📰 What's New in Medicine

Published ahead of print in the American Journal of Gastroenterology on October 5, 2026: 18 GRADE recommendations and 6 key concepts across four domains (cirrhosis complications, infection prevention, lifestyle, and conditions outside the liver). Most of the evidence is low certainty, so most recommendations are conditional.
Antibiotics after an upper GI bleed: CHANGED suggests a short 2-3 day course for any upper GI bleed in cirrhosis, variceal or not. Mortality was noninferior to longer courses in a meta-analysis, but bacterial infections were more common with the short course (risk difference 15.2%), so the panel says to extend it in higher-risk patients. Baveno VII and AASLD guidance still say up to 7 days, and this site's bleeding pages keep that duration while the two positions differ.
Preventing first ascites: NEW suggests carvedilol in compensated cirrhosis with clinically significant portal hypertension (HVPG ≥ 10 mm Hg), based on fewer first episodes of ascites in trials and meta-analyses.
Hepatic encephalopathy: suggests lactulose for minimal HE, adding rifaximin after overt HE on lactulose, and not rifaximin as primary prevention.
Varices: CONFIRMED carvedilol or band ligation for high-risk varices (strong). Carvedilol is favored over propranolol or nadolol because its added alpha-1 blockade lowers portal pressure further.
HCC: CONFIRMED ultrasound plus AFP every 6 months for anyone who would be a candidate for HCC treatment (strong). Suggests against aspirin to prevent HCC.
Vaccines and lifestyle: strong recommendations for hepatitis A testing and vaccination, influenza, SARS-CoV-2, RSV from age 50, and smoking-cessation counseling; conditional ones for hepatitis B vaccination, behavioral therapy after discharge for alcohol use disorder, dietitian-guided protein and calorie targets, structured aerobic and resistance exercise, and bisphosphonates for osteoporosis.
On this site: the cirrhosis topic already covers carvedilol or banding, 6-monthly HCC surveillance and vaccination.

The first update since 2007, released in parts (Clinical Infectious Diseases). Most important first:
Severe acute pulmonary or disseminated disease (September 2026, consensus): liposomal amphotericin B 3 mg/kg/day (5 mg/kg for CNS) for about 7-14 days, shorter if responding (4-6 weeks for CNS), then itraconazole, the preferred step-down azole, with a drug level in every patient after 1-2 weeks (target 1-4 mg/L).
Mild acute pulmonary disease (March 2025): CHANGED no antifungal routinely in immunocompetent patients, treating only if illness lasts over a month or progresses; moderate disease, treat or observe by severity and duration (itraconazole 200 mg TID for 3 days, then BID for 6-12 weeks); immunocompromised patients are treated.
Asymptomatic histoplasmomas: no treatment. Pregnancy: avoid azoles in the first trimester; use liposomal amphotericin B when treatment is needed.

Guideline summary approved September 8, 2026; the full manuscript is pending in Arthritis & Rheumatology and Arthritis Care & Research. Most important first:
New: NEW a GLP-1 receptor agonist for knee OA with obesity, alongside diet and exercise (conditional).
Changed direction since 2019: CHANGED glucosamine and chondroitin are now strongly recommended against at every joint; methotrexate is conditionally recommended for erosive hand OA (still against in knee and hip); massage is conditionally recommended for knee and hip.
Strong recommendations: UNCHANGED exercise and weight loss (knee/hip), topical NSAIDs (knee/hand), oral NSAIDs, intra-articular glucocorticoid (knee/hip), plus ultrasound guidance for hip injections. Strongly against: PRP, stem cells, hydroxychloroquine, bisphosphonates, TNF inhibitors.
Conditional against: opioids, hyaluronic acid, other SNRIs, TCAs, gabapentinoids, colchicine, vitamin D, fish oil.

Seven consensus statements on risk stratification, diagnostic evaluation and treatment, endorsed by ASHP, PIDS, SHEA and SIDP. Most important first:
Risk, not "complicated": CHANGED increased risk is community onset, cultures still positive at 48 h, or an intracardiac device; it widens the search for a deep or metastatic focus.
Echo: TTE in every adult (no very-low-risk group could be defined), then TEE after a negative TTE with a device, predisposing valve condition, cultures positive at 48 h or more, emboli, or more than one focus. Whole-body imaging (such as FDG-PET/CT) when increased risk leaves the focus unknown.
Duration: 14 days when evaluation finds no deep-seated or metastatic focus, whether low or increased risk; longer, guided by the focus, when one is found.
Cultures and lines: at least 2 sets at 48 h after the first positive, then every 24 to 48 h until negative; prompt central line removal; ID consultation strongly encouraged.

