Latest news with #AnnalsofInternalMedicine
TimesLIVE
12 hours ago
- Health
- TimesLIVE
Medical journal rejects Kennedy's call for retraction of vaccine study
An influential US medical journal is rejecting a call from US health secretary Robert F Kennedy Jnr to retract a large Danish study that found aluminium ingredients in vaccines do not increase health risks for children, the journal's editor told Reuters. Kennedy has long promoted doubts about vaccines' safety and efficacy and as health secretary has upended the federal government's process for recommending immunisation. A recent media report said he has been considering whether to initiate a review of shots that contain aluminium, which he said are linked to autoimmune diseases and allergies. The study, funded by the Danish government and published in July in the Annals of Internal Medicine, analysed nationwide registry data for more than 1.2-million children over more than two decades. It did not find evidence that exposure to aluminium in vaccines had caused an increased risk for autoimmune, atopic or allergic or neurodevelopmental disorders. The work is by far the best available evidence on the question of the safety of aluminium in vaccines, said Adam Finn, a childhood vaccination expert in the UK and paediatrician at the University of Bristol who was not involved in the study. "It's solid, [a] massive dataset and high-quality data," he said.
Reuters
16 hours ago
- Health
- Reuters
Exclusive: Medical journal rejects Kennedy's call for retraction of vaccine study
Aug 11 (Reuters) - An influential U.S. medical journal is rejecting a call from Health Secretary Robert F. Kennedy Jr. to retract a large Danish study that found that aluminum ingredients in vaccines do not increase health risks for children, the journal's editor told Reuters. Kennedy has long promoted doubts about vaccines' safety and efficacy, and as health secretary has upended the federal government's process for recommending immunization. A recent media report said he has been considering whether to initiate a review of shots that contain aluminum, which he says are linked to autoimmune diseases and allergies. The study, opens new tab, which was funded by the Danish government and published in July in the Annals of Internal Medicine, analyzed nationwide registry data for more than 1.2 million children over more than two decades. It did not find evidence that exposure to aluminum in vaccines had caused an increased risk for autoimmune, atopic or allergic, or neurodevelopmental disorders. The work is by far the best available evidence on the question of the safety of aluminum in vaccines, said Adam Finn, a childhood vaccination expert in the UK and pediatrician at the University of Bristol, who was not involved in the study. "It's solid, (a) massive dataset and high-quality data," he said. Kennedy described the research as "a deceitful propaganda stunt by the pharmaceutical industry," and said the scientists who authored it had "meticulously designed it not to find harm" in a detailed Aug. 1 opinion piece on TrialSite News, an independent website focused on clinical research. He called on the journal to "immediately retract" the study. "I see no reason for retraction," Dr. Christine Laine, editor in chief of the Annals and a professor of medicine at Thomas Jefferson University, said in an interview. The journal plans to respond to criticism the article has received on its website, Laine said, but it does not intend to respond directly to Kennedy's piece, which was not submitted to the Annals. The lead author of the study, Anders Peter Hviid, head of the epidemiology research department at the Statens Serum Institut in Denmark, defended the work in a response post to TrialSite. He wrote that none of the critiques put forward by Kennedy were substantive and he categorically denied any deceit as implied by the secretary. "I am used to controversy around vaccine safety studies - especially those that relate to autism, but I have not been targeted by a political figurehead in this way before," Hviid said in an emailed response to Reuters. "I have confidence in our work and in our ability to reply to the critiques of our study." Kennedy had a number of critiques, including the lack of a control group, that the study deliberately excluded different groups of children to avoid showing a link between aluminum and childhood health conditions - including those with the highest levels of exposure - and that it did not include the raw data. Hviid responded to the criticisms on TrialSite. He said some of the points were related to study design choices that were reasonable to discuss but refuted others, including that the study was designed not to find a link. In fact he said, its design was based on a study led by Matthew Daley, opens new tab, a pediatrician at Kaiser Permanente Colorado, which did show a link, and which Kennedy cited in his article. There was no control group because in Denmark, only 2% of children are unvaccinated, which is too small for meaningful comparison, Hviid added. The data is available for researchers to analyze, but individual-level data is not released under Danish law, he said. Other prominent vaccine skeptics including those at the antivaccine organization Kennedy previously ran, Children's Health Defense, have similarly criticized the study on the Annals site. TrialSite staff defended the study for its scale, data transparency and funding while acknowledging the limitations of its design, a view seconded by some outside scientists. Laine said that while some of the issues Kennedy raised in his article may underscore acceptable limitations of the study, "they do not invalidate what they found, and there's no evidence of scientific misconduct." An HHS spokesman said the department had "no further comment than what the secretary said."
