Top POEMs of 2025

Oral and topical treatment of male partners of women with bacterial vaginosis reduces recurrence; NNT = 3 (StepUp)

Clinical question

Does treatment of the male partner reduce the likelihood of recurrent bacterial vaginosis in women?

Bottom line

In addition to treating women with BV, treating their male partners with oral and topical metronidazole decreased the likelihood of BV recurrence in the women (NNT = 3). Getting men to adhere to the regimen is key.

Reference

Vodstrcil LA, Plummer EL, Fairley CK, et al, for the StepUp Team. Male-partner treatment to prevent recurrence of bacterial vaginosis. N Engl J Med 2025;392(10):947-957.

Study design: Randomized controlled trial (nonblinded)

Funding source: Government

Allocation: Concealed

Setting: Outpatient (any)

Synopsis

Recurrence of bacterial vaginosis (BV) after treatment is common in women and is thought to be related to re-exposure to bacterial species associated with BV that are harbored by men. However, previous studies of treating male partners with oral metronidazole have had mixed results. These researchers identified 164 women,18 years and older with a regular male partner for at least 8 weeks, who had a diagnosis of BV based on having 3 or more Amsel criteria (vaginal pH > 4.5, fishy odor, clue cells on microscopy, and a homogeneous vaginal discharge) and a microscopic Nugent score of at least 4 out of 10. Of the original 164 women, 27 were lost to follow-up. All women were treated with oral metronidazole 400 mg twice daily — or, if contraindicated, intravaginal 2% clindamycin for 7 nights or 0.75% metronidazole gel for 5 nights. Women were asked to refer their partners, who were then enrolled in the study if they presented within one week. Male partners in the treatment group received oral metronidazole 400 mg twice daily for 7 days, as well as 2% clindamycin cream to be applied twice daily to the glans penis and upper shaft of the penis (under the foreskin, if present). At baseline, the average age of female participants was 29 years, the average age of male participants was 32 years, 80% of men were not circumcised, and women had a median of 3 previous diagnoses of BV. Groups were balanced, and the primary analysis was for the modified intention-to-treat population who took at least one dose of medication and had assessment for recurrence. At 12 weeks, recurrence had occurred significantly less often in the partner treatment group than in the control group (35% vs 63%; P < .05; number needed to treat [NNT] = 3.5). Results were similar for the full intention-to-treat population (n = 159) and the per-protocol population (n = 134). Only 47 of 80 men were at least 70% adherent to the protocol. The number of days to recurrence was also significantly longer in the partner treatment group than in the control group. Adverse events were what you'd expect after a week of metronidazole; only 4 men in the treatment group reported any redness or irritation of penile skin.

Mark H. Ebell, MD, MS
Professor
Michigan State University
East Lansing, MI

Anti-influenza drugs have no effect on mortality or hospitalization and reduce symptoms by 24 hours or less

Clinical question

How effective are anti-influenza drugs for patients with influenza?

Bottom line

We published the first meta-analysis of oseltamivir in 2013, which was followed a year later by a Cochrane review that found essentially the same thing as our initial study and this 2025 meta-analysis. These drugs do not reduce hospitalization or mortality and have only a small effect on the duration of symptoms.

Reference

Gao Y, Zhao Y, Liu M, et al. Antiviral medications for treatment of nonsevere influenza: A systematic review and network meta-analysis. JAMA Intern Med 2025 Jan 13. Epub ahead of print.

Study design: Meta-analysis (randomized controlled trials)

Funding source: Foundation

Setting: Outpatient (any)

Synopsis

This network meta-analysis was funded by the World Health Organization and performed by a team of experienced researchers in Canada and China. These researchers searched multiple databases for studies of influenza-specific antiviral medications in adults and children. They identified 73 unique randomized trials (19 were unpublished) for which clinical trial reports and registry data were used. Most studies were of adults and only included patients within 2 days of symptom onset. Although 8 drugs were studied, only oseltamivir (37 studies) and baloxavir (4 studies) are currently widely used in outpatients. Oseltamivir had no significant effect on mortality or hospitalization in low- or high-risk patients. Similarly, baloxavir had no significant effect on mortality or hospitalization in low- or high-risk patients. Lack of benefit was found in patients with both severe and nonsevere influenza. The authors of the meta-analysis state that baloxavir may reduce the risk of admission for high-risk patients, but this is based on a trend that had broad confidence intervals (16 fewer hospitalizations per 1000 high-risk patients; 95% CI 20 fewer - 4 more). The overall reduction in symptom duration was 1.02 days (.63 - 1.41) for baloxavir and 0.75 days (.57 - .93 days) for oseltamivir. On the harm side of the equation, oseltamivir was associated with significantly more treatment-related adverse events (28 more per 1000 patients; 12 - 48).

Mark H. Ebell, MD, MS
Professor
Michigan State University
East Lansing, MI

COVID-19 vaccines are still effective at reducing emergency department visits, hospitalizations, and (especially) mortality

Clinical question

In the era of Omicron, previous vaccinations, and previous infections, do COVID-19 vaccines still provide a clinically important benefit?

Bottom line

COVID-19 vaccines remained effective at reducing ED visits, hospitalizations, and deaths in this high-quality observational study, with vaccine effectiveness in the same ballpark of earlier randomized trials. Although the absolute benefit is smaller than at the beginning of the pandemic, the NNIs to prevent an event are similar to those seen for respiratory syncytial virus and influenza vaccines.

Reference

Cai M, Xie Y, Al-Aly Z. Association of 2024–2025 Covid-19 vaccine with Covid-19 outcomes in U.S. veterans. N Engl J Med 2025;393(16):1612-1623.

Study design: Cohort (prospective)

Funding source: Government

Setting: Outpatient (any)

Synopsis

These Veterans Affairs researchers identified 164,132 veterans who received both the COVID-19 and influenza vaccines on the same day between September 3, 2024, and December 31, 2024, and a second group of 131,839 veterans who only received the influenza vaccine during the same interval. Groups were quite similar at baseline (probably because they were all willing to get at least one vaccine) with regard to age (mean 72 years), sex (92% were male), tobacco use, frailty, and race. The study authors did additional adjustments using inverse probably weighting for demographics, comorbidities, health care use, and other factors known to be associated with vaccine uptake. This approach is known as a target trial emulation because it uses observational data and statistical adjustment to try to simulate a clinical trial. Almost all the COVID-19 vaccines were mRNA vaccines. Patients were followed up for up to 6 months after the date of vaccination to assess for the occurrence of COVID-19–associated emergency department (ED) visits, hospitalization, or death. The primary outcome was vaccine effectiveness, which is calculated as 1 minus relative risk. So, a relative risk of 0.35 for mortality = 65% vaccine effectiveness. Vaccine effectiveness was 29% for emergency department [ED] visits (95% CI 19% - 39%), 39% for hospitalization (22% - 54%), and 64% for mortality (95% CI 23% - 86%). Because these outcomes are now relatively rare for persons infected with COVID-19, the absolute differences and number needed to immunize [NNI] are less impressive. For hospitalization, the absolute risk reduction (ARR) was 18.3 fewer per 10,000 persons for ED visits (NNI = 546), 7.5 fewer per 10,000 persons for hospitalization (NNI = 1333), and 2.2 fewer per 10,000 for mortality (NNI = 4545). Vaccine effectiveness was consistent across subgroups by age, chronic diseases, and immunocompromise, although persons in these higher risk groups experienced greater absolute reductions in these outcomes and therefore a smaller NNI for each one.

