasal placebo (n=122) O: Geometric mean fold rise in nasal secretory IgA (S-IgA) to B. pertussis whole-cell extract at day 29; serum IgG to pertussis antigens and diphtheria/tetanus seroprotection; reactogenicity and safety |
The primary endpoint was met, with nasal S-IgA geometric mean fold rises at day 29 of 3.8 (95% CI 3.1–4.7) with BPZE1 and 3.5 (2.9–4.3) with BPZE1 plus Tdap (both p<0.0001). Serum IgG responses to pertussis antigens with BPZE1 plus Tdap were comparable with Tdap alone, and all participants in both Tdap groups had protective diphtheria and tetanus antibody levels. Solicited adverse events occurred in 48%, 67% and 72% of the BPZE1, BPZE1 plus Tdap and Tdap groups, respectively, were mostly mild to moderate, and no vaccine-related serious adverse events occurred. Overall, BPZE1 was well tolerated and induced robust nasal mucosal immunity without interfering with Tdap responses, supporting further development in acellular vaccine-primed children. |
Low to moderate risk: Randomisation and observer-blinding with placebo and active comparators reduce selection and detection bias. The outcomes are immunological surrogates rather than protection against infection or disease, and the trial was not designed to assess clinical efficacy, so the clinical relevance of the mucosal responses remains to be established. |
|
P: 112 adults with confirmed Stenotrophomonas maltophilia infections at Siriraj Hospital, Thailand (71.4% requiring intensive care; 66.1% with ventilator-associated pneumonia) I: Minocycline added to monotherapy with levofloxacin or trimethoprim/sulfamethoxazole (n=55) C: Matching placebo added to monotherapy with levofloxacin or trimethoprim/sulfamethoxazole (n=57) O: 28-day mortality; microbiological and clinical response |
Adjunctive minocycline did not significantly reduce 28-day mortality (43.6% vs 38.6%; P=0.590) but achieved higher microbiological response at early assessment (45.5% vs 14.0%; P<0.001) and at the end of treatment (80.0% vs 56.1%; P=0.023). Clinical response did not differ at 72 hours or at the end of treatment. Prior antibiotic use (aOR 7.65) and septic shock (aOR 7.24) were independently associated with 28-day mortality. Overall, adding minocycline enhanced microbiological eradication but did not confer a survival or clinical advantage, and combination therapy was well tolerated. |
Low to moderate risk: The double-blind, placebo-controlled design minimises performance and detection bias. The single-centre setting and modest sample size limit power to detect a mortality difference, and the predominance of ventilator-associated pneumonia in critically ill patients, where colonisation and infection can be difficult to distinguish, may limit generalisability. |
|
P: 425 adults newly diagnosed with Mycobacterium avium complex lung disease (MACLD) who had not started antibiotics (median age 68 years; 79.3% female) I: Amikacin liposome inhalation suspension (ALIS) 590 mg once daily plus azithromycin 250 mg and ethambutol 15 mg/kg for 12 months (n=213) C: Placebo plus azithromycin and ethambutol for 12 months (n=212) O: Change in Respiratory Symptom Score (RSS) from baseline to month 13; culture conversion; durable culture conversion; safety |
ALIS improved respiratory symptoms more than the comparator at month 13 (least squares mean change 17.77 vs 14.66 points; difference 3.11 points; P=0.0299), with the benefit sustained at month 15 (difference 4.80 points; nominal P=0.0015). More patients achieved culture conversion by month 13 (82.4% vs 55.6%) and durable culture conversion at month 15 (76.2% vs 47.6%) with ALIS (both P<0.0001), and no new safety signals were observed. Overall, adding ALIS to azithromycin and ethambutol was superior for improving respiratory symptoms and culture conversion in newly diagnosed MACLD. |
Low risk: The double-blind, placebo-controlled design minimises performance and detection bias for both the patient-reported symptom score and culture outcomes. The modest absolute difference in symptom score may be of uncertain clinical importance, and the comparator was a two-drug regimen, so the benefit compared with standard three-drug therapy including a rifamycin is not directly addressed. |
|
P: 34 febrile brain-injured patients with brain temperature ≥38.5°C requiring targeted temperature management (TTM), monitored with an intracerebral pressure probe with a thermal sensor, at a single centre I: TTM targeted to brain temperature (n=17) C: TTM targeted to systemic (bladder) temperature (n=17) O: Proportion of time with brain temperature within the 36.0–37.0°C target range during the maintenance phase (hours 3–12) |
