PEITHO Trial Teaches an Important Lesson

The current issue of the New England Journal of Medicine contains an important trial- the PEITHO trial. Its important because it tells us what not to do.

In the PEITHO trial patients with intermediate risk pulmonary embolism (right ventricular dysfunction and myocardial injury with no hemodynamic compromise) were randomized to a single weight-based bolus of tenecteplase or placebo. All patients were given unfractionated heparin. Patients were followed for 30 days for the primary outcome of death from any cause or hemodynamic decompensation within 7 days after randomization.

This table shows the efficacy outcomes. Looks promising doesn’t it.

PEITHO efficacy outcomes

The primary outcome was significantly reduced by 56%. This composite outcome is not a good one though. Patients would not consider death and hemodynamic decompensation equal. Also the pathophysiology of the 2 outcomes can be quite different. The intervention should also have a similar effect on all components of a good composite and there is a greater effect on hemodynamic decompensation than death. Thus, don’t pay attention to the composite but look at the composite’s individual components. Only hemodynamic decompensation was significantly reduced (ARR 3.4%, NNT 30). Don’t get me wrong this is a good thing to reduce.

But with all good can come some bad. This trial teaches that we must pay attention to adverse effects. The table below shows the safety outcomes of the PEITHO trial. Is the benefit worth the risk?

PEITHO safety outcomes

You can see from the table that major extracranial bleeding was increased 5 fold (ARI 5.1%, NNH 20) as was stroke, with most of them being hemorrhagic (ARI 1.8%, NNH 55).

This trial teaches a few important EBM points (I will ignore the clinical points it makes):

  1. You must always weigh the risks and benefits of every intervention.
  2. Ignore relative measures of outcomes (in this case the odds ratios) and calculate the absolute effects followed by NNT and NNH. These are much easier to compare.
  3. Watch out for bad composite endpoints. Always look at individual components of a composite endpoint to see what was affected.

I’m Still Not Crazy About the Pooled Risk Equations in the New Cholesterol Guidelines

2 papers got published this week to further validate the pooled risk equations developed for the ACC/AHA Cholesterol Guidelines.
Muntner and colleagues used the REGARDS participants to assess the calibration and discrimination of the pooled risk equations. This study had potential as it oversampled patients from the stroke belt. This is important because the Pooled Risk Equations were developed  to overcome the limitations of the Framingham tool (mainly its lack of minorities).  I have a real problem with this study because the pooled risk equations estimate 10 yr risk of CHD and stroke and this study only has 5 yrs of follow-up for the REGARDS participants. I don’t think their estimates of calibration and discrimination are valid. Risk of CHD and stroke should increase over time so event rates could change with 5 more years of follow-up. The important thing this paper adds is the reminder that observational studies often lack active surveillance. Most observational studies rely on self report of outcomes and obviously silent events would be missed by the patient as would events for which the patient didn’t seek evaluation. Muntner and colleagues also used Medicare claims data to identify events not detected through routine cohort follow-up and found 24% more events. This is a useful lesson from this study.

In a more useful study Kavousi and colleagues compared 3 risk prediction tools (pooled risk equations, Framingham, and SCORE)  using the Rotterdam Study, a prospective population-based cohort of persons aged 55 yrs and older. This cohort does have 10 yrs of follow-up.


This figure shows that at each level of risk the pooled risk equations overestimated risk, though less so in women.

treatment rec

This figure shows the proportion of patients for whom treatment is recommended (red bars), treatment should be considered (yellow bars), and no treatment is recommended (green bars). As you can see the new risk tool leads to the large majority of men “needing treatment” compared to previous guidelines (ATP III) and the current European guidelines (ESC).

calibration curves

Finally, this figure shows the calibration curves and the calibration was not good. The blue dots should lie right upon the red line for good calibration. Furthermore, the c-statistic is 0.67 (a measure of discrimination which means the tool can differentiate diseased from nondiseased patients. A c-statitic above 0.7 is considered moderate to good. The closer to 1 the better).

Why might the pooled risk equations overestimate risk? Maybe they don’t if you believe the Muntner study. It could just be a problem with the lack of active surveillance in the cohort studies used to validate the tool. Or they really do overestimate risk because they aren’t accurate or maybe more contemporary patients receive better therapies that improve overall health or maybe the baseline risk characteristics of the validation cohorts just differ too much from the development cohorts.