Published in Circulation on September 8, 2026 (DeSimone et al). Unlike the 2015 statement it makes consensus suggestions rather than graded recommendations, but it rewrites the antibiotic tables. Most important first:
Shorter courses: CHANGED 4 weeks for native-valve staphylococcal (was 6) and streptococcal IE, 6 weeks for E. faecalis on native or prosthetic valves, and 6 for prosthetic valves, all counted from clearance of bacteremia. After valve surgery, give the planned course minus the days already given, with at least 14 days after the operation; a positive valve culture does not restart the clock.
Less gentamicin: CHANGED consensus that its risk outweighs any benefit in staphylococcal prosthetic valve IE; dropped for penicillin-susceptible streptococci (the 2-week ceftriaxone plus gentamicin course is gone); kept only for fully penicillin-resistant strains for 2 weeks. E. faecalis gets ampicillin plus ceftriaxone. Rifampin for staphylococcal prosthetic valves starts only after bacteremia clears and 5 to 7 days of therapy.
Finishing outside the hospital: NEW partial oral therapy (POET) after at least 10 days of IV treatment for streptococci, E. faecalis, MSSA or coagulase-negative staph, once afebrile 3 days with CRP below 25% of peak and a TEE within 2 days; most of the writing group favors 2 oral agents. Dalbavancin 1500 mg IV on days 1 and 8 when outpatient IV therapy is not possible (DOTS, which excluded left-sided IE).
Surgery: NEW once indicated, within 48 to 72 hours; hemorrhagic or major ischemic stroke delays it at least 2, preferably 4 weeks, while a small stroke without deficits does not. An endocarditis team decides, and the surgeon makes the final call.
Also: the 2023 Duke-ISCVID criteria are the standard; brain MRI before surgery is reasonable even without symptoms; colonoscopy for E. faecalis or S. gallolyticus; complete device extraction in valvular IE with a pacemaker or ICD; percutaneous aspiration for selected right-sided IE at high surgical risk; long-term suppression when surgery is indicated but not offered.
On this site: the endocarditis topic and one-pager #60 now follow the 2026 tables, and the sheet has rejoined the Antibiotics pack.

Published in the September 2026 issue of the American Journal of Respiratory and Critical Care Medicine, the first guideline to cover high-flow nasal cannula, NIV and CPAP together across the whole spectrum of acute respiratory failure. Four questions, four answers:
Preoxygenation before intubation: CHANGED a strong recommendation for HFNC or NIV to prevent peri-intubation hypoxemia. This is the change most likely to alter what a resident does tonight: a non-rebreather cannot deliver FiO₂ much above 0.6 to 0.7 and fails in shunt, and in PREOXI, 2024 NIV halved hypoxemia during intubation (9.1% vs 18.5%), with fewer cardiac arrests as a secondary outcome.
Acute hypoxemic respiratory failure: strong recommendation for HFNC and a conditional one for NIV or CPAP, with close monitoring for the need to escalate, based mainly on fewer intubations.
Acute hypercapnic respiratory failure: strong recommendation for NIV, which reduces both mortality and intubation. HFNC only conditionally, and only for milder hypercapnia with mild acidemia (for example pH above 7.25), where prompt escalation to NIV is available.
After extubation: risk-based, HFNC for low-risk and NIV for high-risk patients to reduce reintubation.
On this site: the RSI, mechanical ventilation, asthma, COPD and procedure pages now preoxygenate with NIV or HFNC, with a non-rebreather as the fallback only.

Released at ESC Congress 2026 and published in the European Heart Journal; the first edition written jointly by all four societies. Most important first:
New classification: CHANGED the numbered types are replaced by primary MI (an acute coronary event, formerly type 1), secondary MI (supply-demand imbalance from another acute condition, formerly type 2) and procedure-related MI (after PCI or cardiac surgery).
Injury: defined by sex-specific 99th percentile troponin limits, with refined acute and chronic injury criteria; accelerated high-sensitivity troponin pathways are described.
MINOCA now stands for myocardial injury with non-obstructive coronary arteries, a working diagnosis that prompts a search for the cause. Also new: objective criteria for silent MI, guidance after sudden death, a low-resource pathway, and ICD-11 codes.

Published in Blood Advances; five 2019 recommendations retired. Most important first:
Initial therapy: CHANGED a corticosteroid plus rituximab or a thrombopoietic agent (with or without IVIG) rather than a corticosteroid alone (conditional), because steroids alone rarely give a lasting remission; a steroid alone only if neither is available.
After corticosteroids: CHANGED a thrombopoietic agent is recommended (the only strong recommendation) and rituximab suggested; if neither suits, a BTK inhibitor (rilzabrutinib, FDA-approved August 2025), mycophenolate or a SYK inhibitor (fostamatinib). The 2019 "splenectomy or TPO agonist" option was retired.
Unchanged: UNCHANGED treat below 30,000 or for bleeding, observe above it; admit a newly diagnosed patient below 20,000; prednisone 0.5-2 mg/kg for no more than 6 weeks, or dexamethasone 40 mg for 4 days.

Every question was revised; the panel uses "suggest" throughout because evidence gaps remain. Most important first:
ESBL complicated UTI: CHANGED TMP-SMX, ciprofloxacin or levofloxacin are preferred when susceptible; a carbapenem or cefepime-enmetazobactam when they cannot be used; piperacillin-tazobactam is now an alternative if the patient is not critically ill and not bacteremic (previously suggested against). Cefepime is still avoided.
ESBL outside the urinary tract: UNCHANGED a carbapenem (meropenem or imipenem if critically ill or hypoalbuminemic), then oral step-down to a fluoroquinolone or TMP-SMX once responding; piperacillin-tazobactam is not suggested even if susceptible.
ESBL cystitis: NEW pivmecillinam, sulopenem and gepotidacin join nitrofurantoin, TMP-SMX and a single aminoglycoside dose as preferred options; oral fosfomycin is an alternative for E. coli.
AmpC organisms (Enterobacter cloacae, K. aerogenes, C. freundii, and now Hafnia alvei): cefepime if MIC 8 or less; avoid ceftriaxone and piperacillin-tazobactam for invasive infection, though finishing ceftriaxone is reasonable in a non-severe infection that is improving.
Carbapenemase producers: KPC, ceftazidime-avibactam, meropenem-vaborbactam or imipenem-relebactam; NEW NDM, aztreonam-avibactam or cefiderocol; OXA-48, ceftazidime-avibactam. DTR Pseudomonas pneumonia: ceftolozane-tazobactam preferred. CRAB: sulbactam-durlobactam with imipenem or meropenem. Stenotrophomonas: cefiderocol alone preferred; TMP-SMX, minocycline or levofloxacin only within a combination.