Mark H. Ebell, MD, MS
Professor
Michigan State University
East Lansing, MI

No association between flu vaccine and most maternal or neonatal outcomes

Clinical question

What adverse effects are associated with influenza vaccine administered during pregnancy?

Bottom line

In this large cohort study conducted right in the thick of the COVID-19 pandemic, influenza vaccine administered during pregnancy was not associated with preeclampsia, antenatal bleeding, or other disorders in mothers or premature/low birth weight, congenital defects, or mortality in their offspring. There were very small increases in gestational diabetes in the mothers and lower respiratory infections in newborns, which may not be clinically relevant.

Reference

Lee H, Yoon D, Kim JH, et al. Association of influenza vaccination during pregnancy with health outcomes in mothers and children: A population-based cohort study. Clin Pharmacol Ther 2025;117(5):1381-1392.

Study design: Cohort (retrospective)

Funding source: Government

Setting: Population-based

Synopsis

This population-based cohort study was conducted using a mother-child linked health insurance database in South Korea. The authors identified 174,008 mothers who received any influenza vaccine between December 2019 and March 2022 and matched them with the same number of women who did not receive a vaccine. Similarly, they identified 53,344 children born between September 2020 and June 2021 whose mothers were vaccinated and matched them with children born to unvaccinated mothers. The investigators used the STROBE criteria for reporting observational studies. After adjusting for a wide range of factors that can affect outcomes, they found no association between influenza vaccination and a variety of pregnancy-related disorders, including preeclampsia; antenatal bleeding; and neurologic, vascular, blood, and lymphatic disorders. There was a slight increase in the risk of gestational diabetes (effect estimate 1.06; 95% CI 1.05 - 1.08) and postpartum hemorrhage (1.05; 1.01 - 1.08). In the infants, there were no increased childbirth risks, such as preterm/low birth weight, congenital malformations, or mortality. There was an increased risk of lower respiratory infection that just squeaked by as being statistically significant (1.06; 1.01 - 1.12).

Allen F. Shaughnessy, PharmD, MMedEd
Professor of Family Medicine
Tufts University
Boston, MA

Zoledronate given twice 5 years apart to women with normal bone density or osteopenia reduces fracture risk (NNT = 9 over 10 years)

Clinical question

In women with normal bone density or osteopenia, does zoledronate given at baseline and 5 years later reduce fracture risk over 10 years?

Bottom line

Zoledronate is normally given annually for the prevention of osteoporotic fractures in adults with elevated fracture risk. This study showed a clinically and statistically meaningful benefit in lower risk postmenopausal women even if just given once, with a bit more benefit if repeated 5 years later. The cost of generic zol ranges from US$40 to US$500 (goodrx.com as of January 27, 2025) and it must be given by infusion.

Reference

Bolland MJ, Nisa Z, Mellar A, et al. Fracture prevention with infrequent zoledronate in women 50 to 60 years of age. N Engl J Med 2025;392(3):239-248.

Study design: Randomized controlled trial (double-blinded)

Funding source: Government

Allocation: Concealed

Setting: Outpatient (any)

Synopsis

The authors of this study point out that most fracture in postmenopausal women occur in women without documented osteoporosis or a previous fracture. Women aged 50 to 60 years with a bone mineral density T-score at the lumbar spine, femoral neck, or total hip between 0 and -2.5 were recruited for this study; women with a previous spine or hip fracture were excluded. A value greater than -1 is considered normal; a value less than -2.5 is defined as osteoporosis. The women were randomized to receive (1) zoledronate (zol) 5 mg intravenous infusion at baseline plus another zol infusion at 5 years, (2) zol at baseline and placebo infusion at 5 years, or (3) placebo infusions at both times. It is not clear that allocation was concealed, although groups were balanced at baseline and analysis was by intention to treat. The mean age of participants was 56 years, 85% were of European ancestry, and the mean T-scores were -0.36 to -0.55. A total of 1054 women were randomized, and follow-up was 95% at 10 years. Approximately 85% of the participants adhered to the second placebo or zol infusion at 5 years. Radiographic vertebral fractures were observed less often in the 2 groups that received zol (6.3% zol-zol, 6.6% zol-placebo, and 11.1% placebo-placebo; relative risk [RR] 0.58; 95% CI .38 to .87 for both zol arms vs placebo; number needed to treat [NNT] = 21 - 22 to prevent one fracture over 10 years). Results for zol-zol were only a bit better than those for zol-placebo, with no statistically significant difference between groups. Women in the zol-placebo group had significantly fewer fractures (27.4% vs 35.3%; RR .77; .63 - .97; NNT = 13 over 10 years) and a trend toward fewer fragility fractures (22.2% vs 28.2%; RR .79; .61 - 1.02). Women who received zol for both infusions had a slightly greater benefit, again with fewer fractures overall (24.7% vs 35.3%; RR .70; .56 - .88; NNT = 9) and fewer fragility fractures (20.2% vs 28.2%; RR .72; .55 - .93; NNT = 12). With regard to important harms, no cases of osteonecrosis of the jaw and no atypical femoral fractures occurred in any group.

Mark H. Ebell, MD, MS
Professor
Michigan State University
East Lansing, MI

Topical vaginal estrogen is not associated with increased risk in breast cancer survivors

Clinical question

Is the use of vaginal estrogen in breast cancer survivors associated with increased recurrence or mortality?

Bottom line

This meta-analysis of observational studies of female breast cancer survivors showed no increase in incidence of breast cancer recurrence, breast cancer–specific mortality, or overall mortality with the use of topical vaginal estrogen compared with non-use. The use of topical vaginal estrogen was associated with better outcomes for breast cancer recurrence and overall mortality. Caution should be used in interpreting these results, however, because of a potential "healthy volunteer effect" among estrogen users.

Reference

Beste ME, Kaunitz AM, McKinney JA, Sanchez-Ramos L. Vaginal estrogen use in breast cancer survivors: A systematic review and meta-analysis of recurrence and mortality risks. Am J Obstet Gynecol 2025;232(3):262-270.

