Issue: Newsletter 5 | September 15, 2025
My Clinical "Take" on a Recent Randomised Controlled Trial
Professor Trisha Peel from Monash University discusses the BARRIER-PROTECT trial
The BARRIER-PROTECT randomised trial tested the impact of iodine-impregnated adhesive drapes (versus no adhesive drape) on end-of-procedure pocket-swab positivity during repeat cardiac implantable electronic device (CIED procedures. CIED infections at one-year follow up were a secondary outcome. The swabs were collected from the open surgical wound immediately prior to wound closure: in the case of participants randomised to the control (no drape) arm, an iodine-impregnated drape was applied before swabbing to maintain blinding. Surgical skin antisepsis consisted of alcoholic-chlorhexidine and patients received antimicrobial prophylaxis in line with the approach outlined in the PADIT trial (Krahn, et al. doi:10.1016/j.jacc.2018.09.068). Antibiotic impregnated envelopes were not used.
The trial analysed 384 patients. Positive swab cultures were identified in 10 % (19 of 189) of patients allocated to the iodine impregnated drapes, compared with 20.5% (40/195) of patients allocated to the no-drape group (relative risk reduction 0.50; 95% CI, 0.24-0.75; P=0.005). Coagulase-negative staphylococci were isolated in 97% of end of procedures swab (57/59). In the control (no drape) arm, Staphylococcus aureus was isolated in 2 swabs. CIED infections were reported in 4 of 195 patients (1.9%) in the control (no-drape) arm: there were no infections iodine-impregnated arm. Two of the four infections isolated coagulase negative Staphylococcus, which the authors reported matched organism isolated in the end-of-procedure pocket swab. The authors concluded that the use of iodine-impregnated drapes might be a low-cost, simple intervention to reduce cardiac device-related infections.
Limitations include:
• Use of extended antibiotic prophylaxis, despite PADIT showing no benefit and inconsistency with guidelines.
• Potential contamination from applying the drape post-procedure in the control group was not addressed.
• CIED infection was a secondary outcome with low event rates.
• No details on species of coagulase-negative staphylococci or use of molecular techniques to confirm linkage between end-of-procedure swabs and infection isolates.
These results differ from previous studies. Earlier Cochrane reviews of adhesive drapes across surgery found no overall benefit with iodine impregnated drapes (Webster, et aldoi:10.1002/14651858.CD006353) underscoring the need for replication in different surgical groups and with adequate powering to examine the impact on surgical site infections. ROSSINI-2, a multi-arm, multi-stage (MAMS) trial led by researchers at the University of Birmingham (ClinicalTrials.gov ID NCT03838575) will be key to confirming whether the use of iodine-impregnated drapes translate into durable reductions in surgical site infections in real-world practice.
| Citation of Articles | PICO | Main Results | Risk of Bias |
|---|---|---|---|
| Hook EW 3rd, Dionne JA, Workowski K, McNeil CJ, Taylor SN, Batteiger TA, et al. One dose versus three doses of benzathine penicillin G in early syphilis. N Engl J Med. 2025;393(9):869–878. doi:10.1056/NEJMoa2401802 • Editorial: Barbee LA, Bachmann LH. Antiquated tools to fight an ancient foe. N Engl J Med. 2025;393(9):921–923. doi:10.1056/NEJMe2508410 |
P: 249 patients with early syphilis (97% men, 62% Black, 61% HIV-positive; 19% primary, 47% secondary, 33% early latent) I: Single intramuscular dose of 2.4 million units benzathine penicillin G C: Three weekly doses of 2.4 million units benzathine penicillin G O: Serologic response at 6 months (RPR seroreversion or ≥2 dilution decline); relapse/treatment failure |
