What's new in hospital medicine

The following represent additions to UpToDate from the past six months that were considered by the editors and authors to be of particular interest. The most recent What's New entries are at the top of each subsection.

GENERAL HOSPITAL MEDICINE

Estimation of free calcium using albumin-adjusted calcium formulas (March 2025)

Albumin-adjusted calcium formulas are often used to estimate free calcium when albumin is abnormal, but none appears to be universally acceptable when compared with ionized calcium. In a new cross-sectional study that included 17,500 patients who had simultaneous testing of albumin, total calcium, and ionized calcium, very few (≤0.3 percent) patients with an albumin <3 g/dL and hypercalcemia by ionized calcium were misclassified as normocalcemic by a commonly used calcium correction formula [1]. However, the formula misclassified 44 percent with hypocalcemia by ionized calcium as normocalcemic, and 6.8 percent with normocalcemia by ionized calcium were misclassified as hypercalcemic. If reliable measurement of ionized calcium is not available, the total calcium may be corrected for any abnormalities in albumin, but the accuracy of the estimate may be poor in a variety of populations. (See "Diagnostic approach to hypercalcemia", section on 'Verify elevated calcium' and "Relation between total and ionized serum calcium concentrations".)

Perioperative management of renin-angiotensin system inhibitors (November 2024)

The optimal perioperative management strategy of renin-angiotensin system inhibitors (RASIs) is unclear. In a trial of over 2000 patients who had been on RASIs for at least three months and were undergoing major noncardiac surgery, intraoperative hypotension was more common in patients randomly assigned to RASI continuation compared with discontinuation 48 hours prior to surgery (54 versus 41 percent), although composite rates of all-cause mortality and major postoperative complications were equivalent in both groups [2]. In patients on RASIs undergoing major noncardiac surgery, management is individualized; discontinuation of RASIs may be favored in patients at higher risk of intraoperative hypotension. (See "Perioperative medication management", section on 'ACE inhibitors and angiotensin II receptor blockers'.)

HOSPITAL CARDIOVASCULAR MEDICINE

Choice of management strategy in older patients with non-ST-segment elevation myocardial infarction (December 2024)

In patients with non-ST-segment elevation myocardial infarction (NSTEMI), the approach to management typically includes early coronary angiography and appropriate percutaneous coronary intervention (PCI), but the efficacy and safety of this approach in older patients are unclear. In a recent trial in over 1500 patients ≥75 years old with NSTEMI, patients randomly assigned to an invasive management strategy (ie, coronary angiography and PCI) had a lower rate of recurrent myocardial infarction but a similar rate of all-cause death when compared with those assigned to a conservative management strategy [3]. The rate of non-fatal bleeding was not significantly higher in the invasive management group. In patients with NSTEMI, we suggest invasive coronary angiography rather than other management strategies. (See "Non-ST-elevation acute coronary syndromes: Selecting a management strategy", section on 'Evidence of infarction (NSTEMI)'.)

Inferior vena cava filter retrieval rates among Medicare beneficiaries (November 2024)

Although retrievable inferior vena cava (IVC) filters should be removed when their protection is no longer needed, removal rates vary widely depending on the population studied. In a review of nearly 271,000 Medicare beneficiaries, the proportion of removed IVC filters increased incrementally after 2014; however, the cumulative incidence of removal was only 17 percent at a maximum follow-up of nine years [4]. This low rate compared with other studies may reflect population characteristics associated with lower removal rates (eg, older age, cancer). Nevertheless, this study underscores the need to periodically reevaluate whether a patient with an IVC filter is a candidate for its removal regardless of age or medical comorbidities. (See "Placement of vena cava filters and their complications", section on 'Filter retrieval'.)

Threshold for potassium repletion to prevent atrial fibrillation after cardiac surgery (November 2024)

General management to reduce the risk of atrial fibrillation (AF) after cardiac surgery includes treatment of hypokalemia; however, the optimum threshold for potassium replacement therapy in this setting has not been established. In a trial in which 1700 patients undergoing coronary artery bypass graft surgery were randomly assigned to potassium supplementation at a potassium concentration trigger of <4.5 mEq/L or <3.6 mEq/L, rates of at least one AF episode after cardiac surgery and inpatient mortality were similar between the groups [5]. These results support the use of a standard potassium threshold (<3.6 mEq/L) for treatment of post-operative hypokalemia. (See "Atrial fibrillation and flutter after cardiac surgery", section on 'General measures'.)