Study design: Meta-analysis (other)

Funding source: Unknown/not stated

Setting: Outpatient (any)

Synopsis

These authors conducted a systematic review and meta-analysis of 8 observational studies that compared outcomes with and without the use of vaginal estrogen among female breast cancer survivors. They assessed outcomes of breast cancer recurrence (local, contralateral breast, or metastasis), breast cancer–specific mortality, and all-cause mortality. The authors were unable to identify any randomized controlled trials on this topic. All included studies showed low risk of bias and were heterogeneous based on differences in regimens and duration. They excluded studies that used systemic hormonal regimens. There were too few studies to assess publication bias. They found that there was no association with increased incidence of any of the outcomes among breast cancer survivors who used topical vaginal estrogen: recurrence (520/4494 [11.6%] among users vs 3086/19,566 [15.8%] among non-users; odds ratio [OR] 0.48; 95% CI 0.23 - 0.98; P = .04), breast cancer–specific mortality (23.4% vs 11.8%) and all-cause mortality (17.2% vs 23.4%). The authors rightly caution that women who used topical estrogen were likely healthier overall, which can make their outcomes more favorable — the "heathy volunteer effect," as seen in early observational studies of hormone replacement therapy.

Linda Speer, MD
Professor and Chair
Department of Family Medicine
University of Toledo
Toledo, OH

Early postpartum insertion of long-acting reversible contraceptive is associated with fewer pregnancies

Clinical question

Does the insertion of a long-acting reversible contraceptive before postpartum hospital discharge reduce pregnancy rates more than delayed insertion?

Bottom line

This meta-analysis of RCTs showed that immediate (before postpartum discharge) insertion of LARC is associated with improved continuation rates and fewer pregnancies at 6 months. On subgroup analysis, implants but not IUDs were significant for reduced pregnancy rate. The expulsion rate of IUDs is higher among patients who had immediate insertion after vaginal but not cesarean delivery.

Reference

Provinciatto H, Meirelles Dias YJ, Abonizio Magdalena SL, et al. Immediate vs delayed postpartum insertion of long-acting reversible contraception methods: meta-analysis of randomized controlled trials. Am J Obstet Gynecol 2025;232(2):139-149.

Study design: Meta-analysis (randomized controlled trials)

Funding source: Unknown/not stated

Setting: Various (meta-analysis)

Synopsis

For this meta-analysis of randomized controlled trials (N = 24), the authors selected studies that compared insertion of long-acting reversible contraceptive (LARC) before postpartum hospital discharge with delayed insertion at 4 to 12 weeks after birth. They included studies with both vaginal and cesarean births, and both intrauterine devices and contraceptive implants. The authors assessed both risk of bias of individual studies and publication bias. Heterogeneity and possible publication bias were concerns. None of the studies were masked because of the nature of the interventions. The researchers included 13 studies that reported utilization rates at 6 months showing increased likelihood of utilization with immediate insertion (relative risk [RR] 1.23; 95% CI 1.09 - 1.37; P < .01). Four studies with 317 participants reported pregnancy within 6 months follow-up, which also favored immediate insertion (RR 0.16; 0.04 - 0.71; P = .02). On subgroup analysis, the reduction was found for implants (RR 0.11; 0.01 - 0.92; P = .04) but not intrauterine devices ([IUDs] RR 0.23; 0.03-1.89; P = .17). There was no association between any breastfeeding and immediate versus delayed insertion (RR 1.01; 0.89 - 1.15; P = .82; 12 studies), and no association with exclusive breastfeeding (RR 0.88; 0.66 - 1.117; P = .38; 4 studies). There was no association between strategies for serious adverse events (pelvic inflammatory disease, perforation) among the 6 studies that reported it. The expulsion rate for IUDs was higher in the immediate insertion group, but significant only in the subgroup of patients with vaginal delivery (RR 5.09; 2.04 - 12.73; P = .01).

Linda Speer, MD
Professor and Chair
Department of Family Medicine
University of Toledo
Toledo, OH

Early medication abortion prior to pregnancy confirmation is safe and effective (VEMA)

Clinical question

Is medication abortion prior to confirmation of an intrauterine pregnancy safe and effective?

Bottom line

Early medication abortion was as effective as delaying treatment until intrauterine pregnancy is confirmed with respect to complete abortion. The reason for failed abortion differed between the groups, with ongoing pregnancy more likely in the early group and surgical intervention needed more often in the standard care group.

Reference

Brandell K, Jar-Allah T, Reynolds-Wright J, et al, for the VEMA (Very Early Medication Abortion) Study Group. Randomized trial of very early medication abortion. N Engl J Med 2024;391(18):1685-1695.

Study design: Randomized controlled trial (nonblinded)

Funding source: Government + foundation

Allocation: Concealed

Setting: Outpatient (specialty)

Synopsis

These Swedish researchers led an international team that recruited adult women who were seeking medication abortion for a suspected but unconfirmed early pregnancy at 26 sites in 9 mostly European countries. To be included, the women had to have an empty sac or a sac-like structure without an embryonic pole or yolk sac on vaginal ultrasound. They were randomized to receive either early medication abortion immediately after enrollment, or standard medication abortion after confirmation of the pregnancy. Women in the standard care group had repeat ultrasounds at 7 days and, if necessary, at 14 days to confirm the pregnancy. If not confirmed at 14 days, the women were considered to have a pathologic pregnancy (eg, early pregnancy loss or ectopic pregnancy); those women were referred for care but were retained in the study. A total of 1504 women were randomized, groups were similar at baseline, and analysis was by intention to treat. The mean age was 29.6 years, mean body mass index was 24.9, and 60% of the women had experienced 3 or more pregnancies (including this one). It was the first pregnancy for 23% of the participants. The primary outcome was the rate of complete abortion, which was nearly identical at slightly more than 95% for both groups. Surgical intervention for incomplete abortion was less common in the early abortion group (1.8% vs 4.5%; NNT = 37). There was 1 fewer day of bleeding and satisfaction with care was higher in the early abortion group. Ongoing pregnancy was more common in the early abortion group (3.0% vs 0.1%; NNT = 34).

Mark H. Ebell, MD, MS
Professor
Michigan State University
East Lansing, MI

For post-stent patients with atherosclerotic coronary vascular disease who are taking an anticoagulant, adding aspirin worsens outcomes (AQUATIC)

Clinical question

For patients with atherosclerotic coronary vascular disease with a stent, who have high residual atherosclerotic risk and an indication for anticoagulation, is aspirin plus anticoagulation preferred over anticoagulation alone?

Bottom line

This excellent trial should close the door on adding supplemental aspirin for patients with ASCVD and atrial fibrillation who are taking an anticoagulant. Even in this very high-risk population, adding aspirin increased mortality (NNTH = 21) over anticoagulation alone.

Reference

Lemesle G, Didier R, Steg PG, et al, for the AQUATIC Trial Investigators. Aspirin in patients with chronic coronary syndrome receiving oral anticoagulation. N Engl J Med 2025;393(16):1578-1588.