The median proportion of time with brain temperature within the target range was similar with brain temperature-targeted TTM and systemic temperature-targeted TTM (84% vs 89%; p=0.88), and target attainment at hour 3 was also similar (p=0.49). The gradient between brain and systemic temperature increased from 0.11°C during induction to 0.30°C during maintenance, and per-protocol analyses were consistent. Overall, targeting brain temperature did not improve brain temperature control compared with conventional systemic temperature-targeted TTM. |
Moderate risk: The single-centre design and small sample of 34 patients limit precision and generalisability, and blinding of clinicians to the temperature target was not feasible. The primary outcome is a physiological measure of temperature control, so the trial does not address clinical outcomes such as neurological recovery. |
|
P: 2247 pregnant women aged ≥18 years enrolled before 27 weeks' gestation (1910 with live births) I: Screening and treatment for Chlamydia trachomatis, Neisseria gonorrhoeae and Trichomonas vaginalis at enrolment with tests of cure (1-time screening); or screening and treatment at enrolment repeated at 30–34 weeks (2-time screening) C: Standard of care (syndromic management) O: Composite of preterm birth (<37 weeks) or low birthweight (<2500 g); each component separately |
The composite outcome occurred in 22.9% with 1-time screening (RR 0.99, 95% CI 0.81–1.21) and 20.6% with 2-time screening (RR 0.89, 95% CI 0.72–1.09) compared with 23.2% with standard of care. Preterm birth was reduced with 2-time screening (14.5% vs 18.8%; RR 0.77, 95% CI 0.60–0.99) but not with 1-time screening (18.9%; RR 1.00), and low birthweight did not differ between groups (14.1%, 12.9% and 12.8%). Overall, neither screening strategy reduced the composite outcome or low birthweight, although repeated screening reduced preterm birth by 23%. |
Moderate risk: Participants and clinicians were aware of the screening strategy, which could introduce performance bias, although birth outcomes are objective. About 15% of enrolled women did not contribute live-birth outcomes, and the preterm birth finding was a secondary outcome with a confidence interval close to 1, so it should be interpreted cautiously. |
|
P: 166 adults with acute leukaemia or haematopoietic stem cell transplant recipients at 11 centres in Spain I: Oral fosfomycin prophylaxis (n=83) C: Oral ciprofloxacin prophylaxis (n=83) O: Rate of febrile neutropenia (non-inferiority margin −5%); bacteraemia; colonisation with multidrug-resistant Gram-negative bacteria; safety |
In the intention-to-treat population, febrile neutropenia occurred in 74.7% with fosfomycin and 73.5% with ciprofloxacin (risk difference in success −1.2%, 95% CI −14.5 to 12.1; lower boundary of the 90% CI −12.3), and in the per-protocol population in 82.2% and 76.3% (risk difference −5.9%, 95% CI −18.8 to 6.9), so non-inferiority was not demonstrated. Gram-negative bacteria caused 90% of bacteraemias with fosfomycin compared with 10% with ciprofloxacin, while colonisation with multidrug-resistant Gram-negative bacteria was less frequent with fosfomycin (3% vs 11%; P=0.04). Overall, fosfomycin was not non-inferior to ciprofloxacin for preventing febrile neutropenia, and the higher rate of Gram-negative bacteraemia discourages its use. |
Moderate risk: The open-label design could introduce performance bias in the assessment and management of fever, and the small sample size gives wide confidence intervals relative to the 5% non-inferiority margin. Exclusions from the per-protocol population were greater in the fosfomycin arm, and the high baseline rate of febrile neutropenia in both arms limits the ability to distinguish between agents. |
|
P: 118 women aged 18–70 years with acute uncomplicated cystitis (111 in the per-protocol set) I: D-mannose as stand-alone treatment (n=57 per protocol) C: Fosfomycin (n=54 per protocol) O: Clinical cure assessed by the Acute Cystitis Symptom Score up to day 8 (non-inferiority margin 15%); recurrence; additional antibiotic use; adverse events |