HOSPITAL HEMATOLOGY

Reduced-dose apixaban and rivaroxaban for indefinite venous thromboembolism treatment (April 2025)

The optimal anticoagulant dose for prevention of venous thromboembolism (VTE) among patients at high risk of recurrence is unknown. In a trial of over 2700 such patients who had completed 6 to 24 months of anticoagulation, patients who transitioned either to reduced-dose apixaban (2.5 mg twice daily) or rivaroxaban (10 mg once daily) had a higher five-year VTE recurrence rate, compared with patients who continued to take full-dose anticoagulants (2.2 versus 1.8 percent) [6]. However, low-dose anticoagulant therapy was associated with a lower risk of major and/or clinically relevant bleeding (9.9 versus 15.2 percent). Low-dose apixaban or rivaroxaban is appropriate for most patients at high risk of recurrent VTE who require indefinite anticoagulation. (See "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation", section on 'Reduced-dose regimens for indefinite anticoagulation'.)

Inappropriate use of plasma in the intensive care unit (January 2025)

Plasma is thought to be over-used, often without a compelling indication. A new multicenter study supports this impression and illustrates the magnitude of misuse [7]. Among 3643 individuals in the intensive care unit (ICU), 10 percent received one or more plasma transfusions; of these, 37 percent did not have a strong indication and might have been avoided, such as mild elevations of the international normalized ratio (INR) in nonbleeding patients. Plasma transfusion should be limited to situations with a strong clinical rationale and/or demonstrated efficacy. (See "Use of blood products in the critically ill", section on 'Plasma indications'.)

Safety of antithrombotic therapy in factor XI deficiency (January 2025)

Spontaneous bleeding is rare in factor XI deficiency, suggesting that anticoagulation or antiplatelet therapy may be safer in these individuals than in those with other bleeding disorders. A new database study evaluated bleeding risk in approximately 50 percent of the general population in Israel [8]. While use of an anticoagulant or antiplatelet agent increased bleeding risk, individuals with factor XI deficiency did not have an appreciably higher rate of bleeding with these treatments than individuals without factor XI deficiency. These results suggest that antithrombotic therapy may be used when needed in individuals with factor XI deficiency, using shared decision-making based on the individual's bleeding phenotype and thrombotic risk. (See "Factor XI (eleven) deficiency", section on 'Anticoagulation or antiplatelet therapy'.)

Anticoagulation for patients hospitalized with COVID-19 (January 2025)

COVID-19 can cause a hypercoagulable state, especially when the disease is severe enough to require hospitalization, but the extent of hypercoagulability has decreased over time. A new meta-analysis of 20 randomized trials reported that, overall, therapeutic-dose anticoagulation was associated with lower 28-day mortality than prophylactic dosing, along with a lower risk of thromboembolic complications and a higher risk of bleeding [9]. However, the more recent trials did not show a significant benefit of therapeutic-dose anticoagulation relative to prophylactic dosing. These data support the use of prophylactic dosing in the intensive care unit (ICU) and in hospitalized (non-ICU) medical patients. (See "COVID-19: Hypercoagulability", section on 'Supporting evidence'.)

Increased risk of pulmonary embolism in sickle cell trait (December 2024)

Sickle cell trait is an asymptomatic carrier state, but risks are increased for certain complications. A new study used genetic data from almost four million individuals to assess risks for venous thromboembolism (VTE), including pulmonary embolism (PE) and deep vein thrombosis (DVT) [10]. Compared to individuals without sickle cell trait, those with sickle cell trait had an approximately 2-fold increased risk for PE; a definite increased risk for DVT was not identified, and the overall VTE risk was increased 1.5-fold. This VTE risk for sickle cell trait was less than that for factor V Leiden (3.3-fold) and is not great enough to warrant management changes. Appropriate VTE prophylaxis during hospitalization should be emphasized. (See "Sickle cell trait", section on 'Venous thromboembolism'.)

