The Curbsiders podcast

#445 DIGEST-Hotcakes: Preventive PCI, Andexanet Alfa for ICH, Aspirin for Fatty Liver, Naltrexone in Cirrhosis, Pivmecillinam for UTI, Cefepime vs Pip-Tazo, MDRO Decolonization, Microplastics and MACE, Pax-NO-vid

June 24, 2024 | By



Transcript available via YouTube

Join us as we review recent practice-changing articles on preventive PCI, andexanet alfa for ICH, aspirin for fatty liver, naltrexone in cirrhosis, pivmecillinam for UTI, cefepime vs pip-tazo, MDRO decolonization, microplastics and MACE, and more! Fill your brain hole with a delicious stack of hotcakes! Featuring Paul Williams (@PaulNWilliamz), Rahul Ganatra (@rbganatra), Nora Taranto (@norataranto) and Matt Watto (@doctorwatto).

Claim CME for this episode at!

Episodes | Subscribe | Spotify | Swag! |Mailing List | Contact | CME!

Show Segments

  • Intro, disclaimer
  • Preventive PCI
  • Andexanet alfa for ICH
  • Aspirin for MASLD
  • Naltrexone for AUD in cirrhosis
  • Pivmecillinam for UTI
  • MDRO decolonization
  • Cefepime vs pip/tazo
  • Microplastics
  • Pax-NO-vid
  • Outro

Hotcake #1 Preventive PCI (Watto)

Park SJ, et al; PREVENT Investigators. Preventive percutaneous coronary intervention versus optimal medical therapy alone for the treatment of vulnerable atherosclerotic coronary plaques (PREVENT): a multicentre, open-label, randomised controlled trial. Lancet. 2024 May 4;403(10438):1753-1765. doi: 10.1016/S0140-6736(24)00413-6. Epub 2024 Apr 8. PMID: 38604213. 

What was the research question? Does optimal medical therapy (OMT) with percutaneous intervention (PCI) of non-flow limiting (fractional flow reserve >0.8), high-risk, vulnerable plaques found on intracoronary imaging, prevent major adverse cardiac events, including death, better than OMT alone?

Why is this study important? COURAGE (Boden, 2007) did not find a benefit of PCI vs OMT, but did not specifically consider vulnerable plaques, which are believed to be a common culprit in fatal thrombotic coronary events (Borovac, 2024).    

How was the study done? This was an open-label, randomized controlled trial at 15 research hospitals in S. Korea, Japan, Taiwan, and New Zealand of patients 18 years or older with non-flow limiting (FFR >0.8), vulnerable coronary plaques based on intracoronary imaging. 

Who were the patients? 5627 patients were screened, and 1606 were enrolled (28.6% of those screened), with n=803 in the PCI+OMT group and n=803 in the OMT group. Follow-up occurred until the last patient reached 2 years after randomization with a 4-year median and 7-year maximum follow-up. Patients were 64-65 yo on average, 25-29% were female, most had HTN and HLD, and stable angina. 

Top-line results: This was a positive trial. The primary outcome (death from cardiac causes, target vessel myocardial infarction, ischemia-driven target vessel revascularization, or hospitalization for unstable or progressive angina) was reduced in the intervention group HR 0.54 (95% CI 0.33 to 0.87). At 2 years (the primary timepoint), there were 3 events in the PCI+OMT group and 27 events OMT group 0.4% and 3.4%, respectively, a 3% absolute difference, NNT of 33. The 3% absolute difference held at 4 years and 7 years follow up. 

Learning points & limitations: Outcomes overall were rare in this trial – probably because most patients had stable CAD and were on good background medical therapy. While preventive PCI did reduce MACE, there was no significant reduction in all-cause death or MI. About 1 in 4 patients screened were ultimately enrolled, raising the possibility that these results may not generalize to an unselected population. 

Bottom line/Hotcakes rating: 3/5. On one hand, I’m excited by the prospect of more individualized and targeted PCI, but I’m worried that this trial’s judicious use of PCI for vulnerable plaques might be hard to replicate in practice given the sophisticated intracoronary imaging techniques. 

Further reading:

Hotcake #2 ANNEXA-I (Rahul)

Andexanet for Factor Xa Inhibitor–Associated Acute Intracerebral Hemorrhage | New England Journal of Medicine

What was the research question? Is andexanet alfa, a recombinant inactive factor Xa analog, safe and effective for limiting hematoma expansion among patients with DOAC-associated intracerebral hemorrhage?