Median time to clinical cure was 5.0 days with D-mannose and 3.0 days with fosfomycin. The point estimate for the difference in clinical cure on day 8 (−10.1%) favoured fosfomycin and was within the non-inferiority margin, but non-inferiority was not statistically confirmed because of wide confidence intervals; a post hoc analysis estimated similar cure rates (84.2% vs 83.5%). Recurrence (13.2% vs 13.3%) and additional antibiotic use (19.3% vs 11.1%) did not differ significantly, and gastrointestinal adverse events were less frequent with D-mannose (9.8% vs 24.6%). Overall, D-mannose may be a promising antibiotic-sparing alternative, but an adequately powered trial is needed. |
Moderate to high risk: Although double-blind, the small sample size meant the prespecified non-inferiority analysis was inconclusive, and comparability was shown only in a post hoc analysis, which increases the risk of selective interpretation. Baseline imbalance in severe symptoms (32.8% vs 22.8%) and missing data may also influence the results. |
|
P: 210 participants of any age with tuberculosis infection at 7 tuberculosis clinics in Sydney, Australia I: Weekly isoniazid plus rifapentine for 3 months (3HP), self-administered with weekly SMS adherence reminders (n=106) C: Daily rifampicin for 4 months (4RIF), self-administered (n=104) O: Treatment completion (ingestion of ≥90% of prescribed doses); adverse events |
Treatment completion was significantly higher with 3HP than with 4RIF (84.9% vs 65.4%; relative risk 1.30, 95% CI 1.22–1.38; P<0.001). Adverse events of any grade occurred in 24.5% of participants receiving 3HP and 20.2% receiving 4RIF, with no treatment-related deaths. Overall, weekly 3HP supported by SMS reminders achieved higher treatment completion than daily 4RIF with similar safety, supporting broader implementation of 3HP. |
Moderate risk: The open-label design and self-reported or self-administered dosing could introduce performance and measurement bias in adherence. Only the 3HP group received SMS reminders, so the effect of the regimen cannot be separated from the effect of the reminders, and the trial was not powered to compare efficacy in preventing tuberculosis disease. |
|
P: 1035 adults mechanically ventilated for ≥48 hours in six medical ICUs of a single tertiary-care academic hospital in Bhubaneswar, India (10,024 ventilator-days) I: Standardised multicomponent oral-care bundle plus 0.12% chlorhexidine (n=398) C: The same oral-care bundle without chlorhexidine (n=637) O: Ventilator-associated pneumonia (VAP) incidence density (CPIS >6) per 1000 ventilator-days; 28-day mortality, duration of ventilation, ICU length of stay, ventilator-free days, oral health and pain scores |
VAP occurred at 16.9 per 1000 ventilator-days without chlorhexidine and 18.8 per 1000 with chlorhexidine (adjusted incidence rate ratio 1.08, 95% CI 0.84–1.38; p=0.56). Twenty-eight-day mortality was 28.7% vs 26.4% (adjusted OR 0.87, 95% CI 0.65–1.15), ICU length of stay was longer with chlorhexidine (adjusted mean difference +0.89 days, 95% CI 0.19–1.58), and oral health and pain scores did not differ. Overall, adding 0.12% chlorhexidine to a standardised oral-care bundle did not reduce VAP or mortality and provided no clear incremental benefit in a mixed medical ICU population. |
Moderate risk: With only six ICUs randomised in a crossover design, the number of clusters is small, and care teams could not be blinded to chlorhexidine use, which could introduce performance bias. VAP was defined using the Clinical Pulmonary Infection Score, which has subjective components, the groups were unequal in size, and the single-centre setting may limit generalisability. |
Target Trial Emulation
Trial Protocols / Trial Ideas
Antibiotic Therapy Reviews
PK/PD and Drug Dosing
Antibiotics - In Vitro Susceptibility
Beta-Lactamases and Other Resistance Mechanisms
Adverse Effects of Antimicrobial Agents
Antibiotic Stewardship and Hospital in the Home
Infection Prevention / Antibiotic Prophylaxis
Bloodstream Infections and Endocarditis
Gastrointestinal Tract Infections
Bone and Joint Infections
Sexually Transmitted Infections
CNS Infections
Respiratory Tract Infections
Urinary Tract Infections
Infections in Children
Mycobacterial Infections
Fungal Infections and Antifungal Agents
Virulence
Diagnostics
Improving Clinical Research
General Interest
Copyright © ADVANCE-ID. All rights reserved.
ADVANCE-ID
Saw Swee Hock School of Public Health
National University of Singapore
Tahir Foundation Building
National University of Singapore
12 Science Drive 2, #10-01, Singapore 117549
+65 6516 4988