Published in the American Journal of Gastroenterology (July 2026), 14 GRADE recommendations aimed at the outpatient and post-episode visit. It confirms rather than reverses the direction of travel since ACP 2022.
Antibiotics: CONFIRMED suggests against routine antibiotics for low-risk acute uncomplicated diverticulitis, and treats patients with high-risk features, immunocompromise, frailty, or no safe outpatient plan. When antibiotics are used, amoxicillin-clavulanate is preferred where possible: it works as well as metronidazole plus a fluoroquinolone and avoids fluoroquinolone harms, including C. difficile. Courses run 4 to 7 days.
Preventing recurrence: CHANGED now recommends against mesalamine (strong, moderate certainty) and against rifaximin and probiotics (conditional). Do not restrict nuts, seeds, corn or popcorn; do encourage a healthy diet, regular activity, weight loss, smoking cessation, moderating heavy alcohol and avoiding regular NSAIDs.
Colonoscopy: recommended after complicated diverticulitis (strong); after uncomplicated disease only with alarm symptoms or when colorectal cancer screening is not current.
Surgery: refer recurrent uncomplicated disease that impairs quality of life to discuss elective resection, rather than operating on an episode count.

The first guideline of its kind, published in JACC on June 9, 2026. It retires and replaces the 2013 AHA/ACC/TOS obesity guideline. The central argument: obesity, type 2 diabetes, chronic kidney disease and cardiovascular disease are not four separate problems managed in four separate clinics, they are one interconnected syndrome, and staging it changes what you screen for and when.
CKM stages 0 to 4 NEW: Stage 0 normal weight, glucose, BP, lipids and kidney function with no CVD. Stage 1 excess or dysfunctional adipose tissue, prediabetes. Stage 2 metabolic risk factors (hypertriglyceridemia, hypertension, metabolic syndrome, T2D) or moderate-to-high-risk CKD. Stage 3 subclinical CVD (CAC ≥ 100, or pre-HF by biomarkers or echo) with CKM risk factors, or a risk equivalent: very high-risk CKD (G4 to G5) or a predicted 10-year CVD risk ≥ 20%. Stage 4 clinical CVD (coronary disease, heart failure, stroke, PAD, atrial fibrillation) with CKM risk factors.
Screening cadence is set by stage, which is the practical change for clinic: anthropometrics and blood pressure annually in everyone; lipids, glucose and eGFR every 5 years in Stage 0, every 2 to 3 years in Stage 1, annually in Stage 2; and urine albumin-to-creatinine ratio annually from Stage 2 onward.
Drug therapy by problem: for obesity, lifestyle first, with GLP-1-based therapy (semaglutide, tirzepatide) as an adjunct and metabolic and bariatric surgery reasonable in Stages 1 to 3 when lifestyle fails (both Class 2a). For T2D with CVD or raised risk (PREVENT-CVD ≥ 7.5% in Stages 2 to 3), an SGLT2 inhibitor or a GLP-1-based therapy with proven cardiovascular benefit, or both (Class 1), chosen by comorbidity: CKD, ASCVD, HF, obesity, MASLD. For CKD with T2D or albuminuria, a RAS inhibitor plus an SGLT2 inhibitor first; if albuminuria persists in CKD with T2D, add finerenone (UACR ≥ 30, eGFR ≥ 25) or a GLP-1-based therapy (UACR ≥ 100), each Class 1.
Risk tool: the PREVENT equations, the same family adopted by the 2026 dyslipidemia guideline, so one calculator now serves both.
Honest caveat the guideline itself makes: its economic value statements rate GLP-1-based therapy in Stage 1 (obesity with no other risk factor) and finerenone in CKD with T2D as not cost-effective at 2025 US prices, even though both are recommended; bariatric surgery and SGLT2 inhibitors in CKD are rated cost-effective. Care is tiered by comorbidity burden, from lifestyle alone up to multidisciplinary teams with a CKM coordinator.

Published in Thyroid (2026;36(5):481-544) and endorsed by ACOG, the Endocrine Society, AACE and nine other societies. Most important first:
Subclinical hypothyroidism turns on timing: CHANGED found in the first trimester, treat or recheck by shared decision (25 to 75 mcg/day if treated); found later, levothyroxine is not started, because the CATS and NIH trials showed no benefit once started after about 13 to 17 weeks. TPO antibody status no longer guides the decision.
Recheck before treating: NEW a raised TSH below 6 mU/L normalizes in about half within weeks; overt hypothyroidism with TSH 6 or more is treated at once.
Known hypothyroidism: increase levothyroxine on the positive test (needs rise about 25% by week 12 and 50% by week 20), tailoring the increase to the pre-pregnancy TSH and dose; target TSH below 2.5; first-trimester range about 0.1 to 4.0 without a local range; back to the pre-pregnancy dose at delivery.
Euthyroid TPO positivity: CHANGED not treated, even with infertility or recurrent miscarriage (three trials, including TABLET); check thyroid function every 3 to 6 months before conception instead.
Graves disease: switch methimazole to PTU once trying to conceive; on a positive test consider stopping the drug; PTU if one is needed in the first trimester; CHANGED after 16 weeks, no recommendation between PTU and methimazole; TRAb in the first trimester, with fetal and neonatal monitoring above 3 times the upper limit of normal; methimazole preferred in breastfeeding.
Also: iodine 150 mcg/day from 3 months before conception through lactation; no universal screening, test by risk factors; postpartum thyroiditis treated with a beta-blocker (never an antithyroid drug) in its thyrotoxic phase.