Why is this study important? Although DAOCs are thought to lead to less intracranial bleeding than vitamin K antagonists like warfarin, the lack of a targeted reversal agent is part of why the morbidity and mortality of intracranial bleeding while on DOACs, when it does occur, remains high. Andexanet alfa has shown promise in reducing hematoma expansion in an uncontrolled study from the same group (ANNEXA-4), but had not been evaluated in an RCT (Connolly, 2019). 

How was the study done? Patients with intracranial hemorrhage (smaller than <60 mL) and without severe neurologic deficits (NIHSS <35) who received a factor Xa inhibitor within the previous 15 hours were randomized 1:1 to receive andexanet alfa as an initial low-dose or high-dose bolus, with the dosing determined by the dose, type, and timing of most recent dose of DOAC, followed by a continuous infusion for 2 hours, or usual care. 85% of patients in the usual care group got prothrombin complex concentrate. The primary outcome was assessed at 12 hours, and neurologic outcomes and deaths assessed out to 30 days. 

Who were the patients? 550 adults with a mean age in the upper 70s were randomized; 20 patients were removed due to problems with consent, so the safety analysis was done in 530, and the efficacy analysis was done in 452 patients. The most frequent DOAC used was apixaban in 60%, the median hematoma size was 10 mL, the median NIHSS was 9, and bleeds were preceded by trauma in 10-15%.

Top-line results: This was (technically) a positive trial with respect to the primary outcome: Among patients with intracerebral hemorrhage on a DOAC, Andexanet alfa, compared with usual care, was associated with a 13% absolute increase in hemostatic efficacy, a composite endpoint defined as hematoma expansion of no more than 35%, worsening in NIHSS of no more than 7 points, and no requirement for rescue therapies within 12h. However, there were no differences between groups in deaths or neurologic function at 30 days (although the trial was not powered to assess these), but there was an absolute increase in thrombosis of 5%, primarily ischemic stroke.  

Learning points & limitations: There are several sources of chance and bias that threaten these results. First, the primary outcome, which was a composite of surrogate outcomes, was changed during the trial (see the protocol) to make it broader (that is, more events could contribute to the primary outcome). Second, the early stopping rule was also changed to make it more permissive (applying the original stopping rule to the new primary outcome wouldn’t have allowed stopping). Finally, the most serious concern I have was that there was clear evidence of harm: a 5% absolute increase in thrombosis (particularly ischemic stroke) in patients who got andexanet alfa.

Bottom line/Hotcakes rating:  Although it was technically positive for its primary outcome, based on the clear evidence of harm, I can’t recommend use of this drug in the population in which it was studied. 1 of 5 hotcakes.

Further reading:

Hotcake #3  Aspirin for MASLD (Nora)

Simon TJ et al; Aspirin for Metabolic Dysfunction Associated Steatotic Liver Disease Without Cirrhosis. JAMA 2024; 331(11):920-929. doi:10.1001/jama.2024.1215. 

What was the research question? Does a daily baby aspirin (81 mg) reduce hepatic fat content at 6 months compared with placebo in people with MASLD? 

Why is this study important? Metabolic dysfunction-associated steatotic liver disease (MASLD) affects >30% of US adults and can progress to fibrosis, cirrhosis and HCC. While there is an FDA-approved drug for patients with at least moderate fibrosis (resmetirom, approved in March 2024 for NASH), there are no approved medications to treat early steatosis.  

How was the study done? This was a phase II, randomized, double-blind, placebo-controlled trial comparing aspirin with placebo in individuals with MASLD (confirmed by imaging or biopsy). Patients were randomized to either once-daily aspirin 81 mg or identical placebo pills. The primary outcome was mean absolute change in hepatic fat at 6-month follow-up, measured by magnetic resonance spectroscopy. 

Who were the patients? 80 people with MASLD (mean age, 48, 55% women, mean BMI 34 kg/m2, mean hepatic fat content 35%, corresponding to moderate steatosis) but without cirrhosis were randomized; people with significant alcohol use and other causes of liver disease were excluded. 

Top-line results: This was a positive trial: the difference in mean absolute change in hepatic fat from baseline to 6 months was 10.2% in favor of aspirin. Aspirin also significantly reduced relative hepatic fat content (mean difference: 39%), reduced hepatic fat measures by MRI, and more people on aspirin achieved a >30% relative reduction in hepatic fat (mean difference: 30%). 

Learning points & limitations: The aspirin group was heavier and had more baseline visceral adipose tissue at baseline, which could bias towards the treatment appearing more effective. This is a good illustration of how small trials are vulnerable to imperfect randomization due to chance alone. Finally, the primary outcome was a surrogate outcome, and it is unclear if disease progression (e.g. to advanced fibrosis) or patient-centered outcomes (symptomatology, mortality) will be affected – more studies are needed.  