Hindustan Times
18 hours ago
- Health
- Hindustan Times
ChatGPT-generated diet plan landed this man in the hospital with rare poisoning- Should you trust AI for health?
With the growing use of Artificial Intelligence (AI) technology across industries, many professionals fear losing their jobs in the coming years. Several AI experts and tech CEOs have claimed that several job titles will be replaced, including some in the healthcare industry. But can AI replace human doctors in the foreseeable future? Well, this new incident will convince people why human intervention is crucial when it comes to human health. Should you rely on ChatGPT for health advice? This case from New York may change your mind.(Pexels) Recently, an elderly man from New York relied on ChatGPT for a healthy diet plan, but ended up in the hospital with a rare poisoning. These cases raise serious concerns about relying on AI for medical advice and why consulting a medical professional is crucial in a world where AI and humans exist together. ChatGPT diet plan gives the user a rare poisoning According to an Annals of Internal Medicine: Clinical Cases report, a 60-year-old man from New York ended up in the ER after following a diet plan generated by ChatGPT. It was highlighted that the man, who has any prior medical history, relied on ChatGPT for dietary advice. In the diet plan, ChatGPT told the man to replace sodium chloride (salt) with sodium bromide in day-to-day food consumption. Believing that ChatGPT can not provide incorrect information, the man followed the substitution and diet suggested by the AI chatbot for over 3 months. He purchased Bromide from an online store and used it as a salt substitute, making major changes to the body. Little did he know, Bromide is considered to be toxic in heavy dosage. Within the 3 months, the man experienced several neurological symptoms, paranoia, hallucinations, and confusion, requiring urgent medical care. Eventually, he ended up in the hospital, where doctors diagnosed him with bromide toxicity, which is said to be a rare condition. Not only was he sick, but he also showed signs of physical symptoms such as bromoderma (an acne-like skin eruption) and rash-like red spots on the body. After three weeks of medical care and restoring electrolyte balance, the man finally recovered. But it raised serious concerns of misinformation from AI chatbots like ChatGPT. While, AI chatbot can provide a great deal of information, it is crucial to check the accuracy of facts or take professional guidance before making any health-related decisions. The technology is yet to evolve to take the place of human doctors. Therefore, it is a wake-up call for users who use ChatGPT for every health-related query or advice.
Digital Trends
a day ago
- Health
- Digital Trends
ChatGPT advice pushed a man into psychosis seen in the 20th century
Earlier this year, an uplifting story detailed how a mother turned to ChatGPT and discovered that her son was suffering from a rare neurological disorder, after more than a dozen doctors had failed to identify the real problem. Thanks to the AI chatbot, the family was able to access the required treatment and save a life. Not every case of ChatGPT medical evaluation leads to a miraculous outcome. The latest case of ChatGPT doling out misleading medical advice ended up giving a person a rare condition called Bromide Intoxication, or Bromism, that leads to various neuropsychiatric issues such as psychosis and hallucinations. Trust ChatGPT to give you a disease from a century ago. A report published in the Annals of Internal Medicine describes a case involving a person who landed himself in a hospital due to bromism after seeking medical advice from ChatGPT regarding their health. The case is pretty interesting because the 60-year-old individual expressed doubt that their neighbour was discreetly poisoning them. Recommended Videos The whole episode began when the person came across reports detailing the negative impact of sodium chloride (aka common salt). After consulting with ChatGPT, the individual replaced the salt with sodium bromide, which eventually led to bromide toxicity. 'He was noted to be very thirsty but paranoid about water he was offered,' says the research report, adding that the patient distilled their own water and put multiple restrictions on what they consumed. The situation, however, soon worsened after being admitted to a hospital, and evaluations were conducted. 'In the first 24 hours of admission, he expressed increasing paranoia and auditory and visual hallucinations, which, after attempting to escape, resulted in an involuntary psychiatric hold for grave disability,' adds the report. Don't forget the friendly human doctor The latest case of ChatGPT landing a person in a pickle is quite astounding, particularly due to the sheer rarity of the situation. 