Released May 7, 2026, it replaces the 2023 guideline and is now split into three parts -presentation and evaluation, medical management, and procedural or surgical management- across 62 recommendations.
The framing change that matters most in general medicine: the guideline states explicitly that urinary symptoms in an older man often arise from several contributing factors at once rather than from prostate enlargement alone. That is the single commonest error on this presentation, and it is why nocturnal polyuria, heart failure, sleep apnea, diabetes, diuretic timing, urethral stricture and neurogenic bladder belong in the workup rather than as afterthoughts.
Two combinations that were previously unsupported now are. NEW 2026 Daily low-dose tadalafil added to an alpha-blocker, or added to finasteride, on evidence of symptom improvement with preserved ejaculatory function. NEW 2026 An antimuscarinic added to an alpha-blocker where storage symptoms persist, with symptom and quality-of-life benefit and a low risk of retention -check the post-void residual first, and weigh anticholinergic burden in older men.
Unchanged and worth restating: UNCHANGED offer a 5-alpha reductase inhibitor when the prostate is above 30 cc and/or PSA is above 1.5 ng/mL, both for symptoms and to reduce progression -and remember it halves the PSA, so the measured value must be doubled.

Released by the Rome Foundation in May 2026, the first revision of the diagnostic criteria since Rome IV in 2016. For IBS the changes are small in wording and large in effect:
Frequency: CHANGED at least 3 days per month over the past 3 months, the Rome III threshold, replacing Rome IV's at least 1 day per week. Rome IV's bar roughly halved prevalence estimates (4.1% vs 10.1% in the Rome Foundation global survey) and selected a more severe minority; Rome V captures more patients, earlier, including a milder group.
Pain or discomfort: CHANGED abdominal discomfort is back alongside pain, because patients who do not describe their sensation as pain, men in particular, were being excluded.
Continuous pain excluded: NEW the pain cannot be continuous, which separates IBS from centrally mediated abdominal pain syndrome. Patients excluded by this rule were the sickest in a UK registry, so they need re-phenotyping and centrally directed treatment, not discharge from the diagnosis.
Supportive criterion: in women, symptoms not related only to menses.
On this site: a new Irritable Bowel Syndrome topic built on Rome V, the ACG 2021 guideline and the AGA 2022 IBS-C and IBS-D pharmacology guidelines, with the places the two societies disagree (antispasmodics, PEG) stated side by side.

Published in Hepatology (April 2026), a GRADE guideline that answers six focused questions. It says the rest of the AASLD 2018 guidance still applies. The immune-active treatment thresholds are unchanged: ALT ≥ 2x the upper limit of normal with HBV DNA ≥ 20,000 IU/mL (HBeAg-positive) or ≥ 2,000 IU/mL (HBeAg-negative), using an ALT upper limit of 35 U/L in men and 25 U/L in women.
Immune-tolerant phase: CHANGED HBeAg-positive patients with DNA above 10 million IU/mL and normal ALT used to be monitored. Treatment is now suggested for those over 40 or with grade 2+ inflammation or F2+ fibrosis, with shared decision-making under 40. Anyone left untreated gets HBV DNA and ALT at least every 6 months.
Indeterminate "gray zone": NEW up to 40% of adults fit no defined phase. For HBeAg-negative patients without cirrhosis, the guideline suggests offering treatment by shared decision and revisiting the decision at every visit.
Pregnancy: STRENGTHENED HBV DNA above 200,000 IU/mL at any point means TDF or TAF from week 28, now a strong recommendation (2018: conditional, TDF preferred, weeks 28-32), because 8-10% of infants are still infected at that viral load despite HBIG and the birth-dose vaccine.
Stopping therapy: NEW in HBeAg-negative patients without cirrhosis whose DNA is suppressed, the guideline suggests not stopping until HBsAg loss. Anyone who insists on stopping must meet strict criteria and needs DNA and ALT every 1-3 months for the first 6 months.
HCC surveillance: NEW now also covers HDV coinfection at any fibrosis stage, HIV coinfection (men from 18, women from 40), and patients after HBsAg loss with cirrhosis, a family history of HCC, or loss after age 40 (men) or 50 (women).
On this site: the viral hepatitis topic and one-pager now treat immune-tolerant patients over 40, add the gray-zone path, and give tenofovir from week 28. Until this update both still said immune-tolerant patients are monitored, not treated.

The first ever AHA/ACC guideline dedicated to acute PE, published in Circulation on March 24, 2026 (epub February 19), written across 11 organizations including ACCP, ACEP, CHEST, SCAI, SHM, SIR, SVM and SVN.
The big change is the vocabulary: the old massive / submassive / low-risk language is replaced by the AHA/ACC Acute PE Clinical Categories, A through E with subcategories NEW, graded from low to high risk of adverse outcome. The point is that "submassive" lumped together patients whose management now diverges sharply.
Disposition follows the category. Category A (asymptomatic PE) can go home from the emergency department and does not need admission. Category C (symptomatic with an elevated severity score, elevated biomarkers and/or RV dysfunction), Category D (incipient cardiopulmonary failure) and Category E (cardiopulmonary failure with persistent hypotension) should be hospitalized.
Advanced therapy thresholds: systemic thrombolysis, catheter-directed thrombolysis, mechanical thrombectomy and surgical embolectomy are reasonable in Category E1 and may be considered in Categories D1 to D2. PE response teams (PERT) are recommended to improve the timeliness of these decisions, which is the part most likely to change how your hospital is organized rather than what you personally prescribe.