HOSPITAL INFECTIOUS DISEASES

Trends of invasive group A Streptococcus infections in the United States (April 2025)

Since 1995, the United States Centers for Disease Control and Prevention (CDC) has tracked rates of invasive group A Streptococcus (GAS) infections, including toxic shock syndrome. In an analysis of data from 10 states from 2013 to 2022, overall incidence of invasive GAS infections increased from 3.6 to 8.2 cases per 100,000 persons [11]. The increase occurred predominantly in the adult population, and patients at highest risk included those over 64 years of age, American Indian or Alaskan Native individuals, residents of long-term care facilities, and people experiencing homelessness or who inject drugs. The percentage of isolates non-susceptible to both macrolides and clindamycin increased from 13 to 33 percent during the study period. Over 21,000 cases were identified, and 9 percent of patients died. This study highlights increasing rates of this severe infection and groups of individuals at the highest risk. (See "Invasive group A streptococcal infection and toxic shock syndrome: Epidemiology, clinical manifestations, and diagnosis", section on 'Incidence'.)

Duration of antibiotics for gram-negative bacillary bacteremia (April 2025)

Although bacteremia had traditionally been treated with at least 14 days of antibiotics, mounting data indicate that shorter courses are often appropriate. In a randomized trial of over 3600 hospitalized patients with bloodstream infections (mostly due to gram-negative bacilli), 90-day mortality was similar in patients treated with 7 or 14 days of antibiotics (15 versus 16 percent) [12]. Patients with immunocompromising conditions, prosthetic heart valves or vascular grafts, a syndrome requiring prolonged therapy (eg, endocarditis), or Staphylococcus aureus bacteremia were excluded from the study. A recent meta-analysis reported similar findings [13]. These data support our approach to treat most patients with uncomplicated gram-negative bacillary bacteremia with seven days of antibiotics. (See "Gram-negative bacillary bacteremia in adults", section on 'Duration and route of therapy'.)

Aztreonam-avibactam for complicated intra-abdominal infections (March 2025)

The US Food and Drug Administration has approved aztreonam-avibactam to be used in combination with metronidazole for treatment of complicated intra-abdominal infections (cIAI) [14]. Approval was based primarily on a randomized trial of 422 patients with either cIAI or hospital-acquired pneumonia that found similar rates of clinical cure and all-cause mortality with aztreonam-avibactam (plus metronidazole for cIAI) compared with meropenem (with or without colistin) [15]. Aztreonam-avibactam may be particularly useful for infections due to carbapenem-resistant Enterobacterales spp that produce metallo-beta-lactamases. The combination drug was approved for use in Europe in 2024. (See "Carbapenem-resistant E. coli, K. pneumoniae, and other Enterobacterales (CRE)", section on 'Carbapenemase testing is positive'.)

HOSPITAL NEPHROLOGY

Association of hyponatremia correction rates with in-hospital mortality (December 2024)

In patients with severe, chronic hyponatremia (ie, serum sodium <120 mEq/L), the goal of initial therapy is to raise the serum sodium level slowly to avoid overcorrection. This recommendation was challenged by a meta-analysis of retrospective cohort studies that found that rapid correction of severe hyponatremia in hospitalized patients was associated with lower in-hospital mortality compared with slow correction [16]. However, the major limitation of this study is its analysis of all-cause in-hospital mortality rather than death from cerebral edema (the cause of death potentially caused by slower correction). Thus, the findings were likely confounded by the fact that severely ill patients, who are at high risk for in-hospital mortality, have hyponatremia that is more recalcitrant and less likely to correct without hypertonic saline; conversely, patients with acute hyponatremia or less severe underlying disease (and a lower risk of in-hospital mortality) are more likely to correct quickly when the cause of hyponatremia is eliminated. We continue to suggest slow correction in patients with severe, chronic hyponatremia. (See "Overview of the treatment of hyponatremia in adults", section on 'Goal rate of correction'.)

Threshold for potassium repletion to prevent atrial fibrillation after cardiac surgery (November 2024)

General management to reduce the risk of atrial fibrillation (AF) after cardiac surgery includes treatment of hypokalemia; however, the optimum threshold for potassium replacement therapy in this setting has not been established. In a trial in which 1700 patients undergoing coronary artery bypass graft surgery were randomly assigned to potassium supplementation at a potassium concentration trigger of <4.5 mEq/L or <3.6 mEq/L, rates of at least one AF episode after cardiac surgery and inpatient mortality were similar between the groups [5]. These results support the use of a standard potassium threshold (<3.6 mEq/L) for treatment of post-operative hypokalemia. (See "Atrial fibrillation and flutter after cardiac surgery", section on 'General measures'.)