Bottom line/Hotcakes rating: = 3.5. I love new uses for old medications, so this trial excites me. However, because it is phase II, small, and uses surrogate endpoints, whether aspirin improves patient-centered liver outcomes remains unknown. 

Further reading:

Hot Cake #4 – Naltrexone remains very good (Paul)

Thompson R, Taddei T, Kaplan D, Rabiee A. Safety of naltrexone in patients with cirrhosis. JHEP Reports. 2024;6(7):101095.

What was the research question?  Is naltrexone to treat alcohol use disorder safe in patients with compensated or decompensated cirrhosis?

Why is this study important?  Treatment of alcohol use disorder in patients with cirrhosis is critical because continued alcohol increases risk of decompensation and mortality. However, naltrexone is under-utilized in this population, owing to concerns over potential hepatotoxicity.  

How was the study done? Veterans enrolled in an outpatient registry from 2008 to 2021 with a diagnosis of cirrhosis (based on ICD codes) were included. Patients newly initiating naltrexone after a diagnosis of cirrhosis and with liver tests collected within 3 months were identified. A tool called the Roussel Uclaf Causality Assessment Method (RUCAM) criteria was used to assign risk of drug-induced liver injury (DILI) among patients with biomarker elevations (>5x ULN for ALT, >2x ULN for AP) in the first 3 months of naltrexone initiation.  

Who were the patients? 2,940 patients (mean age 58; 96% male, alcohol as etiology of cirrhosis in 61%, hepatitis C in 32%) were included. Among them, 31% had a prior history of decompensation, 15% were Child-Pugh class B or C at the time of naltrexone initiation, and 6% had jaundice. Only 1.4% received intramuscular naltrexone.

Author’s claim/Top-line results: Only 2% of patients with labs checked in 3 months following naltrexone initiation had significant liver enzyme elevation. Of these, 48% were classified as “DILI excluded,” and 52% were classified as “DILI unlikely.” No deaths or new decompensations were attributed to naltrexone.

Learning points & limitations: Among the 62 patients with liver enzyme elevation, alternative causes were identified in 77% of them (the most frequent cause was alcohol). These data suggest naltrexone is safe in this population. Two limitations are the retrospective design, since the RUCAM criteria requires information that may have not been easily discernible from chart review, and that we cannot draw strong conclusions about IM naltrexone, since only a small minority of patients received it.

Bottom line/Hotcakes rating:  Naltrexone is safe in patients with liver disease. 4/5 hotcakes

Further Reading:

Hot take #1 Pivmecillinam for UTIs (Watto) 

FDA approves pivmecillinam for uncomplicated UTI treatment

Summary: Pivmecillinam is a penicillin class antibacterial and prodrug of mecillinam, the active antibacterial moiety (moi-eh-tee) included in the 2010 IDSA guidelines as a first-line agent, but only just approved by the FDA in April 2024 for the treatment of uncomplicated urinary tract infection (UTI). Note, while the dose recommended by IDSA is pivmecillinam 400 mg twice daily, the package insert for the FDA-approved compound recommends pivmecillinam 185 mg three times a day; both IDSA and FDA recommend 3 to 7 days as clinically indicated. 

Adverse events and prescribing information: The package insert lists nausea and diarrhea as the most common side effects. Severe allergic reactions have been described, and pivmecillinam should be avoided in patients with a severe allergy to beta-lactam drugs and in those with disordered carnitine metabolism. Pivalate, which is excreted in the urine as pivaloylcarnitine, is generated when taking pivmecillinam. Consider alternative antibacterial therapies in patients at risk for carnitine depletion (e.g., patients with significant renal impairment or decreased muscle mass) and avoid concurrent treatment with valproic acid/valproate or other pivalate-generating drugs due to increased risk of carnitine depletion. Use of pivmecillinam prior to delivery may cause a false positive test for isovaleric acidemia on newborn screening!

Bottom line: It is a welcome alternative to the other first-line drugs for UTI (nitrofurantoin, trimethoprim-sulfamethoxazole, and fosfomycin), though the cost will likely be prohibitive initially.

Further Reading:

Hot take #2 MDRO Decolonization (Nora)

Gussin et al. Reducing Hospitalizations and Multidrug-Resistant Organisms via Regional Decolonization in Hospitals and Nursing Homes. JAMA. 2024;331(18):1544-1557. doi:10.1001/jama.2024.2759. 