At 6 months, serologic response was 76% in the single-dose group vs 70% in the three-dose group (difference –6%, 90% CI –15 to 3), meeting noninferiority criteria. No clinical relapses or treatment failures occurred. Results were consistent across HIV-positive and HIV-negative subgroups. Injection-site pain/tenderness was common in both groups (76% vs 85%). | Moderate risk: Open-label design may introduce performance or detection bias, though objective serologic outcomes reduce this risk. Randomization across multiple centers strengthens validity. Attrition bias appears minimal, but generalizability may be limited given the predominantly male and HIV-positive sample. |
| Hofmann R, James S, Sundqvist MO, Wärme J, Angerås O, Alfredsson J, et al. Helicobacter pylori screening after acute myocardial infarction: the cluster randomized crossover HELP-MI SWEDEHEART trial. JAMA. 2025 Sep 1; doi:10.1001/jama.2025.15047 | P: 18,466 patients with acute myocardial infarction across 35 Swedish hospitals (median age 71, 71% male). I: Routine H. pylori screening with urea breath test during hospitalization. C: Usual care without systematic screening. O: Upper gastrointestinal bleeding (primary); subgroup bleeding risk by anemia status. |
After median 1.9 years, upper GI bleeding occurred in 299 patients in the screening group (16.8 per 1000 person-years; 3-year cumulative hazard 4.1%) vs 336 patients in the control group (19.2 per 1000 person-years; 3-year cumulative hazard 4.6%). Rate ratio 0.90 (95% CI, 0.77–1.05; P=0.18), indicating no significant benefit. Subgroup analysis suggested lower bleeding risk in patients with mild anemia (RR 0.64, 95% CI 0.42–0.98) and moderate/severe anemia (RR 0.44, 95% CI 0.23–0.87), but no effect in those without anemia. | Moderate risk: Open-label, cluster-crossover design may introduce performance and detection bias, although registry-based outcome capture reduces this risk. Large sample size and balanced baseline characteristics strengthen validity. Subgroup findings risk type I error given non-adjusted analyses. Attrition bias minimal due to registry follow-up. |
| Metcalfe JZ, Weir IR, Scarsi KK, Mendoza-Ticona A, Pierre S, Hall L, et al. A 3-month clofazimine–rifapentine-containing regimen for drug-susceptible tuberculosis versus standard of care (Clo-Fast): a randomised, open-label, phase 2c clinical trial. Lancet Infect Dis. 2025 Sep 4; doi:10.1016/S1473-3099(25)00436-0 • Editorial Commentary: Noreña I, Mbeya B, Nalunjogi J, Chimbe O, Heinrich N. Shaping opportunities for future clinical trials in tuberculosis. Lancet Infect Dis. 2025 Sep 4; doi:10.1016/S1473-3099(25)00479-7 |
P: 104 adults with drug-susceptible pulmonary tuberculosis across 6 sites in 5 countries (79% male, 71% advanced disease, 29% HIV-positive). I: 3-month regimen of isoniazid–rifapentine–pyrazinamide–ethambutol–clofazimine (with 2-week clofazimine loading dose). C: Standard 6-month regimen of isoniazid–rifampicin–pyrazinamide–ethambutol followed by rifampicin–isoniazid. O: Primary – time to sputum culture negativity at 12 weeks; safety – grade ≥3 adverse events; secondary – composite unfavourable outcome by week 65. |
By week 12, culture conversion rates were similar: 89% in the 3-month group vs 90% in standard therapy (aHR 1.21, 90% CI 0.82–1.79; p=0.21). However, grade ≥3 adverse events were significantly more common with the experimental regimen (45% vs 16%; difference 30%, 90% CI 14–45; p=0.002). At week 65, unfavourable outcomes occurred in 52% vs 27% (p=0.049). The trial was stopped early for lack of efficacy and excess toxicity. | High risk: Open-label design introduces bias in safety reporting. Early termination reduces statistical power and increases uncertainty in efficacy estimates. Small sample size limits generalizability. Randomization and multicountry design strengthen internal validity, but differential adverse event rates raise concern for treatment tolerability and safety bias. |