First major update since 2021. Published March 23, 2026 by SCCM/ESICM. 129 statements including 46 entirely new recommendations. Key changes:
Screening: NEWS/NEWS2/MEWS/SIRS now recommended over qSOFA for sepsis screening. qSOFA has poor sensitivity and should not be used as a sole screening tool. Sepsis remains a clinical diagnosis -no single biomarker can rule it in or out.
Antibiotics: Broad-spectrum within 1 hour for septic shock and for probable or definite sepsis; only possible sepsis without shock allows a time-limited workup with antibiotics within 3 hours if infection is still suspected. Blood cultures before antibiotics when possible. New section on antibiotic optimization beyond just timing.
Vasopressors: Peripheral vasopressors OK to start -don't delay for central access. Norepinephrine remains first-line, still a strong recommendation over dopamine, epinephrine or selepressin; only the narrower preference over vasopressin or angiotensin II is conditional. Add vasopressin as norepinephrine doses escalate, then epinephrine. MAP target ≥ 65 mmHg; new: MAP 60-65 for adults ≥ 65 years. Prolonged beta-lactam infusion now recommended (strong) for maintenance dosing.
Corticosteroids: IV hydrocortisone 200 mg/day suggested for adults with septic shock (conditional, low certainty; 2026 dropped the 2021 qualifier of an ongoing vasopressor requirement, and 2021 had reversed the 2016 stance against routine steroids). Shortens shock duration but unclear mortality benefit. Trial enrollment (ADRENAL, APROCCHSS) used a ≥ 4h threshold, but no specific wait time is mandated.
Fluids: Balanced crystalloids suggested over 0.9% saline. Initial bolus of at least 30 mL/kg IV crystalloid in first 3 hours for sepsis-induced hypoperfusion. Albumin not recommended as supplement to crystalloids. New guidance on fluid removal after resuscitation -a topic not previously addressed.
New topics: "Code sepsis" / sepsis huddle protocols suggested. Performance improvement programs recommended at institutional level. Post-sepsis discharge care including psychological and physical rehabilitation. Prehospital antibiotics suggested only for probable or definite septic shock when in-hospital evaluation is more than 60 min away (conditional, very low certainty); the largest RCT, PHANTASi 2018, showed no mortality benefit.
Pediatric: 68-expert panel, 20 new topics. POCUS conditionally recommended to guide resuscitation. 30–40% of children surviving ICU-level sepsis face lasting health issues -long-term follow-up now addressed.

First major update since 2018. Released March 13, 2026 by ACC/AHA in conjunction with AACVPR/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA. Retitled from "Blood Cholesterol" to "Dyslipidemia" to reflect broader scope (LDL + TG + remnants + Lp(a)). Core principle: "lower LDL for longer is better" -cumulative-exposure framing analogous to pack-years.
Lower LDL targets in primary prevention NEW: PREVENT ≥10% → LDL <70. PREVENT 5-<10% → LDL <100. 2018 had no numeric primary-prev LDL goal -only "≥50% reduction."
Risk calculator: PREVENT-ASCVD replaces the 2013 Pooled Cohort Equation. Validated ages 30-79, removes race as input, adds CKD/HbA1c/social deprivation, estimates 10- AND 30-year risk. Generally produces lower 10-yr estimates than PCE -especially in Black patients and the elderly.
New risk categories: Low <3% / Borderline 3-<5% / Intermediate 5-<10% / High ≥10% (replaces 2018's <5/5-7.5/7.5-20/≥20%).
Earlier statin starts: ages 30-59 at low 10-year risk with LDL 160-189 or a 30-year PREVENT ≥ 10% can start a moderate-intensity statin (Class 2a), and children with FH get a statin from age 8 if LDL stays ≥ 160 after 3-6 months of lifestyle (Class 1). The first take-home message also asks for early consideration of drug therapy in young adults with LDL ≥ 160 or a strong family history of premature ASCVD.
Routine apoB and Lp(a) testing: Lp(a) once per lifetime in all adults. ApoB when the LDL and non-HDL goals are met, especially TG ≥150, DM, or LDL <70, to expose residual particle burden. The guideline sets no apoB goal tiers: its only apoB numbers are ≥120 as a risk enhancer and ≥140 as a severe-hypercholesterolemia equivalent.
Non-statin add-ons: reframed from strict ladder to parallel options (ezetimibe / bempedoic acid / PCSK9i mAb / inclisiran). Combination therapy is the rule, not the exception, for very-high-risk ASCVD. CLEAR Outcomes 2023 (bempedoic acid) and ORION (inclisiran) data drove this expansion.
HIV (REPRIEVE 2023): pitavastatin 4 mg cut MACE 35% in HIV patients age 40-75 below traditional LDL threshold -statin recommended for primary prevention in HIV regardless of LDL.

Published in Critical Care Medicine (March 2026) and narrowed to ARDS, the most common reason for sustained paralysis in the ICU. Five questions, all answered with conditional recommendations:
Paralyze when P/F stays below 150 NEW in a patient who is persistently hypoxemic and/or not meeting ventilator targets on sedation, after ventilator and sedation optimization, and for no more than 48 hours (conditional, low certainty). Why: pooled across 7 RCTs, paralysis lowered 28-day mortality (RR 0.74) and barotrauma (RR 0.55), with a trend toward more ICU-acquired weakness (RR 1.16).
Equipoise on titratable vs fixed-dose NMBAs, on monitoring sedation depth during paralysis, and on paralysis during prone positioning: the evidence does not favor either option.
How it fits with what you know: ROSE 2019 still shows that routine early paralysis under light sedation adds nothing, and SSC 2026 still suggests intermittent boluses over a continuous infusion. The synthesis: light sedation by default, paralysis as a targeted and time-limited step for the patient who stays hypoxemic.