'Bromism, the chronic intoxication with bromide is rare and has been almost forgotten,' says a research paper. The use of bromine-based salts dates back to the 19th century, when it was recommended for curing mental and neurological diseases, especially in cases of epilepsy. In the 20th century, bromism (or bromide toxicity) was a fairly well-known problem. The consumption of bromide salts has also been documented as a form of sleep medication. Over time, it was discovered that the consumption of bromide salts leads to nervous system issues such as delusions, lack of muscle coordination, and fatigue, though severe cases are characterized by psychosis, tremors, or even coma. In 1975, the US government restricted the use of bromides in over-the-counter medicines. Now, the medical team that handled the case could not access the individual's ChatGPT conversations, but they were able to obtain similar worryingly misleading answers in their test. OpenAI, on the other hand, thinks that AI bots are the future of healthcare. 'When we asked ChatGPT 3.5 what chloride can be replaced with, we also produced a response that included bromide. Though the reply stated that context matters, it did not provide a specific health warning, nor did it inquire about why we wanted to know, as we presume a medical professional would do,' the team reported. Yes, there are definitely cases where ChatGPT has helped a person with health issues, but we can only expect positive results when the AI is provided detailed context and comprehensive information. But despite that, experts suggest that one should exercise extreme caution. 'The ability of ChatGPT (GPT-4.5 and GPT-4) to detect the correct diagnosis was very weak for rare disorders,' says a research paper published in the Genes journal, adding that ChatGPT consultation can't be taken as a replacement for proper evaluation by a doctor. The biggest hurdle, obviously, is that the AI assistant can't reliably investigate the clinical features of a patient. Only when AI is deployed in a medical environment by certified health professionals can it yield trusted results.
Medscape
3 days ago
- Health
- Medscape
Aug 08 2025 This Week in Cardiology
Please note that the text below is not a full transcript and has not been copyedited. For more insight and commentary on these stories, subscribe to the This Week in Cardiology podcast , download the Medscape app or subscribe on Apple Podcasts, Spotify, or your preferred podcast provider. This podcast is intended for healthcare professionals only. In This Week's Podcast For the week ending August 8, 2025, John Mandrola, MD, comments on the following topics: Listener feedback on SURPASS CVOT, AF ablation and the limits of meta-analyses, a Watchman alert from FDA, and oral anticoagulation choices in elderly patients. A listener sent me a huge meta-analysis published in the Annals of Internal Medicine comparing catheter and surgical ablation to no ablation in randomized controlled trials (RCTs). I had seen the study but ignored it, but when avid listeners say they are keen to know my thoughts, I relooked at it and will discuss it now. A group at Washington University in St Louis led the study. It was a massive meta-analysis encompassing 63 trials with more than 11,000 patients. There were 39 catheter ablation trials and 24 surgical ablation trials. The goal was to bring together all RCTs in the field to assess the primary outcome of ischemic stroke at more than 30 days. Pause there. Stop and Think. This is weird. Why ischemic stroke after 30 days. What if a stroke happens at day 29? How is that less important than a stroke at 31 days? The authors also looked at secondary outcomes of mortality and heart failure hospitalizations. Six news outlets picked up the study—all with positive headlines. But we need to look deeper. The main results were that: Compared with medical therapy, catheter ablation reduced risks for ischemic stroke after 30 days (relative risk [RR], 0.63 [95% CI, 0.43 - 0.92]), mortality (RR, 0.73 [CI, 0.60 - 0.88]), and heart failure hospitalization (RR, 0.68 [CI, 0.55 - 0.85]). However, catheter ablation increased the RR for ischemic stroke at or before 30 days (6.81 [CI, 1.56 - 29.8]) such that the RRs were 0.77 (CI, 0.55 - 1.09) for any ischemic stroke and 0.77 (CI, 0.57 - 1.05) for all strokes. Surgical ablation reduced the RRs for ischemic stroke (0.54 [CI, 0.34 - 0.86]) and stroke from any cause (0.54 [CI, 0.35 - 0.82]) but had uncertain benefit for other outcomes. Ischemic stroke after 30 days in surgical ablation trials — RR was 0.63 (CI, 0.37 - 1.06); and non-significant for mortality and HHF. This is lacking, right? You want to know total stroke, mortality and HHF numbers. You also want to know absolute risk reductions. So here they are: For catheter ablation: Total stroke had a non-significant 33% relative risk reduction (RR, 0.77 (0.57-1.05) ARR 0.6% fewer per 1000 to 1 per 1000 more cases. Total mortality had a significant 37% relative risk reduction (RR, 0.73 (0.6-0.88); ARR was 16 per 1000 fewer. HHF had a 32% relative risk reduction (RR, 0.68 (0.55-0.85) and the ARR was 24 fewer per 1000 HF hospitalization For surgical atrial fibrillation (AF) ablation vs cardiac surgery alone: Total Stroke (Any Cause): Relative Risk: 0.54 (95% CI, 0.35 - 0.82; Absolute difference: 25 fewer per 1000 (35 fewer to 10 fewer) As I said above