Study design: Randomized controlled trial (double-blinded)

Funding source: Government

Allocation: Concealed

Setting: Outpatient (specialty)

Synopsis

These French investigators identified 872 patients with atherosclerotic coronary vascular disease (ASCVD), previous stent implantation (PCI) at least 6 months ago, and who also had an indication for anticoagulation (atrial fibrillation for 90%). They all also had to have a high residual atherothrombotic risk based on conditions like diabetes, diffuse multivessel disease, chronic kidney disease, previous stent thrombosis, or multiple or complicated stents. All patients received an anticoagulant chosen by their physician (62% were given apixaban, 25% rivaroxaban), and then they were randomized to receive aspirin 100 mg once daily or matching placebo. The randomization was stratified by type of anticoagulant, study site, and whether they were taking an antiplatelet agent at baseline. Analysis was by intention to treat and allocation was concealed. Groups were balanced at baseline with a mean age of 72 years, 85% were male, 72% had a history of myocardial infarction, and a median time since their PCI of 3 years. Enrollment was stopped early after 4 years (before reaching the target enrollment of 2000 patients) due to a significant excess of deaths in the aspirin group. The median follow-up at that time was 2.2 years. The composite of 6 adverse cardiovascular events occurred more often in the aspirin group (16.9% vs 12.1%: P = .02; number needed to treat to harm [NNTH] = 21), as were all-cause mortality (13.4% vs 8.4%; P = .01; NNTH = 20) and major bleeding (10.2% vs 3.4%; P < .001; NNTH = 15). None of the primary or secondary outcomes favored aspirin over placebo.

Mark H. Ebell, MD, MS
Professor
Michigan State University
East Lansing, MI

Discontinuing anticoagulant therapy after catheter ablation for atrial fibrillation is safe if atrial arrhythmia does not recur after 1 year (ALONE-AF)

Clinical question

Is it safe to discontinue oral anticoagulant therapy after one year in adults without recurrence of atrial arrhythmia following catheter ablation for atrial fibrillation?

Bottom line

This study found that discontinuing oral anticoagulant therapy in adults at intermediate or high risk of thromboembolism (CHA₂DS₂-VASc score 1 - 3) without recurrent atrial arrhythmia one year after catheter ablation for AF results in a lower risk of stroke, systemic embolism, and major bleeding compared with continuing anticoagulant therapy.

Reference

Kim D, Shim J, Choi EK, et al, for the ALONE-AF Investigators. Long-term anticoagulation discontinuation after catheter ablation for atrial fibrillation. The ALONE-AF randomized clinical trial. JAMA 2025;334(14):1246-1254.

Study design: Randomized controlled trial (single-blinded)

Funding source: Government + foundation + industry

Allocation: Concealed

Setting: Outpatient (any)

Synopsis

The risk of long-term thromboembolic outcomes following catheter ablation for atrial fibrillation (AF) remains uncertain. These investigators identified 840 adults, aged 19 to 90 years, previously treated for AF with catheter ablation. Eligibility criteria included only patients at intermediate or high risk of thromboembolism (defined as a CHA₂DS₂-VASc score of 1 or more for men or 2 or more for women) and without any recurrence of atrial arrhythmia for at least 1 year after ablation. Study patients randomly received assignment to either continue or discontinue oral anticoagulant therapy. Individuals masked to treatment group assignment assessed outcomes. Complete follow-up occurred for 88.7% of participants at 2 years.

Using both intention-to-treat and per-protocol analyses, the primary outcome of a composite of stroke, systemic embolism, and major bleeding occurred significantly less often in the discontinue oral anticoagulant group vs the continue oral anticoagulant group (0.3% vs 2.2%; number needed to treat [NNT] for discontinuing vs continuing anticoagulation = 53; 95% CI 29 - 333). In addition, both ischemic stroke and major bleeding individually occurred significantly less in the discontinue group.

David C. Slawson, MD
Professor
The University of Virginia School of Medicine
Charlotte, NC

After ablation for atrial fibrillation, left atrial appendage closure is superior to oral anticoagulation (OPTION)

Clinical question

For patients who have undergone catheter ablation for atrial fibrillation: Is left atrial appendage closure with a device superior to oral anticoagulation?

Bottom line

Following catheter ablation for AF, implantation of a device that occludes the left atrial appendage obviates the need for oral anticoagulation after 3 months and reduces the risk of nonmajor bleeding (NNT = 10) while being noninferior for mortality and vascular events. The device company estimates the out-of-pocket costs of device implantation for a Medicare patient at US$2600.

Reference

Wazni OM, Saliba WI, Nair DG, et al, for the OPTION Trial Investigators. Left atrial appendage closure after ablation for atrial fibrillation. N Engl J Med 2025;392(13):1277-1287.

Study design: Randomized controlled trial (single-blinded)

Funding source: Industry

Allocation: Concealed

Setting: Outpatient (specialty)

Synopsis

Left atrial appendage occlusion is an effective way to reduce the risk of stroke in patients with atrial fibrillation (AF), but has primarily been compared with warfarin. In this study, researchers identified 1600 adults who had undergone catheter ablation for AF and who had a CHA2DS2-VASc score of 2 or higher for men or 3 or higher for women, indicating a moderate or greater bleeding risk and therefore a candidate for oral anticoagulation. Patients were randomized to receive implantation of the left atrial appendage closure device (Watchman FLX) or oral anticoagulation chosen by their physician (59% apixaban, 27% rivaroxaban, 8.5% other, and 5% none). The device group received 3 months of oral anticoagulation plus aspirin, and then aspirin alone for the next 9 months (the aspirin dose is not given). Groups were balanced at baseline: mean age 69 years, 66% male, 40% persistent AF, and 60% paroxysmal AF. Analysis was by intention to treat. Patients were followed up for a total of 36 months, during which time 41% of participants required additional catheter ablation. At 36 months, there were significantly more clinically relevant major or nonmajor bleeds in the anticoagulation group (18.1% vs 8.5%; P < .001; number needed to treat [NNT] = 10 over 3 years). The left atrial appendage closure was noninferior to oral anticoagulation for the composite efficacy outcome of death, stroke, or systemic embolism (5.3% vs 5.8% for anticoagulation; P < .001 for noninferiority) and for major bleeding events (3.9% vs 5.0%; P < .001). Complications related to the device occurred in 23 patients (3.0%).

Mark H. Ebell, MD, MS
Professor
Michigan State University
East Lansing, MI

For patients with provoked venous thromboembolism and at least one ongoing risk factor, 12 months of apixaban is better than 3 months

Clinical question

For patients with a provoked episode of venous thromboembolism (VTE) and at least one ongoing risk factor for VTE, is 12 months of apixaban better than 3 months?