Published in Stroke on January 26, 2026 (print August 2026), it retires the 2018 guideline and its 2019 update. The practice-changing parts:
Thrombolytic agent: CHANGED tenecteplase 0.25 mg/kg (max 25 mg) or alteplase 0.9 mg/kg, both Class 1 within 4.5 h. Treat a disabling deficit regardless of NIHSS, without waiting for advanced imaging. For a non-disabling deficit (for example, isolated sensory symptoms), thrombolysis has shown no benefit and DAPT is recommended instead (ARAMIS, 2023).
Thrombectomy, broadened: NEW 6 to 24 h with NIHSS 6 or more and ASPECTS 6 or more is Class 1 on CT and CTA alone; perfusion or MRI mismatch is adjunctive (Class 2a) when immediately available. Large cores: ASPECTS 3 to 5 is Class 1 up to 24 h (beyond 6 h only under age 80 and without significant mass effect; SELECT2, 2023, ANGEL-ASPECT, 2023), and ASPECTS 0 to 2 within 6 h is Class 2a in selected patients under 80. Basilar occlusion with NIHSS 10 or more and PC-ASPECTS 6 or more: Class 1 within 24 h (ATTENTION, 2022, BAOCHE, 2022). Tirofiban before thrombectomy: no benefit.
Extended-window thrombolysis: reasonable (Class 2a) at 4.5 to 9 h, or on waking within 9 h of the midpoint of sleep, when automated perfusion imaging shows salvageable penumbra (EXTEND, 2019); MRI DWI-FLAIR mismatch for unknown onset. Adding argatroban or eptifibatide to thrombolysis is not recommended (MOST, 2024: more deaths, no benefit).
Blood pressure: CHANGED keep 185/110 or lower before, and 180/105 or lower during and for 24 h after thrombectomy (Class 2a). An intensive SBP target below 140 after successful recanalization is harmful for the first 72 h (Class 3: Harm; OPTIMAL-BP, 2023), and after thrombolysis it adds nothing (Class 3: No Benefit).
Glucose: 140 to 180 mg/dL; IV insulin to 80 to 130 is not recommended (Class 3: No Benefit; SHINE, 2019).
Antithrombotics: 21 days of DAPT extended to NIHSS 5 or less within 24 to 72 h (or NIHSS 4 to 5 within 24 h) of presumed atherosclerotic cause, without thrombolysis (Class 2a; INSPIRES, 2023). Early anticoagulation for atrial fibrillation is reasonable in carefully selected, milder strokes (Class 2a).
Systems and new populations: mobile stroke units endorsed; direct transport to the closest thrombectomy-capable hospital where transfer systems are slow; baseline ECG and troponin without delaying treatment; never delay CTA waiting for a creatinine; and the first pediatric recommendations, including thrombectomy from age 6 by experienced neurointerventionalists (Class 2a).
The tension worth knowing: 20 investigators from the DAWN and DEFUSE trials formally asked the writing group to downgrade the plain-CT 6 to 24 h recommendation to Class 2b. The writing group declined in Stroke (July 31, 2026), citing DAWN, DEFUSE 3, MR CLEAN-LATE and the AURORA meta-analysis, and the NO-SELECT trial is now testing the question directly.

Annual update covering all aspects of diabetes care. Key 2026 changes: GLP-1 RA / dual GIP-GLP-1 (tirzepatide, semaglutide) elevated to first-line for obesity-related T2DM ahead of sulfonylureas / TZDs, especially when weight management is a goal.
SGLT2 inhibitors recommended for ALL patients with CKD (eGFR ≥ 20, regardless of A1c) -mirrors KDIGO 2024.
Continuous glucose monitoring (CGM) expanded to T2DM on any basal insulin (previously intensive insulin only). Time-in-range targets formalized.
DM screening starts at age 35 for all adults regardless of risk; age 25 if BMI ≥ 25 + risk factors (down from 35 with BMI criterion).
Inpatient targets: glucose 100-180 mg/dL on the wards (a separate ward goal since 2025) and 140-180 in the ICU; basal-bolus preferred over sliding-scale-only; SQ insulin acceptable for mild-moderate DKA.
Lp(a) screening once per lifetime in adults with DM (mirrors ACC/AHA 2026 Dyslipidemia Guideline).
GLP-1 RA + SGLT2i preferred over insulin in many T2DM scenarios when ASCVD, HF, or CKD coexist -insulin is not "default escalation" anymore.

Released in November 2025 by the Global Initiative for Chronic Obstructive Lung Disease (v1.3, December 8, 2025). Most important first:
Group E starts at one exacerbation: CHANGED one moderate or severe exacerbation in the past year now puts a patient in group E (it was 2 moderate, or 1 leading to hospitalization), because even one exacerbation before maintenance therapy predicts more. One moderate exacerbation is the new threshold to consider escalating treatment, with the aim of no exacerbations at all.
Biologics: NEW FIGURE the evidence for dupilumab (BOREAS, NOTUS: eosinophils ≥ 300 with chronic bronchitis) and mepolizumab (MATINEE). In the follow-up algorithm a biologic is for patients on triple therapy with 2 moderate or 1 severe exacerbation and eosinophils ≥ 300. Benralizumab is not included.
Structure: a new section on disease activity; INITIAL treatment (for patients new to maintenance therapy) is now separated from FOLLOW-UP treatment; the exacerbation and multimorbidity chapters are fully revised; a new chapter on AI and digital tools.
Vaccines: UPDATED RSV at 50 or older and/or with chronic heart or lung disease; yearly influenza; one dose of PCV21 or PCV20.
Unchanged: initial therapy by group (A a bronchodilator, B LABA + LAMA, E LABA + LAMA with triple if eosinophils ≥ 300), escalation to triple at eosinophils ≥ 100, roflumilast and azithromycin, ensifentrine for persistent dyspnea, and ICS never without a LABA.