Summary: SHIELD-OC was a public health, quasi-experimental, quality improvement initiative to reduce MDROs in healthcare facilities in Orange County, CA. 19 nursing homes (NH) and long-term acute care hospitals (LTACHs) and 16 hospitals participated in the QI initiative for MDRO prevention, with a 25-month baseline period, a 4-month phase in period, and a 25-month intervention period (2017 – 2019). The intervention was universal decolonization at NH/LTACHs, using chlorhexidine bathing on admission and routinely afterwards for usual bathing (bath or shower 3x/week at nursing homes, daily in LTACHs or hospitals), as well as twice daily nasal iodophor for 5 days on admission, and then M-F every other week. Hospitals adopted targeted decolonization (5 days chlorhexidine baths, twice daily nasal iodophor) for all non-ICU patients on contact precautions. Participating facilities were compared across baseline and intervention periods. 

Bottom line: MDRO colonization was significantly lower at LTACHs, NHs, and hospitals after implementation of the protocol compared with baseline. There was also a significant decrease in MDRO-positive clinical cultures (a 30% reduction) at participating NHs compared with non-participating NHs. The rate of infection-related hospitalization decreased during the intervention for participating but not for non-participating nursing homes (~27% reduction), and costs were reduced. This study adds to the literature using large-scale, multi-institution, quality improvement methodology that decolonization is time consuming, but seems to reduce MDRO incidence–and the consequences of them–when it’s actually done.  

Further Reading:

Hot take #3 The Cefepime vs Pip-tazo debate continues (Rahul)

Mortality of Patients With Sepsis Administered Piperacillin-Tazobactam vs Cefepime | Less is More | JAMA Internal Medicine 

Summary: This study tested the hypothesis that empiric therapy with piperacillin/tazobactam for patients with sepsis but without a clear indication for anaerobic coverage would, compared with empiric therapy with cefepime, be associated with higher 90-day mortality. The idea is that depletion of anaerobes leads to dysbiosis and worse clinical outcomes. In terms of the top-line results, this was a positive study: using a shortage of pip-tazo as an opportunity for a natural experiment, the authors found that among patients with sepsis without an indication for anaerobic coverage, use of pip-tazo was associated with a 5% absolute increase in mortality at 90 days (22.5% vs 17.5%) compared with cefepime.  

Bottom line: Despite this being a well-done target trial emulation, it uses retrospective observational data, which is vulnerable to bias, and unfortunately there is lots of evidence of confounding by indication that threatens our interpretation of these results. First, the raw data didn’t reveal a significant mortality difference (20.8% for outside the shortage and 19.8% within the shortage, p=0.3). I also worry that physicians prescribe anaerobic coverage for patients who appear sicker, which could partially explain these results. Ultimately, I give this 3 out of 5 hotcakes because it’s a good paper and a well-done study, but I don’t think it has persuaded me that confounding by indication doesn’t explain these results. 

Further reading: 

Hot take #4 Microplastics – delicious treat, or harbinger of cardiac doom? (Paul)

Microplastics and Nanoplastics in Atheromas and Cardiovascular Events  – Marfella et al. NEJM.

Summary: Prior studies have linked microplastics and nanoplastics (MNPs) to cardiovascular disease through increased oxidative stress and inflammation, but the role that MPNs play in formation of atherosclerotic plaques in humans is poorly understood. In this Italian study, researchers assessed the presence of MNPs in atherosclerotic plaques in 257 patients undergoing intervention for asymptomatic carotid artery stenosis. Excised plaques were examined by gas chromatography-mass spectroscopy, electron microscopy, and stable isotope analysis, and outcomes were observed prospectively and compared between patients with and without evidence of MNPs in their plaques. Patients with MNPs were younger, more likely to be men; less likely to have hypertension; more likely to have diabetes, cardiovascular disease, and dyslipidemia; and more likely to smoke. A primary endpoint event (MI, stroke, or death from any cause) occurred in 7.5% of patients without MNPs, compared with 20% of patients with MNPs. This corresponded to a higher risk of a primary endpoint event in the MNP group (HR 4.53; 95% CI, 2.00 to 10.27; P<0.001).

Bottom line: Among patients with asymptomatic high-grade carotid artery stenosis undergoing endarterectomy, those patients with evidence of MNPs in their carotid plaques were more likely to have a heart attack, stroke, or die, but whether this association is causal remains unclear. The letters to the editor raised alternate hypotheses, including the possibility of contamination in the lab or from prior endovascular stents, and that macrophages present in plaques may have migrated from elsewhere in the body. Regardless, the presence of MNPs certainly seems to be a marker for increased risk for cardiovascular disease and death. The other question is: what do we do with this information? We can’t advise people to not eat plastic; it’s everywhere. Perhaps at some point this may factor into risk factor reduction for certain populations, and it does point to some interesting theoretical possibilities that are too abstract for my concrete brain.

