A Large Multisite Cohort Study

[Posted 6/Jun/2022]

AUDIENCE: General Surgery, Ob/Gyn

KEY FINDINGS: Although the risk for uterine perforation with IUD insertion 4 days to 6 weeks or less post partum is nearly seven times that of insertion non-post partum, perforation remains an incredibly rare event for all clinical time points. Despite a slight increased risk of perforation with breastfeeding at IUD insertion, the benefits of breastfeeding and effective contraception generally outweigh risks and should have little clinical impact. Therefore, IUD insertion timing should be based on individual desire for IUD contraception and patient convenience to assure an IUD insertion can occur. Careful follow-up of individuals at higher risk of uterine perforation is warranted.

BACKGROUND: Reports of perforation risk related to intrauterine devices (IUDs) inserted immediately post partum and among non-post-partum individuals are scarce, and previous studies with only 12-month follow-ups underestimate the risk. Breastfeeding at IUD insertion and insertion within 36 weeks post partum have been associated with increased risk of uterine perforation. The aim of these analyses was to compare the incidence and risks of IUD-related uterine perforations by non-post-partum and post-partum intervals at IUD insertion, and among post-partum individuals, to assess the impact of breastfeeding on these outcomes.

DETAILS: Performed a multisite cohort study in the USA, using electronic health records (EHR). Study sites were three health-care systems and a site that used data from a health-care information exchange. The study population included individuals who were aged 50 years or younger and had an IUD insertion between Jan 1, 2001, and April 30, 2018. Individuals were excluded if they had not been in the health-care system for at least 12 months before IUD insertion. The primary outcome for this analysis was any IUD-related uterine perforation diagnosis for the first IUD insertion in this time period. Both complete and partial IUD-related perforations were identified. Chart abstraction was done to validate EHR-based algorithms or confirm perforations. The crude rate and cumulative incidence of uterine perforation were evaluated by non-post-partum and post-partum intervals at IUD insertion in the full cohort, and by breastfeeding status in a subcohort of post-partum individuals. Cox models estimated crude and adjusted hazard ratios (aHRs). Data from 326,658 individuals in the full cohort and 94,817 individuals in the post-partum subcohort were analysed. In the full cohort, we identified 1008 uterine perforations (51.2% complete), with the 5-year cumulative incidence being the lowest in the non-post-partum group (0.29%, 95% CI 0.26-0.34). The aHR for the post-partum interval relative to non-post partum ranged from 2.73 (95% CI 1.33-5.63; 0 to 3 days post partum) to 6.71 (4.80-9.38; 4 days to >=6 weeks post partum). The post-partum subcohort of individuals with breastfeeding information had 673 uterine perforations (62% complete), with a 5-year cumulative incidence of 1.37% (95% CI 1.24-1.52) and an increased risk with breastfeeding (aHR 1.37, 95% CI 1.12-1.66).

Copyright © Elsevier Ltd. All rights reserved.

Source: Reed, S. D., Zhou, X., Ichikawa, L., et al. (2022). Intrauterine Device-related Uterine Perforation Incidence And Risk (Apex-Iud): A Large Multisite Cohort Study. The Lancet. 2022; 399(10341): 2103-2112. Published: June 4, 2022. DOI: 10.1016/S0140-6736(22)00015-0.

A Randomized Clinical Trial

[Posted 20/Jul/2026]

AUDIENCE: General Surgery, Oncology

KEY FINDINGS: This randomized clinical trial demonstrated that single-encounter AR-guided localization was noninferior to conventional CT-guided localization for achieving successful sublobar resection of suspected early-stage lung cancer. In addition to maintaining localization accuracy, the AR-guided approach reduced radiation exposure, procedural pain, and workflow time, supporting its potential as an alternative localization strategy for minimally invasive lung surgery.

BACKGROUND: Accurate preoperative localization is critical for successful sublobar resection of small pulmonary nodules suspicious for early-stage lung cancer. Conventional computed tomography (CT)-guided localization typically requires multiple encounters, including localization in the CT suite followed by transfer to the operating room, increasing patient discomfort, radiation exposure, and procedural complexity. This randomized clinical trial evaluated whether a single-encounter augmented reality (AR)-guided localization strategy could achieve surgical outcomes comparable to standard CT-guided localization while improving procedural efficiency and patient experience.