Bottom line

For patients with a provoked VTE and at least one ongoing risk factor (obesity, chronic lung disease, or chronic inflammatory disease), 12 months of apixaban 2.5 mg twice daily reduced recurrent VTE compared with only 3 months of treatment (NNT = 11).

Reference

Piazza G, Bikdeli B, Pandey AK, et al. Apixaban for extended treatment of provoked venous thromboembolism. N Engl J Med 2025;393(12):1166-1176.

Study design: Randomized controlled trial (double-blinded)

Funding source: Industry

Setting: Outpatient (any)

Synopsis

Current guidelines recommend 3 months of anticoagulation with a direct oral anticoagulant (DOA) for patients with a provoked VTE, defined as a VTE occurring in association with major surgery, trauma, acute medical illness, or immobility. The safety profile and convenience of DOAs such as apixaban is more favorable than warfarin, so it is possible that a longer period of anticoagulation may be beneficial. These US industry–sponsored researchers identified 600 adults with an episode of VTE provoked by major surgery, acute medical illness, or immobility. The patients also had at least one persistent risk factor for VTE, such as body mass index of at least 30, chronic inflammatory disease, or chronic lung disease. Patients had been anticoagulated for at least 3 months and were randomized to receive apixaban 2.5 mg twice daily or matching placebo for a total of 12 months. Groups were balanced at baseline with a mean age of 59 years, 58% were female, 48% had only deep vein thrombosis (DVT), 23% had only pulmonary embolism (PE), and 29% had both. At one year, the researchers had an amazing 100% follow-up. The likelihood of symptomatic recurrent VTE was significantly lower in the apixaban group (1.3% vs 10.0%; P < .001; number needed to treat [NNT] = 11). There were reductions in both DVT (4 vs 23) and PE (0 vs 11) in the apixaban group. With regard to harm, major bleeding occurred in 1 patient in the apixaban group and none in the placebo group; clinically relevant nonmajor bleeding was numerically more common with apixaban (4.8% vs 1.7%; P = .06).

Mark H. Ebell, MD, MS
Professor
Michigan State University
East Lansing, MI

Clopidogrel is slightly more effective than aspirin for the secondary prevention of cardiovascular and cerebrovascular events

Clinical question

Is clopidogrel more effective than aspirin for the secondary prevention of cardiovascular and cerebrovascular events?

Bottom line

Based on this analysis of pooled individual participant data, clopidogrel is slightly more effective than aspirin in the secondary prevention of major cardiovascular and cerebrovascular events without increasing the risk of major bleeding. The main difference was in lower rates of subsequent myocardial infarction and stroke of any etiology.

Reference

Valgimigli M, Choi KH, Giacoppo D, et al. Clopidogrel versus aspirin for secondary prevention of coronary artery disease: A systematic review and individual patient data meta-analysis. Lancet 2025;406(10508):1091-1102.

Study design: Meta-analysis (randomized controlled trials)

Funding source: Government

Setting: Various (meta-analysis)

Synopsis

These authors searched 4 databases and the websites of various cardiology societies to identify published randomized trials that directly compared clopidogrel with aspirin monotherapy in adults with established coronary artery disease. The participants could be antiplatelet therapy naïve, have recently discontinued dual antiplatelet therapy (DAPT), or have undergone DAPT run-in before randomization. The authors pooled the individual participant data to evaluate benefits and harms. The main outcome of interest was major adverse cardiovascular and cerebrovascular events (MACCE): cardiovascular death, myocardial infarction, or stroke. The main harm was major bleeding. The authors identified 7 trials (N = 28,982) with a median of 2.3 years of follow-up. Four studies were at low risk of bias, and the others had some concerns about protocol deviations. Using intention-to-treat analysis, the authors report that the overall rate of MACCE was lower in the clopidogrel-treated participants (2.6 vs 2.99 per 100 person-years; number needed to treat [NNT] = 257) and they found no significant difference in the rate of major bleeding (0.71 vs 0.77 per 100 person-years). They also found no significant difference in death from any cause or in cardiovascular death (2 and 1.2 per 100 person-years, respectively). For the individual MACCE outcomes, clopidogrel was more effective than aspirin only for subsequent myocardial infarctions (0.99 vs 1.27 per 100 person-years; NNT = 358) and for stroke of any etiology (0.73 vs 0.88 per 100 person-years; NNT = 667). Although the authors detected no heterogeneity among the MACCE data, they found significant heterogeneity among the data on major bleeding (I² = 50.7%). They also did a bunch of fancy statistics to look at event rates over time, and these supported the main findings. According to GoodRx, a 90-day supply of 75 mg clopidogrel costs between US$14 and US$31 and a 90-day supply of 81 mg aspirin costs between US$5 and US$12.

Henry C. Barry, MD, MS
Professor
Michigan State University
East Lansing, MI

Intensive blood pressure lowering decreases cardiovascular events (NNT = 58) and increases hypotension and syncope (NNTH = 55)

Clinical question

What are the benefits and harms from intensive blood pressure lowering?

Bottom line

The authors of this study report that the cardiovascular benefits of intensive therapy outweigh the harms. A heat map in the appendix of the paper can be used to guide individual decision-making.

Reference

Guo X, Sun G, Xu Y, et al, for the BPRULE Study Group. Benefit-harm trade-offs of intensive blood pressure control versus standard blood pressure control on cardiovascular and renal outcomes: An individual participant data analysis of randomised controlled trials. Lancet 2025;406(10507):1009-1019.

Study design: Meta-analysis (randomized controlled trials)

Funding source: Government

Setting: Various (meta-analysis)

Synopsis

These researchers pooled the individual participant data from 6 large randomized trials (N = 80,220) that compared intensive blood pressure lowering (< 120 mm Hg or < 130 mm Hg) with standard lowering and reported cardiovascular outcomes. The authors used a composite of myocardial infarction, stroke, heart failure, and cardiovascular death (we really dislike composite outcomes). This outcome was variably defined in the individual trials, causing some heterogeneity in the data; the researchers standardized these measures to take this into account. They also estimated the frequency of adverse events: hypotension, syncope, injurious fall, arrhythmia, angioedema, and various measures of renal function. On average, the participants were 64 years of age, overweight (body mass index 26.9 kg/m²), and had baseline blood pressure of 148/83. Slightly more than half were female (51%), more than 80% were Asian, and 35% had at least a high school education. The median follow-up was 3.2 years, during which 4969 participants had one of the cardiovascular events and 5056 experienced one of the adverse events. The authors don’t report data on deaths from all causes. Overall, the rate of cardiovascular events was lower in the intensively controlled participants than in the standard controlled participants (5.3% vs 7.1%; number needed to treat [NNT] 58; 95% CI 48 - 71) but harms occurred more often (7.2% vs 5.4%; number needed to treat to harm [NNTH] = 56; 47 - 68), especially in adults 65 years or older. The individual cardiovascular outcomes also occurred less frequently in the intensive therapy group (NNTs range from 88 to 370). The authors don’t report data on individual harms other than the renal outcomes. They also did multiple subgroup analyses, including analyses based on age thresholds (65 and 80 years of age) and the data on benefits and harms did not change appreciably. There is a heat map in the appendix that shows the net balance of benefits and harms for various subgroups. A recent study reported that higher doses and more medication results in greater degrees of blood pressure lowering. The authors of that paper provide an online calculator to estimate the effectiveness of different medication regimens.