Published in Circulation on October 22, 2025, the first full rewrite of the CPR and ECC guidelines since 2020. For the ICU the part that changes practice is post-cardiac arrest care.
Temperature: CHANGED one constant target anywhere from 32 to 37.5°C (89.6 to 99.5°F) for at least 36 hours (2020: 32 to 36°C (89.6 to 96.8°F) for at least 24), in every adult who does not follow commands, any rhythm, any location. Oxygen: CHANGED SpO₂ 90–98% (2020: 92–98%). Pressure: CHANGED MAP at least 65, with the SBP target removed and no preferred vasopressor.
Cath lab: NEW 2025 emergency angiography without ST elevation is Class 3 (no benefit) unless there is shock, recurrent VT/VF or ongoing ischemia; angiography before discharge for a suspected cardiac cause is Class 1; culprit-only PCI in shock. Temporary mechanical circulatory support is not routine.
Brain: NEW 2025 a myoclonus section (do not treat myoclonus that has no EEG correlate), neurofilament light beside NSE, and, for the first time, predictors of a favorable outcome. UNCHANGED multimodal prognostication no earlier than 72 hours after normothermia and off sedation.

First major update since 2017. Replaces the prior hypertension guideline with updated evidence through 2025. Key changes include earlier risk stratification, expanded screening for secondary causes including primary aldosteronism, updated lifestyle and psychosocial intervention recommendations, and more individualized BP targets across the lifespan. Addresses tailored strategies for older adults, CKD, and resistant hypertension. → HTN Emergency Topic • → Outpatient Hypertension Topic

ATS released updated community-acquired pneumonia guidelines. IDSA agreed with 8 of 10 recommendations but declined endorsement over antibiotic use in viral pneumonia. IDSA argues that nondiscriminatory antibiotic use in patients with positive respiratory virus tests (flu, RSV, COVID) confers more harm than benefit, since most do not have bacterial coinfection. Practice impact: think twice before reflexively adding antibiotics to confirmed viral CAP. → Pneumonia Topic

Published in Clinical Infectious Diseases on July 17, 2025, covering pyelonephritis, catheter-associated UTI and other complicated UTI. Most important first:
Shorter courses: CHANGED 5 to 7 days of a fluoroquinolone or 7 days of another agent, and 7 days rather than 14 with gram-negative bacteremia. Men with febrile UTI and suspected prostatitis may need 10 to 14 days. Count from the first day of effective therapy.
Oral switch: NEW once improving and able to take oral drugs, including with bacteremia.
Empiric choice: with sepsis, a 3rd- or 4th-generation cephalosporin, a carbapenem, piperacillin-tazobactam or a fluoroquinolone rather than newer agents; without sepsis, the same list but not a carbapenem. Slow improvement should prompt a search for a source needing control.
On this site: the UTI topic and the urosepsis one-pager now follow these durations.

Published together in the June 2025 American Journal of Gastroenterology (Rubin et al for UC, Lichtenstein et al for Crohn). Most important first:
Every new mechanism is recommended: NEW the IL-23 inhibitors guselkumab, mirikizumab and risankizumab (UC and Crohn), upadacitinib (UC, and Crohn after anti-TNF exposure) and the S1P modulators ozanimod and etrasimod (UC), alongside anti-TNF, vedolizumab, ustekinumab and tofacitinib, with no single first line.
Crohn: treat early, and drop mesalamine: CHANGED do not require conventional therapy to fail before an advanced therapy; oral mesalamine is recommended against; thiopurines maintain but do not induce remission.
Head-to-head positioning: vedolizumab over adalimumab in UC (VARSITY), risankizumab over ustekinumab after anti-TNF exposure in Crohn (SEQUENCE), and infliximab as the preferred anti-TNF in UC, given with a thiopurine.
Acute severe UC: CONFIRMED methylprednisolone 60 mg a day or hydrocortisone 100 mg 3 to 4 times daily, C. difficile testing, pharmacologic VTE prophylaxis, and infliximab or cyclosporine by day 3. NEW no routine broad-spectrum antibiotics or TPN for bowel rest; consider a higher infliximab dose when albumin is below 2.5 g/dL; JAK inhibitors are not routine rescue.

First unified ACS guideline replacing the 2013 STEMI and 2014 NSTE-ACS guidelines. Biggest practice change: 1 month DAPT followed by ticagrelor monotherapy is now Class 1, LOE A (TWILIGHT, TICO, T-PASS), cuts clinically relevant (BARC 2, 3 or 5) bleeding by about 44% without raising ischemic risk. Ticagrelor or prasugrel preferred over clopidogrel post-PCI (clopidogrel is the preferred dual-therapy partner only when stacked on top of an OAC). For AFib + ACS + PCI: short triple therapy (1 to 4 weeks), then drop ASA, then OAC + clopidogrel through month 12 (Class 1; AUGUSTUS). LDL targets pushed lower: non-statin add-on if LDL ≥ 70 (Class 1), reasonable to intensify if LDL 55 to <70 (Class 2a). Intravascular imaging upgraded to Class 1 for PCI guidance. Complete revascularization Class 1 in multivessel disease (COMPLETE, MULTISTARS-AMI). Microaxial flow pump (Impella CP) Class 2a in select AMI cardiogenic shock per DanGer Shock 2024 (first positive shock RCT, 180-day mortality 45.8% vs 58.5%). Radial access preferred over femoral (Class 1; MATRIX). 0/1h and 0/2h hs-troponin algorithms formalized. → Full 2025 ACS Changes Comparison • → STEMI Topic • → NSTEMI Topic • → Read Full Guideline (Circulation)