DETAILS: This multicenter, randomized, noninferiority trial was conducted at 5 centers in China between August 8, 2024, and September 30, 2025. Among 296 randomized patients, 270 were included in the modified intention-to-treat analysis, with 134 assigned to single-encounter AR-guided localization and 136 to conventional multiple-encounter CT-guided localization. Eligible participants had CT-detected pulmonary nodules suspicious for early-stage lung cancer requiring preoperative localization before sublobar resection. The primary endpoint was successful sublobar resection, defined as R0 resection with protocol-specified adequate surgical margins. Secondary outcomes included localization accuracy, radiation exposure, localization-related complications, patient-reported pain, and procedural efficiency. The median (IQR) patient age was 59 (50-67) years, and 172 participants (63.7%) were women. Successful sublobar resection was achieved in 132 of 134 AR-guided procedures (98.5%) and 135 of 136 CT-guided procedures (99.3%), yielding a risk difference of -0.8 percentage points (95% CI, -2.7 to 3.9) and meeting the prespecified criterion for noninferiority. Localization accuracy was similar between groups, with a median (IQR) localization error of 3.0 (0.0-5.0) in the AR group and 3.0 (2.0-6.0) in the CT group. Compared with CT guidance, AR-guided localization significantly reduced radiation exposure (456.50 [378.75-631.85] vs 1260.11 [1026.48-1544.53] mGy·cm; P<.001), preoperative pain (0 [0-0] vs 5 [4-6]; P<.001), puncture time (0.63 [0.50-0.83] vs 6.50 [5.00-8.75] minutes; P<.001), and localization-to-incision interval (2.00 [1.50-2.00] vs 33.50 [18.00-63.00] minutes; P<.001). Pneumothorax occurred in 40 of 136 CT-guided cases (29.4%), whereas no clinically significant complications associated with AR-guided localization were reported.

Copyright © Skyscape. All rights reserved.

Source: Song, Z., Wang, Z., Yao, H., et al. Single-Encounter Augmented Reality-Guided Localization for Resection of Suspected Early-Stage Lung Cancer: A Randomized Clinical Trial. JAMA Surgery. Published: July 8, 2026. DOI: 10.1001/jamasurg.2026.2516.

[Posted 10/Jul/2026]

AUDIENCE: Infectious Disease, Internal Medicine

KEY FINDINGS: Early initiation of a 5-day oral ensitrelvir regimen within 72 hours after symptom onset in an index patient significantly reduced the risk of developing Covid-19 among household contacts while maintaining a safety profile comparable to placebo. Clinical benefit was observed across major patient subgroups, including individuals at increased risk for severe disease, and treatment was associated with reduced household transmission. These findings support ensitrelvir as an effective postexposure prophylactic option for household contacts and suggest potential utility in other high-risk exposure settings where rapid outbreak control is needed.

BACKGROUND: Household transmission remains a major contributor to the spread of SARS-CoV-2, particularly among individuals at increased risk for severe Covid-19. Although vaccination and prior infection have reduced disease severity, waning immunity and emerging variants continue to sustain transmission. Previous trials evaluating oral antiviral agents for postexposure prophylaxis in household contacts have not demonstrated significant protection, highlighting the need for effective preventive therapies. This phase 3 trial evaluated whether oral ensitrelvir, a SARS-CoV-2 3C-like protease inhibitor, could prevent Covid-19 among household contacts exposed to an infected index patient.