Henry C. Barry, MD, MS
Professor
Michigan State University
East Lansing, MI

Good evidence that duloxetine, milnacipran, and pregabalin provide meaningful pain reduction in adults with fibromyalgia

Clinical question

What are the most effective drugs for providing meaningful pain relief in adults with fibromyalgia?

Bottom line

In this synthesis of Cochrane Reviews, the authors found moderate- to good-quality evidence that duloxetine, milnacipran, and pregabalin were each effective in meaningful pain relief at 4 to 12 weeks. Long-term effectiveness data are lacking.

Reference

Moore A, Bidonde J, Fisher E, et al. Effectiveness of pharmacological therapies for fibromyalgia syndrome in adults: an overview of Cochrane Reviews. Rheumatology (Oxford) 2025;64(5):2385-2394.

Study design: Meta-analysis (other)

Funding source: Government

Setting: Various (meta-analysis)

Synopsis

These authors did an umbrella review — a synthesis of other systematic reviews and meta-analyses — of 21 Cochrane Reviews of pharmacologic treatments of fibromyalgia in adults. To be included, the analyses had to address outcomes 3 months after starting treatment. Because of differences in outcomes, potential inclusion of both randomized and nonrandomized trials, and so forth, the authors chose not to pool data. Seven of the Cochrane reviews evaluated antidepressants, 9 evaluated anticonvulsants, and one each assessed antipsychotics, cannabinoids, oxycodone, nonsteroidal anti-inflammatory drugs (NSAIDs), and combination therapy. Overall, the reviews found no trials of carbamazepine, clonazepam, lamotrigine, phenytoin, oxycodone, topiramate, or valproate. The authors identified very limited evidence on the effectiveness of antipsychotics, cannabinoids, combination therapy, gabapentin, lacosamide, monoamine oxidase inhibitors, and NSAIDs. They also found significant publication bias for studies of amitriptyline and selective serotonin reuptake inhibitors. More helpful, however, was the discovery of moderate- to good-quality evidence that duloxetine, milnacipran, and pregabalin improved pain by at least 30% (range of numbers needed to treat [NNT] 6 – 10) and similarly for pain reduction at least 50% (NNTs from 7 – 14). Finally, patients taking duloxetine, milnacipran, or pregabalin were more likely than control patients to report meaningful global improvement (NNTs from 5 – 22). Sadly, the authors found no efficacy data beyond 6 months. Finally, the frequency of serious adverse effects was no different than with placebo.

Henry C. Barry, MD, MS
Professor
Michigan State University
East Lansing, MI

Amyloid-directed monoclonal antibodies provide no clinically meaningful cognitive benefits at the risk of clinically important harms

Clinical question

Are amyloid-directed monoclonal antibodies effective and safe in improving cognition in adults with mild cognitive impairment or Alzheimer disease?

Bottom line

This meta-analysis demonstrates (as do previous studies) that amyloid-directed monoclonal antibodies have tiny, clinically insignificant effects on various scales that measure cognition. These minimal “improvements” come at great physical and economic costs.

Reference

Tonegawa-Kuji R, Hou Y, Hu B, et al. Efficacy and safety of passive immunotherapies targeting amyloid beta in Alzheimer's disease: A systematic review and meta-analysis. PLoS Med 2025;22(3):e1004568.

Study design: Meta-analysis (randomized controlled trials)

Funding source: Government

Setting: Various (meta-analysis)

Synopsis

These researchers searched PubMed, Embase, and clinicaltrials.gov to identify high-quality (Jadad score > 3) phase III randomized trials with more than 200 participants that compared placebo with monoclonal antibodies against amyloid in adults with mild cognitive impairment or Alzheimer disease. Although they don’t describe other efforts at identifying the gray literature, the authors graphically and statistically assessed the potential for publication bias. They report the potential for publication bias with the Clinical Dementia Rating Scale Sum of Boxes, which is one of the research scales that measures cognition, and they report that small studies that favored antibodies were potentially missing. Overall, the authors found that on various scales of cognition monoclonal antibodies were superior to placebo, but that the effect sizes were tiny and the data heterogeneous. While there was no difference in all-cause mortality, the antibodies caused more adverse events than placebo: amyloid imaging abnormalities (ARIA) with edema (16.6% vs 1.4%; number needed to treat to harm [NNTH] = 7) and with hemorrhage (18.4% vs 9.2%; NNTH = 10), and headaches (12.8% vs 9.9%; NNTH = 54). These findings are similar to another analysis published more than a year ago that was co-authored by 2 Essential Evidence Plus authors.

Henry C. Barry, MD, MS
Professor
Michigan State University
East Lansing, MI

Bright light therapy is effective for nonseasonal depressive disorders

Clinical question

Is bright light therapy effective treatment for adults with nonseasonal depressive disorders?

Bottom line

A previous review found moderate-quality evidence that supports the effectiveness of BLT for the treatment of nonseasonal depressive disorders. This updated meta-analysis of 11 individual trials published since 2000 found additional evidence to support the benefit of BLT as an effective treatment/adjunctive treatment for adults with nonseasonal depressive disorders.

Reference

Menegaz de Almeida A, Aquino de Moraes FC, Cavalcanti Souza ME, et al. Bright light therapy for nonseasonal depressive disorders. A systematic review and meta-analysis. JAMA Psychiatry 2025;82(1):38-46.

Study design: Meta-analysis (randomized controlled trials)

Funding source: Unknown/not stated

Setting: Various (meta-analysis)

Synopsis

These investigators thoroughly searched multiple databases, including MEDLINE, EMBASE, SCOPUS, Web of Science, PsychINFO, and the Cochrane Register, for English language–only randomized trials that compared bright light therapy (BLT) alone or BLT plus antidepressant with placebo, antidepressant monotherapy, or dim red light in adults with nonseasonal depression. BLT was defined as using a fluorescent light box that produces 10,000 lux white light for at least 30 minutes daily. Two individuals independently assessed individual trials for inclusion criteria and risk of bias using the Cochrane Risk of Bias tool. Disagreements were resolved by consensus discussion with a third reviewer. All 11 trials (N = 858 patients) that met inclusion criteria were scored at low risk of bias.

Statistically significant higher remission and response rates occurred in the BLT group than in the non-BLT group (40.7% vs 23.5%, and 60.4% vs 38.6%, respectively). A subgroup analysis found a similar benefit to BLT for improving both remission and response rates in trials lasting less than 4 weeks and in trials lasting more than 4 weeks. Additional analyses found minimal heterogeneity among the individual trials and a low probability of publication bias.