Further Reading

  • Correspondence on Microplastics and Nanoplastics in Atheromas – various contributors and author response

Hot take #5 Pax-NO-vid (Watto)

Paul Sax on “The Rise and Fall of Paxlovid” 

Summary: In June 2022, Pfizer sent a press release with negative interim results from EPIC-SR and in December 2022 they increased the sample size. Since then, we’ve known that nirmatrelvir/ritonavir (Paxlovid) does not improve symptoms or prevent hospitalization or death for low-risk patients with COVID19 infection. However, was not updated with results until August 2023, and EPIC-SR was not published until April 2024. I saw indirect advertisements for Paxlovid during the fall of 2022 (World Cup, World Series, and NFL games) and direct advertisements in the fall-winter 2023-2024. These advertisements made it seem like Paxlovid was recommended for any adult with COVID19 symptoms, which has generated a large number of calls and inappropriate prescriptions despite risk of rebound symptoms, drug interactions, taste disturbance, cost, and probable lack of benefit. 

Bottom line: We have to earn patients’ trust and the last four years of COVID have been a mess. I have a high threshold to prescribe Paxlovid and reserve it for my oldest, sickest, and highest-risk patients. We will look out for the results of the PANORAMIC study (enrolling through Sep 2024), but I am not hopeful. 

Further Reading:


Listeners will review recent practice-changing articles and medical news.

Learning objectives

After listening to this episode listeners will…

  1. Describe the evidence in favor of stenting non-hemodynamically significant coronary stenoses and its limitations
  2. Recommend for or against the use of andexanet alfa to treat DOAC-associated intracerebral hemorrhage
  3. State the effect of aspirin on hepatic fat content in patients with MASLD
  4. Characterize the risk of drug-induced liver injury from naltrexone in patients with cirrhosis


The Curbsiders report no relevant financial disclosures. 


Ganatra RB, Taranto N, Williams PN, Watto MF. “#445 DIGEST-Hotcakes: Preventive PCI, Andexanet Alfa for ICH, Aspirin for Fatty Liver, Naltrexone in Cirrhosis, Pivmecillinam for UTI, Cefepime vs Pip-Tazo, MDRO Decolonization, Microplastics and MACE, Pax-NO-vid”. The Curbsiders Internal Medicine Podcast. Final publishing date, June 24, 2024.


  1. June 25, 2024, 11:12am Kathleen Haggerty, MD Post Covid writes:

    It is important that you continue to keep in mind the indications for Paxlovid and prescribe it for adults over 50 or who are otherwise at increased risk of infection from COVID infection. Paxlovid is highly effective at preventing hospitalization and death and continues to be widely under prescribed as it requires some considerable care and trouble. Paxlovid does not cause rebound. COVID causes rebound. Paxlovid does not change the incidence of that phenomenon occurring. There is no evidence that the SARS-COV2infection is any less virulent. We are just seeing most cases in an immunized population. Also, recall that this virus even in un immunized people presented as anything from asymptomatic to deadly. Even now, we see COVID deaths continue. Don’t minimize this illness or it’s proper and important therapies.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Episode Sponsors


You can try Freed for free right now by going to And listeners of Curbsiders can use code
CURB50 for $50 off their first month.


Learn more about ACP MKSAP at

Litter Robot

Whisker is currently offering $50 off Litter-Robot bundles. As a special offer to listeners of the show, go and use promo code CURB to save an additional $50 on any Litter-Robot bundle.

Episode Credits

Written and Hosted by: Rahul Ganatra MD, MPH; Nora Taranto MD; Paul Williams, MD, FACP, Matthew Watto MD, FACP
Cover Art: Matthew Watto MD, FACP
Reviewer: Rahul Ganatra MD, MPH; Emi Okamoto MD
Technical Production: Pod Paste
Showrunners: Matthew Watto MD, FACP; Paul Williams MD, FACP

CME Partner


The Curbsiders are partnering with VCU Health Continuing Education to offer FREE continuing education credits for physicians and other healthcare professionals. Visit and search for this episode to claim credit.

Contact Us

Got feedback? Suggest a Curbsiders topic. Recommend a guest. Tell us what you think.

Contact Us

We love hearing from you.


We and selected third parties use cookies or similar technologies for technical purposes and, with your consent, for other purposes as specified in the cookie policy. Denying consent may make related features unavailable.

Close this notice to consent.