there were not significant reductions in HHF or mortality with surgical AF ablation. The authors concluded: 'Catheter ablation reduced the risks for ischemic stroke at more than 30 days, mortality, and HF hospitalization. Surgical ablation had uncertain benefit, except for stroke.' A neutral reader could conclude that total stroke, mortality and HHF all trended lower for catheter ablation. I am quite surprised that the accompanying editorial for Beth Isreal Boston key opinion leaders was quite positive. I am not at all positive about this meta-analysis. And I am surprised by how many key opinion leaders in AF there were as sub-authors. I think the main, perhaps only, value of this meta-analysis is how it displays the weaknesses of meta-analyses and what should not be done. Before I mention the specifics of this analysis, it's a good time to do some basics of meta-analyses, which combine studies to, hopefully…ideally…come to a summed effect size. Meta-analyses in my mind can really shine when individual trials are underpowered because of low event rates. If 2 or 3 or 4 trials are similar, you can combine and increase the power. I think of NOAH and ARTESIA, the trials of DOAC in sub-clinical atrial fibrillation. Each had low stroke rates; each were similar and combining them increases power. Another example is the risk of AF with fish oil. Each trial of fish oil showed an elevated risk of AF. Why, I do not know, but together the signal of AF is stronger. Another area where meta-analyses may be helpful is finding a dose-response relationship, like in the fish oil and AF example: trials with the highest fish oil concentration had the highest risk. Another area of meta-analysis might be when individual trials show conflicting results. Here, if the trials are similar, you might sort out the direction and magnitude of effect. But inherent in deriving any benefit from meta-analyzing trials, you need the combined trials to enroll similar populations who have similar interventions, ideally, with similar outcomes. You want the included studies to be high in quality, and of the same clinical scenario. For instance, it's madness to combine ICD trials from the 2000s to now—as HF therapy has changed so much. This meta-analysis fails in nearly all of these facets. It's so weird that this many smart academics thought it a good idea. First problem…the endpoint is mind-boggling. You don't look at stroke before and after 30 days. You look at stroke overall. One group gets a procedure that could increase short term risk of stroke (manipulation of the left atrium) but might reduce stroke in the long-term, by decreasing AF. You only care about stroke—overall. The example in the left atrial appendage occlusion (LAAO) trials where the only positive result was excluding the week after the procedure. This is ridiculous, because no patient can exclude the time after a procedure. Second problem….the authors combined trials from 20 years ago to trials done now. This is almost equally ridiculous, because modern ablation is very different from the hocus pocus we did in 2004. Third problem…the vast majority of included studies had a handful of patients. Combining small studies is a recipe for overestimating effect size. Combining 10, 20, 30, or 40 single-center studies cannot come close to estimating effect size compared with a 2000-patient strong CABANA trial. It's foolish to combine tiny studies with massive studies. Fourth problem…the authors looked at outcomes that were not primary outcomes of the major trials. Without individual patient level data from each trial, who knows if the results are accurate. Fifth problem…the trials varied too much in trial conduct, endpoints, time of study, to combine them together. Heterogeneity is off the charts. In sum, this massive effort yields exactly zero new information. I mean no malice to the authors, but this was destined to fail. Meta-analyses have a role, but this is not it. Catheter ablation should be studied; we should have proper primary endpoints, such as stroke, death, and total (not heart failure) hospitalizations. If you want to study quality of life, there is only one way: you must have a placebo or sham control arm. Otherwise you fail as hard as the tricuspid valve interventionalists have failed. I think air embolism here is not likely related exclusively to the Watchman device; it sounds more like a physiologic phenomenon wherein patients on conscious sedation can take these huge breaths, which, if timed unfortunately during opening of the valve of the sheath could suck in air. We worry about this a lot during AF ablation, and sheath management is crucial. I always remove catheters from sheaths slowly. And I hate catheter exchanges. It's one of the negatives of pulsed field ablation (PFA) technology now, that we have to use a separate mapping catheter and PFA catheter. That will soon change. What Boston Scientific is recommending is basic procedural hygiene when in the LA. I wonder, though, whether this new warning has to do with the new system or is it just more doctors, many of whom are low-volume operators, doing more cases