David C. Slawson, MD
Professor
The University of Virginia School of Medicine
Charlotte, NC

Meta-analysis of platelet-rich plasma injections for knee osteoarthritis: High-quality studies find no meaningful pain improvement

Clinical question

Are platelet-rich plasma injections effective in improving pain or function in adults with degenerative joint disease of the knee?

Bottom line

This meta-analysis of randomized trials finds that most studies of PRP injections in adults with knee DJD are not of high methodologic quality. Although, overall, the studies suggest that PRP injections improve pain and function, the high-quality studies demonstrated no significant pain reduction and showed improved function only after 12 months of follow-up. Finally, the authors provide no data on adverse events.

Reference

Bensa A, Previtali D, Sangiorgio A, Boffa A, Salerno M, Filardo G. PRP injections for the treatment of knee osteoarthritis: The improvement is clinically significant and influenced by platelet concentration: A meta-analysis of randomized controlled trials. Am J Sports Med 2025;53(3):745-754.

Study design: Meta-analysis (randomized controlled trials)

Funding source: Unknown/not stated

Setting: Various (meta-analysis)

Synopsis

These authors searched multiple databases and registries to identify randomized trials of platelet-rich plasma (PRP) injections versus placebo in treating adults with knee degenerative joint disease (DJD). Other than the registry search, the authors don’t describe other efforts at searching the gray literature nor do they report the potential impact of publication bias. Ultimately, they included 18 trials (1995 participants); only 7 of which (39%) were at low risk of bias. The main study outcomes (pain on a visual analog scale and the Western Ontario McMaster Universities Osteoarthritis Index [WOMAC]) were reported at 1, 3, 6, and 12 months of follow-up. The authors report the minimum clinically important difference on each of these scales these is 1.37 and 6.4, respectively. Overall, PRP resulted in clinically meaningful improvements in the VAS score at 3 and 6 months and in the WOMAC score at each interval. The authors report many subgroup analyses — the most important of which is buried in the appendix, which might be a reflection of their intellectual conflict of interest. Among the studies at low risk of bias, there was no statistically significant difference in pain at any follow-up interval and the WOMAC was significantly improved only after 12 months. Since biased studies tend to falsely overestimate benefits, the overall data are dubious. Finally, although the authors report that they extracted data on adverse effects, they don't report any.

Henry C. Barry, MD, MS
Professor
Michigan State University
East Lansing, MI

Water exposure at 6 hours after cutaneous surgery does not increase the risk of complications

Clinical question

Is water exposure safe at 6 hours after cutaneous surgical treatment of benign and malignant lesions, including standard excisional surgery and Mohs surgery with immediate reconstruction?

Bottom line

This study found that early water exposure, as soon as 6 hours after cutaneous surgery of dermatologic lesions, did not increase the risk of infection or bleeding compared with standard instructions to keep wound dressing dry for 48 hours. In addition, scar quality at 6 months was similar in both treatment groups.

Reference

Samaan C, Kim Y, Zhou S, Kirby JS, Cartee TV. Early postoperative water exposure does not increase complications in cutaneous surgeries: A randomized, investigator-blinded, controlled trial. J Am Acad Dermatol 2024;91(5):896-903.

Study design: Randomized controlled trial (single-blinded)

Funding source: Foundation

Allocation: Concealed

Setting: Outpatient (specialty)

Synopsis

Patients are traditionally advised to keep initial postoperative dressings dry for 24 to 72 hours following cutaneous dermatologic surgery. These investigators identified adults,18 years or older, who were undergoing surgical treatment of benign and malignant lesions, including standard excisional surgery or Mohs micrographic surgery with immediate reconstruction as needed. Eligible participants (N = 437) randomly received postoperative instructions for early water exposure (ie, remove the dressing after 6 hours and wet the wound for at least 10 minutes in a shower, bath, or pool) or standard advice to keep the dressing dry for 48 hours. The individuals who assessed all outcomes remained masked to treatment group assignment. Complete follow-up occurred for 100% of patients at 14 days. A smaller subset of participants (n = 146) was asked to return at 6 months.

Using intention-to-treat analysis, the rate of postoperative infection, defined by a positive wound culture was similar in both the standard and early water exposure groups (1.4% vs 1.8%, respectively). Similarly, no significant group differences occurred in the rate of bruising or hematoma. At the 6-month follow-up, scar quality as measured by a standard scoring tool was similar in both groups.

David C. Slawson, MD
Professor
The University of Virginia School of Medicine
Charlotte, NC

Oral challenge safe with risk assessment tool to remove erroneous sulfonamide allergy label

Clinical question

Is an oral provocation challenge safe and accessible for distinguishing an erroneous sulfonamide allergy label from a true one in adults?

Bottom line

This study reports the outcome of a trimethoprim/sulfamethoxazole oral provocation challenge in 104 adults who were initially labeled as having a sulfonamide allergy and subsequently identified by their history to be at low risk of a true allergy. A total of 100 patients (96%) completed the oral provocation challenge with no reaction, resulting in the removal of the erroneous allergy label. Four patients had reactions that were all mild and required minimal intervention (ie, oral antihistamine/ibuprofen/acetaminophen and no epinephrine). As with the penicillin allergy risk tool, primary care clinicians can use this method to identify the millions of adults who are incorrectly labeled as having a sulfonamide allergy.

Reference

Stehlin F, Vogrin S, Mitri E, Isabwe GAC, Trubiano JA, Copaescu AM. International validation of the SULF-FAST risk-stratification tool for sulfonamide antibiotic allergy. JAMA Network Open 2025;8(7):e2519113.

Study design: Cohort (retrospective)

Funding source: Self-funded or unfunded

Setting: Outpatient (any)

Synopsis

These investigators assessed the accuracy of the SULF-FAST risk-stratification tool (applying the same questions used for assessing a true penicillin allergy from an erroneous one) in 125 adults initially labeled as having a sulfonamide allergy. Of these, 104 with a SULF-FAST score of less than 3 (considered low risk) were given an oral challenge of trimethoprim 80 mg or 160 mg/sulfamethoxazole 400 mg or 800 mg. A mild reaction (ie, skin rash, joint pain, transient fever) occurred in 4 patients (4%) requiring the administration of oral antihistamine and/or ibuprofen/acetaminophen. The remaining 100 patients (96%) had the sulfonamide allergy label removed from their medical record.

David C. Slawson, MD
Professor
The University of Virginia School of Medicine
Charlotte, NC

Guidelines suggest new diabetes medicine choice based on cardiovascular and chronic kidney disease risk

Clinical question

What are the benefits and harms of medications for adults with type 2 diabetes at varied risks of cardiovascular and kidney-related complications?