| Khodair HA, Elmahdy HS, Zidan LK, et al. Assessment of the role of probiotics in prevention of ventilator-associated pneumonia in neonates. Eur J Pediatr. 2025;184:597. doi:10.1007/s00431-025-06380-6 | P: 80 full-term neonates requiring invasive mechanical ventilation >48 hours in a single NICU (Tanta University Hospitals, Egypt); randomized 1:1 (Probiotics n=40; Control n=40). I: Twice-daily lactic acid bacteria probiotic (1×10^9 CFU per dose) from day 1 until discharge, plus standard care. C: Standard care without probiotics. O: Incidence of ventilator-associated pneumonia (VAP); feeding intolerance, vomiting, abdominal distension; duration of mechanical ventilation; NICU length of stay. |
Probiotics reduced VAP compared with control (20% vs 47.5%; OR 0.28, 95% CI 0.10–0.75), lowered feeding intolerance (17.5% vs 44.0%; OR 0.26), vomiting (12.5% vs 40.0%; OR 0.21), and abdominal distension (10.0% vs 44.0%; OR 0.18), and shortened mechanical ventilation (mean difference –10 days, 95% CI –13.70 to –6.30) and NICU stay (mean difference –8 days, 95% CI –12.71 to –3.29). | Moderate risk: Single-center, small sample size, and likely open-label design may introduce performance and detection bias for clinically assessed outcomes; randomization reported but allocation concealment and blinding are unclear; outcome definitions (e.g., VAP) may be susceptible to diagnostic variability. |
| Gagnon DJ, Burkholder KM, Weissman AJ, Riker RR, Ryzhov S, May TL, et al. Ceftriaxone to prevent early-onset pneumonia in comatose patients after out-of-hospital cardiac arrest: a pilot randomized controlled trial and resistome assessment (PROTECT). Chest. 2025 Aug 28. doi:10.1016/j.chest.2025.08.007 | P: Comatose adult survivors of out-of-hospital cardiac arrest undergoing targeted temperature management without pneumonia at admission; randomized 1:1 (Ceftriaxone n=26; Placebo n=26); randomized N=53 with one withdrawal, analyzed N=52. I: Ceftriaxone 2 g IV every 12 hours for 3 days (prophylaxis). C: Matching placebo every 12 hours for 3 days, plus standard care. O: Early-onset pneumonia ≤4 days after intubation (blinded adjudication); open-label antibiotic use; acquisition of antibiotic resistance genes (ARGs) by metagenomic sequencing of rectal swabs pre/post study drug; safety (serious adverse events). |
Ceftriaxone prophylaxis reduced early-onset pneumonia numerically but not definitively (38% vs 69%; RR 0.57, 95% CI 0.21–1.001; p=0.05), decreased receipt of open-label antibiotics (54% vs 85%; RR 0.64, 95% CI 0.43–0.94), and—after adjustment for baseline resistome—reduced acquisition of ARGs to commonly used ICU antibiotics (IRR 0.30, 95% CI 0.13–0.70); no serious adverse drug effects were reported. Overall, the trial was underpowered for conclusive effects on pneumonia but suggests stewardship and resistome benefits. | Moderate risk: The randomized, placebo-controlled design with blinded outcome adjudication and minimal attrition supports internal validity; however, very small sample size and low enrollment limit precision and generalizability, and substantial concomitant open-label antibiotics could dilute or confound treatment effects despite blinding. |