DETAILS: This phase 3, double-blind, randomized, placebo-controlled trial was conducted between June 2023 and mid-September 2024 across the United States, Argentina, Japan, South Africa, and Vietnam. Eligible household contacts were 12 years of age or older, had a negative SARS-CoV-2 test at enrollment, and were randomized within 72 hours after symptom onset in the index patient. Participants received either ensitrelvir 375 mg on day 1 followed by 125 mg once daily on days 2-5 or matching placebo. The primary endpoint was laboratory-confirmed Covid-19 by day 10 in the modified intention-to-treat population, defined as RT-PCR positivity accompanied by at least one prespecified Covid-19 symptom lasting 48 hours or longer. Secondary endpoints included laboratory-confirmed SARS-CoV-2 infection regardless of symptoms, subgroup analyses, and safety outcomes. Overall, 2,387 household contacts were randomized, including 1,030 participants in the ensitrelvir group and 1,011 in the placebo group within the modified intention-to-treat population. The mean participant age was 42.4 years, 71.1% were enrolled within 48 hours after symptom onset in the index patient, and 37.0% had at least one risk factor for severe Covid-19. By day 10, laboratory-confirmed Covid-19 occurred in 2.9% of participants receiving ensitrelvir compared with 9.0% receiving placebo, corresponding to a risk ratio of 0.33 (95% CI, 0.22-0.49; P<0.001) and an approximate 67% relative risk reduction in the modified intention-to-treat population. In the intention-to-treat population, Covid-19 developed in 4.4% and 10.2% of participants, respectively (risk ratio, 0.43; 95% CI, 0.32-0.59; P<0.001). Laboratory-confirmed SARS-CoV-2 infection irrespective of symptoms was also lower with ensitrelvir (14.0% vs. 21.5%; risk ratio, 0.66; 95% CI, 0.55-0.79). Subgroup analyses demonstrated generally consistent efficacy across age groups and participants with risk factors for severe disease. Among participants with risk factors, Covid-19 developed in 2.4% of the ensitrelvir group compared with 9.9% of the placebo group. Ensitrelvir was also associated with a 34% relative reduction in household SARS-CoV-2 transmission. Adverse events occurred at similar frequencies in the ensitrelvir and placebo groups (15.1% vs. 15.5%), with serious adverse events reported in 0.2% of participants in each group. No Covid-19-related hospitalizations or deaths occurred. Transient reductions in high-density lipoprotein concentrations were observed with ensitrelvir but returned toward baseline by day 15 and were not associated with clinical events.

Source: Hayden, F. G., Shinkai, M., Clark, T. W., et al. Ensitrelvir for Covid-19 Postexposure Prophylaxis in Household Contacts. New England Journal of Medicine. 2026; 394(19): 1905-1915. Published: June 22, 2026. DOI: 10.1056/NEJMoa2509306.

[Posted 9/Jul/2026]

AUDIENCE: Gastroenterology, Oncology

KEY FINDINGS: This prospective randomized tandem study demonstrates that AI-assisted colonoscopy significantly improves overall polyp detection, with the greatest benefit observed for diminutive and Paris type 0-IIa lesions. The enhanced detection was achieved without increasing procedure withdrawal time, supporting the feasibility of real-time AI integration into routine colonoscopic practice. Although false-positive alerts occurred, they were infrequent and readily distinguishable by experienced endoscopists. Given the single-center design, relatively small sample size, and absence of adenoma detection rate analysis, larger multicenter studies are warranted to confirm the clinical impact of AI-assisted colonoscopy on colorectal cancer prevention.

BACKGROUND: Colonoscopy remains the standard procedure for detecting and removing colorectal polyps, thereby reducing the incidence of colorectal cancer (CRC). However, lesions may be overlooked because of operator-dependent factors and the subtle appearance of small or flat polyps. Artificial intelligence (AI)-based computer-aided detection systems have been developed to provide real-time assistance during colonoscopy, but clinical evidence supporting their effectiveness in routine practice remains limited. This prospective randomized study evaluated whether AI-assisted colonoscopy improves polyp detection compared with conventional colonoscopy in a real-world clinical setting.

DETAILS: This prospective, randomized tandem cohort study was conducted at the Endoscopy Center of Nanfang Hospital, China, between April 2019 and September 2019. A total of 150 patients aged 18-70 years underwent same-day back-to-back colonoscopies performed by two experienced endoscopists, each with experience exceeding 3,000 colonoscopies. Participants were randomly assigned to undergo either conventional colonoscopy or AI-assisted colonoscopy first, followed immediately by the alternate procedure. The AI system employed a convolutional neural network based on the YOLO architecture and operated during withdrawal to identify suspected polyps in real time. The primary endpoint was the polyp detection rate (PDR), while secondary outcomes included the total number of detected polyps, detection of diminutive polyps (<6 mm), detection according to Paris classification, withdrawal time, and false-positive findings. Baseline demographic characteristics, bowel preparation quality, and withdrawal times were comparable between study groups. Mean withdrawal time was 370.15 ± 31.44 seconds with conventional colonoscopy and 373.17 ± 33.37 seconds with AI-assisted colonoscopy (p = 0.102). AI assistance significantly improved the PDR from 34.0% to 38.7% (p < 0.001). The total number of detected polyps increased from 80 with conventional colonoscopy to 105 with AI-assisted colonoscopy (p = 0.020). The benefit was primarily attributable to enhanced detection of diminutive polyps, with 91 lesions identified using AI compared with 69 during conventional examination (p < 0.001). Detection rates for patients with at least one diminutive polyp also increased from 30.0% to 34.7% (p < 0.001). In contrast, detection of polyps measuring >=6 mm did not differ significantly (11 vs 14; p = 0.319). AI-assisted colonoscopy also improved detection of Paris type 0-IIa polyps, increasing the number detected from 61 to 87 (p = 0.010) and the proportion of patients with at least one such lesion from 26.0% to 32.0% (p < 0.001). The AI system generated 52 false-positive alerts, averaging 0.35 false positives per colonoscopy, most commonly due to feces and mucosal folds, without prolonging withdrawal time.