Bottom line

This guideline, from an international group unattached to any professional guild, continues the march away from the blood-glucose-control-is-everything approach to diabetes and suggests using the newer classes of medicines for type 2 diabetes solely based on CVD and CKD risk (see also this 12 December 2025 POEM Research Summary). The guidelines recommend against the use of an SGLT-2 inhibitor or a GLP-1 agonist in patients at low risk, consider them in patients at moderate risk, and strongly recommend them for patients at high risk. The MATCH-IT tool provides a quick visual summary of the recommendations. The group offers a caveat that these guidelines were based on clinical trials that enrolled participants with reasonably controlled diabetes (Hb A1c < 8%) and may not be applicable to people at the extremes.

Reference

Agarwal A, Mustafa R, Manja V, et al. Cardiovascular, kidney related, and weight loss effects of therapeutics for type 2 diabetes: A living clinical practice guideline. BMJ 2025;390:e082071.

Study design: Practice guideline

Funding source: Government + foundation

Setting: Various (guideline)

Synopsis

This guideline was produced as part of the BMJ Rapid Recommendations series. The international working group comprised patients, clinicians, and methodologists and used the GRADE approach to balance benefits and harms when making recommendations. The guideline is based on a systematic review and network meta-analysis. The guideline moves away from blood glucose control as a raison d’être and suggests selecting whether to use new medications based on baseline risk. In obese people with type 2 diabetes, the working group recommends the glucagon-like peptide-1 (GLP-1) receptor agonist tirzepatide (Zepbound, Mounjaro) across the board (weak recommendation). For people with diabetes at lower risk — for example, those without cardiovascular disease (CVD) or chronic kidney disease (CKD) and 3 or fewer cardiovascular risk factors — the authors recommend against routine use of the newer diabetes medicines (weak recommendation). For people with established CVD or CKD or more than 3 risk factors, they recommend sodium-glucose cotransporter 2 (SGLT-2) inhibitors or GLP-1 receptor agonists (weak recommendation). For people at higher risk (with CVD and/or CKD and at higher risk of complications or with heart failure), they strongly recommend an SGLT-2 inhibitor or GLP-receptor agonist, along with a weak recommendation of finerenone.

Allen F. Shaughnessy, PharmD, MMedEd
Professor of Family Medicine
Tufts University
Boston, MA

American College of Physicians reaffirms de-emphasis on glucose control and adds new treatments for type 2 diabetes

Clinical question

What is the role of newer medicines for type 2 diabetes, according to the American College of Physicians?

Bottom line

Continuing to (perhaps) be ahead of the curve compared with other professional societies, the American College of Physicians published this guideline to reaffirm an HbA1c goal of 7% to 8% for most people with type 2 diabetes. Rather than focusing solely on blood glucose control, they also suggest a quick addition of newer treatments, an SGLT-2 inhibitor (eg, dapagliflozin, empagliflozin) or GLP-1 agonist (eg, liraglutide, semaglutide) for most patients because of their demonstrated benefit on morbidity and mortality, with the choice of treatment based on co-morbidity. The guideline recommends against using a DPP-4 inhibitor because of a lack of demonstrated benefit on morbidity and all-cause mortality.

Reference

Qaseem A, Obley AJ, Shamliyan T, et al, for the Clinical Guidelines Committee of the American College of Physicians. Newer pharmacologic treatments in adults with type 2 diabetes: A clinical guideline from the American College of Physicians. Ann Intern Med 2024;177(5):658-666.

Study design: Practice guideline

Funding source: Self-funded or unfunded

Setting: Various (guideline)

Synopsis

This guideline from the American College of Physicians updates their 2017 guideline for the treatment of patients with type 2 diabetes, considering the role of newer treatments that have demonstrated benefit on all-cause mortality and morbidity. The authors re-iterate that the goal of long-term blood glucose control is a hemoglobin A1c of 7% to 8%, with a specific number individualized to each patient based on their goals, co-morbidities, life expectancy, and risks associated with hypoglycemia. Glucose control should be relaxed in patients with an HbA1c of less than 6.5%. The guideline continues to recommend metformin as first-line therapy, with rapid addition of a sodium–glucose cotransporter-2 (SGLT-2) inhibitor or glucagon-like peptide-1 (GLP-1) agonist (strong recommendation, high-certainty evidence). Based on evidence of no benefit — or, perhaps, increased morbidity — the authors suggest against routine treatment with a dipeptidyl peptidase-4 (DPP-4) inhibitor (strong recommendation, high-certainty evidence). This guideline is based on a systematic review of recent literature by a working group without intellectual, professional, or financial conflicts of interest. The group included internists, an epidemiologist, and a panel of public members.

Allen F. Shaughnessy, PharmD, MMedEd
Professor of Family Medicine
Tufts University
Boston, MA

Updated guidelines from the Society of Critical Care Medicine strongly recommend steroids for hospitalized adults with severe community-acquired pneumonia

Clinical question

Should steroids be administered to hospitalized patients with sepsis, acute respiratory distress syndrome, and community-acquired pneumonia?

Bottom line

An updated guideline from a panel of experts convened by the Society of Critical Care Medicine strongly recommends the administration of steroids for adult patients who are hospitalized with severe CAP. Conditional recommendations also suggest the use of steroids for hospitalized adults with septic shock or ARDS.

Reference

Chaudhuri D, Nei AM, Rochwerg B, et al. 2024 focused update: Guidelines on use of corticosteroids in sepsis, acute respiratory distress syndrome, and community-acquired pneumonia. Crit Care Med 2024;52(5):e219-e233.

Study design: Practice guideline

Funding source: Self-funded or unfunded

Setting: Inpatient (any location)

Synopsis

The Society of Critical Care Medicine assembled an expert panel, including intensivists, pulmonologists, endocrinologists, nurses, and pharmacists, to provide evidence-based recommendations on the use of corticosteroids for hospitalized adults and children with sepsis, acute respiratory distress syndrome (ARDS), and community-acquired pneumonia (CAP). The panel conducted systematic reviews to find the best evidence for a series of questions. A team of reviewers screened titles and abstracts, performed data extraction, and assessed risk of bias both independently and in duplicate. Following a voting process, the panel issued 4 recommendations. A strong recommendation with moderate certainty of evidence supports the use of steroids for hospitalized adults with severe bacterial CAP, primarily based on evidence of decreased hospital mortality (relative risk 0.62; 95% CI 0.45 - 0.85). This mortality benefit was not seen in patients with less severe CAP. Two conditional recommendations suggest the use of steroids for adults with septic shock or ARDS, again based on reduced short-term mortality. A final recommendation spoke against the use of high-dose/short-duration steroids for septic shock because of the risk of adverse effects. This dosage is defined as greater than 400 mg per day of hydrocortisone equivalent for less than 3 days.

Nita Shrikant Kulkarni, MD
Assistant Professor in Hospital Medicine
Northwestern University
Chicago, IL