I am still not sold on the new pooled risk equations but they might not be much better than what we have been using based on the Kavousi study (Framinham also overpredicted risk and had poor calibration). I think we need more study and tweaking of the tool or we use the tool as is and focus more on cardiovascular risk reduction (with exercise, diet, tobacco cessation, diabetes and HTN control) and don’t focus so much on starting a statin right away.

The Mayo Clinic has a nice patient decision aid that you can use to help patients decide if a statin is right for them:



Answering Clinical Questions at the Point of Care- Its Time to Stop Making Excuses!

Del Fiol and colleagues published a systematic review of studies examining the questions raised and answered by clinicians in the context of patient care.  The studies they examined used several methodologies including after-visit interviews, clinician self-report, direct observation, analysis of questions submitted to an information service, and analysis of information resource search logs. Each of these methodologies has their pros and cons. I’ll review their findings following the 4 questions that they asked.

How often do clinicians raise clinical questions? On average, clinicians ask 1 question for every 2 patients (range 0.16-1.85).

How often do clinicians pursue questions they raise? On average, they only pursued 47% (range 22-71%).

How often do clinicians succeed at answering the questions they pursue? They were pretty successful when they decided to pursue an answer: 80% of the time they were able to answer the question. Interestingly, clinicians spent less than 2-3 minutes seeking an answer to a specific question.  They were clearly choosing questions that could be answered fairly quickly when they decided to pursue the answer to a  question.

What types of questions were asked? Overall, 34% of questions were related to drug treatment while 24% were related to the potential causes of a symptoms, physical finding, or diagnostic test finding.

I find 3 other findings (from the Box in the manuscript) interesting:

  • Most questions were pursued with the patient still in the practice (not sure if the clinicians searched in front of the patient or left the room- more about this later)
  • Most questions, as you would expect, are highly patient-specific and nongeneralizable.  This is unfortunate for long-term learning.
  • Also unfortunately, clinicians mainly used paper and human resources (more on this in a minute)


Even though Del Fiol examined barriers to answering questions I refer to another study by Cook and colleagues that more closely examined barriers to answering clinical questions at the point of care. Cook did focus groups with a sample of 50 primary care and subspecialist internal medicine and family medicine physicians to understand barriers and factors that influence point of care learning/question answering. Of course the main barrier is time. This study was done in late 2011 into early 2012 and included a wide range of ages of participants. With the resources available on both the desktop and handheld devices this barrier should be declining, especially when you consider the most common question clinicians ask is about drug treatment.

Physicians frequently noted patient complexity as a barrier. Complex patients require more time and often lead to more complex questions that are harder to answer with many resources. Almost all guidelines and studies are focused on single disease patients. Multimorbidity is rarely covered. Thus many answers will likely rely on clinical expertise and judgment. This is where using human resources is likely to occur. I bet few of us question how up to date our colleagues are that we ask questions of.

Interestingly, Cook’s study participants identified the sheer volume of information as a barrier. As a result, these physicians used textbooks more than electronic resources. I wonder if they understand that a print textbook is at least a year out of date by the time it hits market. How often do they update their textbooks? (likely rarely…just look at a private practice doctor’s bookshelf and you will often see books that are at least 2 or more editions out of date).

Finally, the physicians in Cook’s study felt that searching for information in front of the patient might “damage the patient-physician relationship or make patients uncomfortable.” They couldn’t be more wrong. Patients actually like when we look things up in front of them.  I always do this and I tell them what I am looking up and admit my knowledge limitation. I show them what I found so they can participate in decision making. No one can know everything and patients understand that. I would be wary of a physician who doesn’t look something up.

So, how should a busy clinician go about answering clinical questions?

  1. You must have access to trustworthy resources.  2 main resources should suffice: a drug resource (like Epocrates or Micromedex- both are free and available for smartphones and desk tops) and what Haynes  labels as a “Summary” (Dynamed or UpToDate).  I leave guidelines out here (even though they are classified as a “Summary” resource) because most guidelines are too narrowly focused and many are not explicit enough in their biases.
  2. Answer the most important questions (most important to the patient #1 and then most important to improving your knowledge #2). If the above resources can’t answer your question and you must consult a colleague challenge them to support their opinion with data. You will learn something and likely they will too.
  3. Answer the questions you can in the available time. Many questions should be able to be answered in 5 min or less using the above resources. You are more likely to search for an answer to a question while the patient is in your office than waiting until the end of the day (the above cited studies can attest to that).
  4. Be creative in answering questions. I saw a great video by Paul Glasziou (sorry can’t remember which of his videos it was to link) where he discussed a practice-based journal club. Your partners likely are developing similar questions as your are. This is how he recommends organizing each journal club session: step 1 (10 min) discuss new questions or topics to research, step 2 (40 min) read and appraise (if needed) a research paper for last week’s problem, and step 3 (10 min) develop an action plan to implement what you just learned. This is doable and makes your practice evidence-based and feel somewhat academic. If you follow the Haynes hierarchy and pick the right types of journal articles (synopses and summaries) you can skip the appraisal part and just use the evidence directly.