Tirzepatide (Mounjaro) significantly reduced AHI in obese patients with OSA. Weight loss of ~20% with improvement in sleep apnea severity. May change first-line management of obesity-related OSA. → Obesity Topic

National Lipid Association moved Lp(a) from "consider checking in selected patients" to routine once-per-lifetime measurement in all adults. Lp(a) ≥50 mg/dL (or ≥125 nmol/L) is a major risk enhancer. Two years later the 2026 ACC/AHA guideline (not this statement, which does not grade by class) made a PCSK9 antibody the Class 1 add-on when Lp(a) ≥50 with established ASCVD leaves LDL-C and non-HDL-C goals unmet on a maximally tolerated statin; elevated Lp(a) does not by itself redefine very high risk. Statins don't lower Lp(a); PCSK9i and inclisiran lower it ~20-25%. Apheresis for refractory. Set the stage for the 2026 ACC/AHA Dyslipidemia Guideline adoption of routine Lp(a) testing. → Outpatient Lipid Management Topic

Updated GFR categories and albuminuria staging. SGLT2 inhibitors now first-line for CKD with albuminuria regardless of diabetes status. Finerenone added for diabetic kidney disease. → CKD Topic

Updated management of immune-related adverse events from checkpoint inhibitors. Expanded guidance on steroid dosing, when to hold immunotherapy, and specialty referral triggers. → irAE Topic

Semaglutide reduced MACE by 20% in overweight/obese adults with CVD, regardless of diabetes status. First GLP-1 RA with cardiovascular benefit independent of diabetes. → Obesity Topic

Semaglutide improved symptoms, exercise function, and reduced body weight in obese HFpEF patients. New treatment paradigm -treating the obesity driving HFpEF. → HFpEF Topic

Replaced the 2012 Stable Ischemic Heart Disease guideline. New "Chronic Coronary Disease" (CCD) framework consolidates stable angina, post-MI/post-revasc, asymptomatic plaque, and silent ischemia into one umbrella. Key changes: SGLT2i and GLP-1 RA codified for ASCVD patients with DM (diabetes-class effect on MACE). ISCHEMIA trial integrated -no benefit of routine revascularization in stable CCD with moderate ischemia; med therapy first. Icosapent ethyl (REDUCE-IT) recommended for residual TG-driven ASCVD risk in patients with TG 150-499 on statin. DAPT duration after PCI individualized by bleeding/ischemic risk balance. Lifestyle reinforcement, depression screening, and structured cardiac rehab strongly recommended. → Stable Angina / CCD Topic

No difference between restrictive and liberal fluid strategies in early septic shock. Supports individualized fluid resuscitation over mandatory 30 mL/kg. → Sepsis Topic

Rapid uptitration of all 4 GDMT pillars before discharge improved 180-day outcomes. Changed practice from slow sequential titration to rapid parallel initiation. → Heart Failure Topic

Empagliflozin reduced CKD progression and cardiovascular death regardless of diabetes status. SGLT2i now standard for CKD with albuminuria. → CKD Topic

Aggressive LR resuscitation did not improve outcomes in acute pancreatitis and increased fluid overload. Goal-directed fluid therapy is now preferred. → Pancreatitis Topic

SGLT2 inhibitors now benefit ALL heart failure regardless of ejection fraction. Dapagliflozin reduced worsening HF and cardiovascular death in HFpEF. → HFpEF Topic

30 mL/kg fluid bolus downgraded to weak recommendation. Balanced crystalloids suggested over NS. Hour-1 bundle replaces 3-hour and 6-hour bundles. Norepinephrine confirmed first-line vasopressor. → Sepsis Topic

Fidaxomicin (Dificid) now preferred over oral vancomycin for initial C. diff episodes -lower recurrence rate. Oral metronidazole is NO LONGER recommended for initial episodes. → C. difficile Topic

33°C hypothermia showed no benefit over normothermia with early fever treatment (cooling started at 37.8°C) after out-of-hospital cardiac arrest (1,900 randomized; 6-month death 50% vs 48%), and caused more arrhythmia with hemodynamic compromise (24% vs 17%). Preventing fever is what matters, not active cooling to 33°C. Changed post-arrest temperature management. → Post-Cardiac Arrest Topic

Early rhythm control (within 1 year of AF diagnosis) reduced cardiovascular death, stroke, and HF hospitalization vs rate control alone. Shifted the rate vs rhythm debate for newly diagnosed AF. → Atrial Fibrillation Topic

Early RRT initiation in AKI showed no benefit over a watchful waiting strategy. Delayed approach avoided unnecessary dialysis in ~50% of patients. Only start for absolute indications. → AKI Topic

Trough-based monitoring (target 15-20 mcg/mL) replaced by AUC/MIC-guided dosing (target 400-600). AUC monitoring reduces nephrotoxicity by ~30% with equivalent efficacy. → Sepsis Topic

Healthcare-Associated Pneumonia (HCAP) removed as a category. Led to massive overtreatment with unnecessarily broad antibiotics. Instead, assess individual MDR risk factors and use nasal MRSA PCR to guide de-escalation. → Pneumonia Topic

Oral antibiotics non-inferior to IV for bone and joint infections after initial IV therapy. Changed the dogma of "6 weeks IV only" for osteomyelitis. IV-to-oral step-down is now standard practice. → Osteomyelitis Topic