HOSPITAL NEUROLOGY

Tenecteplase versus alteplase for intravenous thrombolysis of acute ischemic stroke (November 2024, Modified March 2025)

The benefit of intravenous thrombolysis (IVT) for acute ischemic stroke was initially established by randomized trials using alteplase. However, tenecteplase is easier to administer than alteplase, and there is mounting evidence that tenecteplase has efficacy and safety outcomes similar to alteplase. This evidence is bolstered by a recent meta-analysis of 11 randomized trials of IVT within 4.5 hours of acute ischemic stroke comparing tenecteplase (n = 3788) with alteplase (n = 3757) [17]. Tenecteplase and alteplase had similar rates of both excellent and good functional outcomes and similarly low rates of symptomatic intracerebral hemorrhage and mortality. For eligible patients with acute ischemic stroke, we recommend IVT using either alteplase or tenecteplase for treatment within 3 hours of the time last known well, and we suggest IVT with alteplase or tenecteplase for treatment within 3 to 4.5 hours of the time last known well. Tenecteplase was approved by the US Food and Drug Administration for this purpose in February 2025 [18]. (See "Approach to reperfusion therapy for acute ischemic stroke", section on 'Tenecteplase'.)

Timing of resumption of antiplatelet therapy after intracerebral hemorrhage (December 2024)

Antiplatelet therapy in patients with established atherosclerotic disease who develop intracerebral hemorrhage (ICH) is often resumed to reduce the risk of early ischemic complications, but the optimal time for safe resumption is uncertain. In a multicenter clinical trial that included 269 patients from China with acute nontraumatic ICH and atherosclerotic risk factors who underwent hematoma evacuation, the rate of new ischemic cerebral, cardiac, or peripheral vascular events within 90 days was lower in patients assigned to start aspirin on the third day following surgery than in those assigned to start aspirin 30 days after surgery (20 versus 31 percent) [19]. Intracranial hemorrhage was rare, occurring in one patient in the early aspirin group and four patients in the late group. These results support our preference for early resumption of antiplatelet therapy after ICH in patients with atherosclerotic disease. (See "Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis", section on 'Antiplatelet therapy'.)

Transfusion threshold for patients with aneurysmal subarachnoid hemorrhage (December 2024)

Anemia in aneurysmal subarachnoid hemorrhage (SAH) has been associated with cerebral infarctions and poor outcomes, but the benefit of a liberal transfusion strategy in this setting is uncertain. In a clinical trial of 742 patients with acute SAH that compared strategies of transfusion thresholds at hemoglobin ≤10 g/dL (liberal) versus ≤8 g/dL, functional outcomes were similar between groups (34 versus 38 percent with an unfavorable outcome at 12 months) [20]. Nearly all patients in the liberal threshold group received transfusions compared with 35 percent in the lower threshold group, and adverse event rates were similar. These results support our practice of using a restrictive transfusion threshold for patients with SAH and anemia. (See "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis", section on 'Anemia'.)

Timing of anticoagulation after acute ischemic stroke in patients with atrial fibrillation (November 2024)

The timing of anticoagulation after acute ischemic stroke in patients with atrial fibrillation (AF) is controversial. Because of concern for intracranial hemorrhage, the start of anticoagulation is often delayed by one to two weeks for certain patients, such as those with large infarcts. However, recent findings from the OPTIMAS randomized trial and the CATALYST meta-analysis challenge this approach [21-23]. In a meta-analysis of patient-level data on over 6700 patients from trials (including OPTIMAS) evaluating the timing of direct oral anticoagulant (DOAC) initiation in patients with stroke and AF, rates of symptomatic intracranial hemorrhage were similarly low for both early (≤4 days from stroke onset) and late (≥5 days) DOAC administration, and the composite outcome of stroke or hemorrhage with early initiation was marginally lower at 30 but not 90 days [22]. Approximately 15 percent of patients had large or severe infarcts in the three largest trials. These data support the safety of early DOAC administration in patients with AF and ischemic stroke, but whether earlier DOAC treatment reduces the risk of recurrent ischemic stroke remains to be settled. (See "Early antithrombotic treatment of acute ischemic stroke and transient ischemic attack", section on 'Timing of anticoagulation after acute ischemic stroke or TIA in patients with atrial fibrillation'.)

HOSPITAL PULMONOLOGY AND CRITICAL CARE MEDICINE

Securing central venous catheter dressings (May 2025)

A central venous catheter (CVC) must be secured to the skin to stabilize it, but the optimal dressing is unclear. In a randomized trial of patients undergoing jugular CVC insertion, application of medical liquid adhesive (MLA) under the standard CVC dressing border resulted in fewer dressing failures due to lifting edges at seven days compared with a standard dressing alone (28 versus 50 percent) [24]. Skin complications were similar between the groups. While MLA improved catheter securement and dressing integrity, a larger trial is needed to evaluate clinically important outcomes such as infection, catheter loss, and other complications. (See "Routine care and maintenance of intravenous devices", section on 'Device securement'.)