because of the device's popularity. I mean there have been tens of thousands of cases of LAAO since 2016, and now in 2025, we get an alert on specific catheters. If I were an implanter, I would be curious about this question. We've done this procedure for a long time, but now, with the new systems, we see a problem. Why? Are we sure about those valves? I am asking, I don't know. That said, FDA warnings are common with procedures. We learn things during the iterative process—for all procedures. For instance, we did not know about atrioesophageal fistula after AF ablation until tragedies occurred. Thermal ablation evolved to avoid heat stacking in the area of the esophagus. I am glad companies issue such warnings. Transparency is laudable. The larger lesson here is that procedures can cause harm. Their evidentiary basis should be as strong as a new drug. Often it is not. AF ablation has only one placebo controlled trial. Tricuspid interventions have zero, and you all know how I feel about LAAO procedures: Failure of the seminal Watchman trials to show non-inferiority to warfarin. And no compelling evidence for the device in patients not able to take oral anticoagulants. None. And it's been 9 years. But…But….when we intervene on patients with symptomatic AF, or symptomatic tricuspid regurgitation, or symptomatic myocardial infarction, we intervene on patients with an active problem. When we intervene on the LAA, we are doing prevention. We seek probabilistic benefit in the future. Prevention must always have a higher not lower bar of evidence. It doesn't with LAAO, and that is blemish on our profession. Journal of the American College of Cardiology has a nice paper on the issue of switching oral anticoagulants in older patients who have frailty. This comes up a lot. Anticoagulant- naive patients, at least in my neighborhood, get a direct oral anticoagulant (DOAC). I can hardly recall the last time I started warfarin de novo—in any patient with AF. We do this mostly for convenience, but also for efficacy and safety. The issue at hand is whether a patient on warfarin should be switched to DOAC. I reported in 2023 on the FRAIL-AF study, a Dutch study that randomized 1330 patients who were age ≥75 years, frail, and treated with a vitamin K antagonist (VKA) to either switch from VKA to a DOAC (rivaroxaban 54%, apixaban 19%, edoxaban 18%, or dabigatran 9%) or to continue VKA. The trial was stopped prematurely for futility as there was a significant increase in major or clinically relevant nonmajor (CRNM) bleeding, mainly gastrointestinal bleeding, in patients switched to DOAC (HR: 1.69; 95% CI: 1.23-2.32) This was provocative because it was surprising. You would not have thought bleeding would have gone so much higher when switching. Critical appraisal of FRAIL-AF included the very wide confidence intervals, and low number of events, made wider by the early termination of the trial for futility. The TIMI group, led by Dr Robert Gugliano, has formed a COMBINE-AF collaboration, which is patient-level meta-analysis of the four DOAC vs warfarin trials. RELY, ARISTOTLE, ROCKET and ENGAGE. That's a lot of data, and this is a much better use of meta-analyses than the first paper I discussed on AF ablation. This substudy of COMBINE-AF set out to replicate FRAIL-AF in a larger population of the four trials. It's a neat analysis. Here is what they did: They took patients who were on VKA, the so-called VKA experienced arm, when entering the trial, and elderly (older than 75 years old), and frail. For frailty, the investigators had to make a frailty scale based on a number of factors (which is different from the scale used in FRAIL-AF). Also note: VKA and warfarin are almost interchangeable. This group — warfarin-experienced, old, and frail — were then randomized in the trials to DOAC or warfarin. There were about 6000 of these FRAIL-AF like patients in the four trials. Split equally into DOAC or warfarin arms. The authors called these the test groups. They also looked at patients without these characteristics. These were either VKA naive, or non-elderly or non-frail. There were 52,000 of these, and the authors called these the non-test group. The main outcomes were stroke/systemic embolism, major bleeding, and mortality. They also looked at GI bleeding and ICH, but I think these are not as important as the big categories: stroke, bleeding, mortality. Here were the findings: In the FRAIL-AF like population, over 2.5 years: there was a 17% reduction in stroke and systemic embolism that did not reach statistical significance. Absolute risk reduction (ARR) was 0.4%, no difference in mortality and no difference in major bleeding. In the FRAIL-AF like population, randomization to DOAC vs warfarin was associated with an 83% increase in GI bleeding that was significant, and 1.5% absolute risk increase, but a 54% reduction in fatal bleeding, though the ARR was only 0.3%, of course because there were fewer events. When looking at net clinical outcome in the FRAIL-AF-like group the HR was right at 1.01. When compared to the non-FRAIL-AF population, the 52,000 patients who did not