Ultimately you have to develop a culture of answering questions in your practice. It has to be something you truly value or you won’t do it.  Resources are available to answer questions at the point of care in a timely fashion. At some point we have to stop making excuses for not answering clinical questions.

Should Natriuretic Peptide Guide Chronic CHF Therapy?

Guideline recommended heart failure care is generally followed by most primary care doctors. They prescribe ACE inhibitors, beta-blockers, and occasionally mineralocorticoid receptor antagonists. Despite prescribing these proven medications many patients remain symptomatic, are frequently admitted to the hospital and maybe even die prematurely. One of the main problems is that doctors often don’t titrate medications to the doses used in the landmark studies of these agents. There are several reasons for this but one is just plain lack of knowledge of what the target doses are. I know I am guilty of not teaching my house staff what the target doses are. We often start a medication in the hospital or titrate it slightly but leave the final dose adjustments to the outpatient physicians (we of course assume they will do what’s right). A missed teaching opportunity on my part which I will be correcting this month on my inpatient service. I am going to use a nifty app on my iPad called MyStudies (no affiliation). The app is free but the full complement of journal articles costs $10/year. I plan to use it to review landmark trials with my house staff and I am focusing on making sure they know the target doses of drugs. Most of us adjust med doses based on symptoms or physical findings. Could using a lab test (that seems to always get checked by my ER no matter what the complaint) help us primary care physicians do a better job taking care of CHF patients? This month in the European Heart Journal Troughton and colleagues published an individual patient data meta-analysis on using B-type natriuretic peptide to guide therapy in patients with chronic, largely systolic CHF.  Previous meta-analyses on this topic used study level data and had limitations in what they could adjust for- something that a patient-level meta-analysis is much better suited for. The methodology was fair but not Cochrane level (somewhat limited search strategy, not very explicit about how they did things, no information on publication bias, etc). They wanted to compare BNP guided therapy with clinically guided therapy and they found 9 studies with 2151 total patients that met their inclusion criteria (RCTs reporting all cause mortality). Despite the lack of statistical heterogeneity there is definitely clinical heterogeneity in their included studies ( different target levels of BNP, different study periods, different durations of follow up, different treatment algorithms).  Most studies were fairly small with 69-499  patients enrolled.


What did they find? BNP guided therapy reduced all-cause mortality by 3.4% (19.3% mortality in clinically guided therapy vs 15.9% in BNP guided therapy). This difference was only seen in those under 75 yrs of age as can be seen in this figure. Heart failure hospitalizations were also reduced by 4.7% (27.5% in clinically guided therapy vs 22.8% in BNP guided therapy). Interestingly, there were similar levels of decline in proBNP levels in both groups. I had anticipated BNP guided therapy would result in greater reductions in proBNP levels.

What explains their findings? There are several possible explanations:

  1. The most obvious is that patients in the BNP guided arm received more dose titrations of medications. They found that only ACE or ARBs doses increased in the included studies and only by 8.4%.  No dose increases were seen with beta-blockers or mineralocorticoid receptor antagonists. Loop diuretic doses also stayed the same. The authors found that increasing doses of each of these medications was significantly associated with reduced all-cause mortality.
  2. I wonder if cointerventions (like diet counseling, medication compliance counseling) were intensified in the BNP guided arms more so than in the clinically guided arms. This would not likely be captured in these studies and could explain lower event rates despite minimal increases in ACE inhibitor doses.
  3. Could other agents not measured (like digoxin) been added in the BNP guided arms. Digoxin does lower hospitalization rates. No information is given in the meta-analysis about this and I didn’t go back to the individual studies to see how they handled cointerventions.
  4. Referral to specialty care- this is possible I guess but one of the main effects would be greater titration of meds by cardiologists. ICDs or pacing could have also been done but again this would be related to cointerventions and I would hope this would not have been done differentially.
  5. BNP guided patients could have been seen more frequently by their providers- possible but this should mostly lead to dose titrations or lifestyle modification counseling.
  6. Hawthorne effect