European Respiratory Society guidelines on benign pleural effusion (February 2025)

The European Respiratory Society has published its first consensus guideline on the management of nonmalignant pleural effusion [25]. The guideline emphasizes the moderate specificity of Light's criteria for distinguishing exudates from transudates and promotes the use of alternate tests (eg, pleural fluid lactate dehydrogenase, cholesterol); when heart failure (HF) is suspected, they suggest using the serum-effusion albumin gradient and N-terminal pro-brain natriuretic peptide (NT-BNP) levels. The guideline also supports a low threshold for pleural interventions (eg, therapeutic thoracentesis, indwelling catheter) for refractory pleural effusions due to HF, hepatic hydrothorax, and end-stage kidney failure and for biopsy for benign asbestos-related pleural effusions. We agree with these recommendations. (See "Management of nonmalignant pleural effusions in adults", section on 'Introduction' and "Pleural fluid analysis in adults with a pleural effusion".)

Advanced respiratory support in COVID-19 (January 2025)

In patients with COVID-19 who need advanced respiratory support, choosing between high-flow oxygen delivered via nasal cannulae (HFNC) and noninvasive ventilatory (NIV) support is challenging. In a recent trial of 1800 patients with acute respiratory failure from COVID-19 randomized to HFNC versus NIV, similar rates of endotracheal intubation (approximately 30 percent) and death (12 percent) were reported at the end of one week [26]. Subgroup analyses were not helpful in modality selection. In patients with COVID-19 who need advanced respiratory support, we use the patient's comorbidities and the tolerability of the device to help choose between NIV and HFNC. (See "COVID-19: Respiratory care of the nonintubated hypoxemic adult (supplemental oxygen, noninvasive ventilation, and intubation)", section on 'Choosing oxygen via high-flow nasal cannulae versus noninvasive ventilation'.)

Intensity level of noninvasive ventilation in acute exacerbation of COPD (December 2024)

In patients with acute hypercapnia due to acute exacerbations of chronic obstructive pulmonary disease (AECOPD), bilevel noninvasive ventilation (NIV) reduces mortality and intubation rates, but the ideal level of pressure support is uncertain. In a recent randomized trial of patients with AECOPD who tolerated an initial six hours of low-intensity NIV, patients assigned to receive continued low-intensity NIV (peak inspiratory airway pressure [IPAP] <20 cm H₂O) were more likely to meet criteria for intubation compared with those assigned to high-intensity NIV (peak IPAP 20 to 30 cm H₂O; 13.7 versus 4.8 percent, respectively) [27]. However, the rates of actual intubation were similar in both groups. Crossover from the low- to the high-intensity group may have prevented intubation but complicates interpretation of the study. In addition, the study was stopped early for benefit and underpowered for any mortality difference. Until more data are available, we continue to initiate bilevel NIV at low levels and increase as needed to relieve respiratory distress or improve ventilation. (See "Noninvasive ventilation in adults with acute respiratory failure: Benefits and contraindications", section on 'Acute exacerbation of chronic obstructive pulmonary disease with hypercapnic respiratory acidosis'.)

Nonsurgical biopsy modalities for peripheral pulmonary nodules (December 2024)

The optimal non-open surgical biopsy modality for peripheral pulmonary nodules (PPNs) is unclear. One recent network meta-analysis reported that computed tomography (CT)-guided transbronchial needle aspiration (CT-TBNA) had the highest diagnostic yield (89 percent), followed by robot-assisted bronchoscopy (RAB; 85 percent) and radial endobronchial ultrasound (rEBUS; 72 percent) [28]. However, CT-TBNA had the highest rate of pneumothorax requiring chest tube drainage (1.6 percent versus <1 percent for RAB or rEBUS) and bleeding (5.2 percent versus <1.5 percent RAB or rEBUS). This study summarizes two important factors that we take into consideration when choosing a non-open surgical biopsy modality for PPNs. Additional factors include PPN location and local expertise. (See "Image-guided bronchoscopy for biopsy of peripheral pulmonary lesions", section on 'Image-guided bronchoscopy techniques'.)