Source: Luo, Y., Zhang, Y., Liu, M., et al. Artificial Intelligence-Assisted Colonoscopy for Detection of Colon Polyps: a Prospective, Randomized Cohort Study. Journal of Gastrointestinal Surgery. Journal of Gastrointestinal Surgery. 2021; 25(8): 2011-2018. Published: June 22, 2026. DOI: 10.1007/s11605-020-04802-4.

Findings from the PRESIDE Double-Blind Randomized Controlled Trial

[Posted 7/Jul/2026]

AUDIENCE: Family Medicine, Psychiatry

KEY FINDINGS: In this pragmatic double-blind randomized controlled trial conducted in Australian primary care, pharmacogenomic-informed antidepressant prescribing did not improve depressive symptoms compared with prescribing guided by national therapeutic guidelines. Clinical outcomes, medication tolerability, adherence, and quality of life were similar between groups, while cost-effectiveness analyses did not support routine implementation of pharmacogenomic testing in this setting. These findings suggest that broad pharmacogenomic-guided antidepressant prescribing in general practice does not provide additional clinical benefit over guideline-based management, although targeted evaluation in carefully selected patient populations may warrant further investigation.

BACKGROUND: Pharmacogenomic testing has been proposed as a strategy to individualize antidepressant selection by identifying CYP2D6 and CYP2C19 variants that influence drug metabolism. Although previous studies have suggested potential clinical benefits, evidence from pragmatic primary care settings remains limited. The Pharmacogenomic-Informed Antidepressant Prescribing for Moderate-to-Severe Depressive Symptoms in Australian General Practice (PRESIDE) trial evaluated whether pharmacogenomic-guided prescribing improves depression outcomes compared with guideline-based prescribing in routine general practice.

DETAILS: PRESIDE was a multicenter, double-blind, randomized controlled trial conducted in Australian general practice between May 26, 2021, and September 28, 2023. Of 5185 patients approached, 552 were randomized, and 550 participants (275 per group) were included in the intention-to-treat analysis. Participants with moderate-to-severe depressive symptoms were assigned in a 1:1 ratio to receive antidepressant prescribing recommendations based either on pharmacogenomic testing combined with Australian Therapeutic Guidelines or on Australian Therapeutic Guidelines alone. Pharmacogenomic reports incorporated CYP2D6 and CYP2C19 metabolizer phenotypes derived from saliva-based genotyping. The primary endpoint was change in Patient Health Questionnaire-9 (PHQ-9) score from baseline to 12 weeks after the prescribing report was received. Secondary outcomes included remission, treatment response, antidepressant-related adverse effects, medication adherence, prescribing congruence, quality of life, and cost-effectiveness. A total of 479 participants (87%) completed the primary 12-week outcome assessment. Depressive symptoms improved over time in both groups. At 12 weeks, the adjusted between-group difference in PHQ-9 change was 0.90 (95% CI, 0.06-1.75; standardized mean difference 0.23 [95% CI, 0.02-0.45]; p=0.036), indicating a small but greater improvement in depressive symptoms in the guideline-based control group. No significant between-group differences were observed at 4, 8, or 26 weeks. Remission at 12 weeks occurred in 11% of participants receiving pharmacogenomic-guided prescribing compared with 18% in the control group, corresponding to an adjusted difference of -7.22% (95% CI, -13.25 to -1.19) and an odds ratio of 0.530 (95% CI, 0.311-0.903; p=0.020). Treatment response rates did not differ significantly between groups (29% vs 32%; OR 0.874; 95% CI, 0.581-1.315; p=0.518). Approximately two-thirds of participants had an actionable CYP2D6 or CYP2C19 phenotype, yet subgroup analyses demonstrated no differential treatment benefit based on genotype. Antidepressant adherence, side-effect burden, health-related quality of life, and health-care utilization were comparable between groups. Economic analyses indicated that pharmacogenomic-guided prescribing was more costly and less effective than standard guideline-based care, although differences in costs and quality-adjusted life years were not statistically significant. No grade 2-5 adverse events occurred, and only one grade 1 adverse event related to an administrative reporting error was documented.