| George CM, Sanvura P, Bisimwa JC, Endres K, Namunesha A, Felicien W, et al. Effects of a water, sanitation, and hygiene program on diarrhea, cholera, and child growth in the Democratic Republic of the Congo: a cluster-randomized controlled trial of the preventative intervention for cholera for 7 days (PICHA7 WASHmobile) mobile health program. Clin Infect Dis. 2025 Sep 11; doi:10.1093/cid/ciaf417 | P: 2334 participants from diarrhea patient households in urban Bukavu, DRC, randomized into 2 arms (1138 standard arm, 1196 PICHA7 arm). I: PICHA7 WASHmobile program – weekly voice and text mHealth messages plus quarterly in-person visits, in addition to standard government message. C: Standard government message on oral rehydration solution use and basic WASH message (single in-person visit). O: Primary – diarrhea prevalence in the past 2 weeks assessed monthly for 12 months; Safety – not specifically reported (no grade ≥3 adverse events mentioned); Secondary – healthcare visits for diarrhea, Vibrio cholerae serological response, diarrhea with rice water stool, child growth outcomes (stunting, underweight, wasting). |
Diarrhea prevalence was significantly lower in the PICHA7 arm (PR 0.39, 95% CI 0.32–0.48). Participants had reduced odds of healthcare visits for diarrhea (OR 0.44, 95% CI 0.25–0.77) and diarrhea with rice water stool (OR 0.48, 95% CI 0.27–0.86). PICHA7 arm showed lower V. cholerae IgA serological responses (coefficient –0.85, 95% CI –1.60 to –0.09). Children under 5 in the PICHA7 arm were less likely to be stunted at 12 months (52% vs 63%; OR 0.65, 95% CI 0.43–0.99). High adherence to WASH components was reported. | Moderate risk: The cluster-randomized design strengthens internal validity, and adherence was high. However, lack of blinding of participants and outcome assessors may introduce performance and detection bias. Self-reported diarrhea outcomes could be subject to recall bias, though objective serological and anthropometric measures support robustness of findings. |
| Song Z, Zhou L, Wang W, Lan C, Tang T, Xie J, et al. Rifasutenizol-based triple therapy versus bismuth plus clarithromycin-based triple therapy for first-line treatment of Helicobacter pylori infection in China (EVEREST-HP): a phase 3, multicentre, randomised, triple-dummy, double-blind, controlled, non-inferiority trial. Lancet Infect Dis. 2025 Sep 10; doi:10.1016/S1473-3099(25)00438-4 • Editorial Commentary: Chen J, Lu H. Rifasutenizol-based triple therapy for Helicobacter pylori infection. Lancet Infect Dis. 2025 Sep 10; doi:10.1016/S1473-3099(25)00504-3 |
P: 700 treatment-naive adults aged 18–65 years with confirmed H. pylori infection, randomized at 40 hospitals in China (RTT n=353, BCTT n=347). I: Rifasutenizol-based triple therapy (RTT: rifasutenizol 400 mg, amoxicillin 1 g, rabeprazole 20 mg; all BID for 14 days). C: Bismuth plus clarithromycin-based triple therapy (BCTT: bismuth potassium citrate 240 mg, clarithromycin 500 mg, amoxicillin 1 g, rabeprazole 20 mg; all BID for 14 days). O: Primary – H. pylori eradication rate at 4–6 weeks post-treatment ([13C]UBT, mITT population); Safety – grade and frequency of treatment-emergent adverse events, labs, vitals, ECG; Secondary – antimicrobial resistance profile, adverse event patterns by treatment group. |
In the modified intention-to-treat analysis, eradication rates were non-inferior for RTT vs BCTT (92.0% [95% CI 88.7–94.6] vs 87.9% [83.9–91.1]; absolute difference 4.2%, 95% CI –0.3 to 8.8). Clarithromycin resistance was common (41%), but all isolates were susceptible to rifasutenizol. Treatment-emergent adverse events were less frequent in RTT (37%) than BCTT (53%). The most frequent AEs were diarrhea (7%), nausea (6%), and dizziness (6%) in RTT vs taste perversion (36%), nausea (6%), and diarrhea (5%) in BCTT. Most AEs were mild or moderate, and no drug-related serious adverse events were reported. | Low risk: The trial was randomized, double-blind, and triple-dummy, minimizing performance and detection bias. Central randomization and stratification by site strengthen allocation concealment. Attrition was very low with nearly all randomized patients included in mITT and safety analyses. The industry funding may introduce potential bias in interpretation, though objective eradication outcomes reduce this risk. |
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