have one of the three criteria of VKA naive, elderly, or frail. These results were exactly similar to what you would expect and what was seen in the main trials: DOAC has statistically significant reductions in the risks of stroke and systemic embolism (SE) vs warfarin (HR, 0.81), death (HR, 0.91), and major bleeding (HR, 0.82). GI bleeding was increased with DOAC but it was only a 23% increase rather than the 83% increase. Net clinical outcome was statistically significant at HR 0.89 for DOACs. The authors did a special analysis based on type of DOAC in the FRAIL-AF like group. Recall that in FRAIL-AF, the main DOAC that patients on warfarin were switched to was rivaroxaban. In COMBINE-AF it was apixaban and edoxaban. When the authors simulated this in their data, they found that when using the 'same DOAC mix as FRAIL-AF (rivaroxaban 54%, apixaban 19%, edoxaban ,18% and dabigatran 9%) they found no statistical differences in stroke/SE but now a statistically significant increase in major bleeding HR 1.21 and even larger 2.2x increase in GI bleeding. But on the other hand, when they did an apixaban and edoxaban-only analysis in this FRAIL-AF group, they found similar stroke/SE rates but now a 17% significant reduction in major bleeding and only a 36% non-significant increase in GI bleeding. Like FRAIL-AF, switching from warfarin to DOAC had no sig effect on stroke or SE. Unlike FRAIL-AF, which had to be terminated early for bleeding, the COMBINE-AF FRAIL-AF-like group did not sustain higher rates of major bleeding when switching. GI bleeding was higher with the switch, but fatal bleeding was less. FRAIL-AF therefore does not replicate in this data set. The authors dispute the Euro guidelines which give a new Class IIb recommendation to maintain warfarin in elderly frail patients based on FRAIL-AF. They write (strongly, I would add): 'The findings from this study suggest that switching to a DOAC is a reasonable strategy for frail, elderly, and VKA-experienced patients, particularly for reducing the risk of stroke, systemic embolism, death, and the most severe bleeding events, such as intracranial and fatal bleeding.' First, I think it's worth thinking about why the results were divergent. One that the authors put a lot of weight on is the different frailty scores used in the two trials. COMBINE authors say their patients might have been frailer, or at least different enough to drive results. I am not sure this is true. To be in a global trial requires a degree of robustness. FRAIL-AF, however, included social isolation, inability to walk around in the home, and cognitive impairment, which were not available in the COMBINE-AF data set. Close your eyes and imagine an isolated Dutch patient vs a patient well enough to come in the office and be randomized in a global RCT. Another possibility is that DOAC mix is important. The subanalysis with apixaban and edoxaban looked a lot better for bleeding than did rivaroxaban heavy analysis. But I would be extremely suspect: this is a sub-analysis of a sub-analysis that is not adjusted for multiplicity. I think invoking specific mixes of DOAC is a stretch. The main reason for the difference, in my opinion, comes in Figure 4 of the supplement where the COMBINE authors provide the numbers of events along with Forest plots in the two studies For FRAIL-AF there were only 29 stroke events vs 201 in the COMBINE-AF study. For FRAIL-AF there were only 40 major bleeds vs 551 in COMBINE-AF. If you have access to the paper, take a look at the width of the error bars. It's remarkable. This is not a criticism of FRAIL-AF; the authors did a nice trial of a very special patient population and were looking for a reduction in bleeding. When they found the opposite —an increase — it made ethical sense to stop the trial early. Yet I think the lessons here from the two trials are that single trials with small numbers of events should be translated carefully. You could argue that the Euro guideline authors choice of a weak IIb recommendation for remaining on warfarin is careful. I was enthusiastic when FRAIL-AF came out because it was so darn provocative. I would not have predicted it. And I think it still is a great trial, but this analysis tells us that a) we should be cautious of small numbers of events, and b) as it so often is in clinical medicine, perhaps there is no exact right answer in these patients. So, how I see the clinical question given this new data is that, in sum, for elderly frail patients on warfarin, I don't think we are mandated to do any one specific thing. FRAIL-AF said don't switch when doing well on warfarin. COMBINE says that if you switch, outcomes are similar. There is not a major advantage or disadvantage. But if an elderly, frail patient wanted to switch, say for convenience, COMBINE results would argue that outcomes would be similar. Hence, switching should neither be proscribed nor recommended. It can be a shared decision with the patient and his or her family. Recall also that this is not about initial choice, which I think remains a DOAC. This is about patients on warfarin and considering switching.