Source: Saya, S., Chondros, P., Abela, A., et al. Pharmacogenomic-informed antidepressant prescribing for moderate-to-severe depressive symptoms in Australian general practice (PRESIDE): a double-blind, randomised controlled trial. The Lancet Primary Care. 2026; Published: June 25, 2026. DOI: 10.1016/j.lanprc.2026.100158

Sustained 12-Month Improvements in Sexual Function and Vaginal Health

[Posted 4/Jul/2026]

AUDIENCE: General Surgery, Ob/Gyn

KEY FINDINGS: This prospective study demonstrates that fractional CO2 laser therapy provides sustained improvements in sexual function, perceived vaginal tightness, vaginal mucosal health, pelvic floor muscle strength, and patient-reported satisfaction in women with mild to moderate vaginal relaxation syndrome. Clinical benefits persisted for up to 12 months following treatment and were observed across diverse patient subgroups. The procedure was well tolerated, with only mild, self-limited adverse events reported. While these findings support FxCO2 laser therapy as a promising non-surgical treatment option for VRS, confirmation through larger multicenter randomized controlled trials with longer follow-up remains necessary.

BACKGROUND: Vaginal relaxation syndrome (VRS) is a common gynecologic condition characterized by reduced vaginal support and perceived vaginal laxity, often resulting in impaired sexual function and diminished quality of life. Although pelvic floor muscle training and other conservative therapies are available, treatment adherence and effectiveness are often limited. Fractional carbon dioxide (FxCO2) laser therapy has emerged as a minimally invasive alternative; however, prospective evidence evaluating its long-term efficacy and safety remains limited. This study assessed the clinical effectiveness and safety profile of FxCO2 laser therapy in women with mild to moderate VRS over a 12-month follow-up period.

DETAILS: This prospective, self-controlled study enrolled 101 women with mild to moderate VRS between February 2024 and August 2024. Participants received three FxCO2 laser treatment sessions administered at 4-week intervals. Clinical assessments were performed at baseline and 1, 3, 6, and 12 months after the final treatment session. Primary endpoints included the Female Sexual Function Index (FSFI), Vaginal Laxity Questionnaire (VLQ), and Vaginal Health Index (VHI). Secondary outcomes included the Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire-12 (PISQ-12), pelvic floor muscle strength, Sexual Satisfaction Questionnaire (SSQ), Patient Global Impression of Improvement (PGI-I), and overall treatment satisfaction. Safety was evaluated by documenting treatment-related adverse events throughout follow-up. Subgroup analyses examined treatment responses according to parity, age, and urinary incontinence status. Fractional CO2 laser therapy produced significant improvements across all primary clinical outcomes. The mean FSFI score increased from 21.4 to 25.9, VLQ improved from 2.68 to approximately 3.0, and VHI increased from 17.0 to 19.25, with all improvements reaching p <0.001. Sexual quality-of-life also improved, as demonstrated by an increase in PISQ-12 scores from 36.0 to 39.7 (p <0.001). Pelvic floor muscle strength improved for both sustained and rapid contractions throughout follow-up, while SSQ scores also increased significantly at each assessment (all p <0.001). Patient-reported improvement reflected favorable outcomes, with a mean PGI-I score of 2.82 ± 0.57, and overall treatment satisfaction indicated that most participants were somewhat satisfied to satisfied with therapy. Treatment benefits remained evident through 12 months after the final laser session. Subgroup analyses demonstrated that clinical improvements were consistent regardless of parity, age, or the presence of urinary incontinence. No serious treatment-related adverse events occurred. Mild transient discomfort was reported in 4 of 101 patients (3.96%), including pelvic soreness, vaginal dryness, or burning sensation, all of which resolved spontaneously or with minimal symptomatic management. No vaginal or urinary tract infections, persistent pain, scarring, or long-term complications were observed.

Copyright © Wiley Periodicals LLC. All rights reserved.

Source: Li, Q., Meng, J., Wang, Y., et al. Fractional CO2 Laser Treatment for Female Vaginal Relaxation Syndrome: A Prospective Study. Lasers in Surgery and Medicine. 2026; 58(5): 402-411. Published: July, 2026. DOI: 10.1002/lsm.70128.