I am not planning to order and follow BNPs on all my CHF patients. Why? Isn’t that what the evidence would say to do?  I think the key is not necessarily the lab test but making sure you titrate doses to those used in the studies. Dose increasing is all that these studies seem to suggest improves mortality.  I think BNP is just a reminder to do that. We need a study of BNP guided therapy vs a clinical pathway that titrates patients to goal doses independent of monitoring BNP. I suspect that study would show no differences in outcomes. You should also remember that BNP is not specific and is affected by renal function. None of the included studies enrolled patients with AKI or CKD (a problem I deal with often).

Intellectual Conflicts of Interest Are Harder to Deal With Than Financial

A colleague of mine wrote a blog post on conflicts of interest that I want to expand upon and discuss my recent experience with.

I am a member of a guideline panel (sponsoring organization not to be named) on screening for prostate cancer. I am the only primary care/internist on the panel (as best I can tell). The majority are urologists. Recently a revised guideline manuscript was sent around for comment. From my biased point of view (my intellectual conflict of interest) I was against several of the recommendations not only because the evidence didn’t strongly support it but also because I have to deal with the unhappy patients who undergo prostate cancer screening and are found to have something and ultimately get a procedure that worsens their quality of life. The urologists just couldn’t understand how I could be against screening all men and getting a baseline PSA at age 40. At one point I was referred to as “pathetic” that I would have such thoughts and teach my residents to follow the UPSTF recommendation against screening for prostate cancer in average risk men. Even the American Urological Association takes the stance to participate in shared decision making with men about prostate cancer screening.


So why all the push back from my urological colleagues? The easy answer is financial. They make money from prostate biopsies and the surgical and hormonal treatment of prostate cancer. But I think it goes deeper than that especially since many are academic urologists and probably don’t have as great a financial incentive to evaluate and treat more prostate cancer (though I could be wrong). I think their intellectual conflict of interest is the main problem. Their research and academic beliefs are so strong that prostate cancer screening is good that they can’t see anyone else’s point of view (or view those views are equally meritorious). They can’t understand how I give greater value to the risk side of the risk/benefit equation. At least financial conflicts of interest are visible (when disclosed) and understandable. Intellectual conflicts of interest are usually subconscious and hard to overcome as I have found out. It will be an interesting face-to-face meeting this fall when we get together for another update.

N-of-1 Trial for Statin-Related Myalgia: Consider Conducting These Studies in Your Practice

The March 4th edition of the Annals of Internal Medicine contains an article by Joy and colleagues in which they conducted an N-of-1 trial in patients who had previously not tolerated statins. This is important because patients often complain that they cannot tolerate statins despite needing them.  I have wondered how much of this was a self-fulfilling prophecy because they hear a lot about this from friends and various media outlets. A N-of-1 trial is a great way to determine if the statin-related symptoms are real or imagined.

First, lets discuss N-of-1 trials. What is a N-of-1 trial? It’s a RCT of active treatment vs. placebo in an individual patient. The patient serves as his or her own control thus perfectly controlling for a variety of biases. When is a N-of-1 trail most useful? This design is not useful for self-limited illnesses, acute or rapidly evolving illnesses, surgical procedures or prevention of irreversible outcomes (like stroke or MI). It’s most useful for conditions that are chronic and for with therapy is prolonged. It’s best if the effect you are looking for occurs fairly quickly and goes away quickly when treatment is stopped. These trials are a good way to determine the optimal dose of a medication  for a patient. They are also good to determine if an adverse effect is truly due to a medication. Finally, they are good way to test a treatment’s effect when the clinician feels it will be useless but the patient insists on taking it. How is a N-of-1 trial conducted? Get informed consent from the patient and make sure they understand that a placebo will be part of the study. Next the patient randomly undergoes pairs of treatment periods in which one period of each pair applies to the active treatment and one to placebo. A pharmacist will need to be involved to compound the placebo and to develop the randomization scheme (so as to keep clinician and patient blinded). Pairs of treatment periods are replicated a minimum of 3 periods. There needs to be a washout period between moving from active to placebo and vice versa. The length of the treatment period needs to be long enough for the outcome of interest to occur. Use the rule of 3s here (if an event occurs on average once every x days, then observe 3x days to be 95% confident of observing at least 1 event). What outcome should be measured? Most commonly these types of trials will be conducted to determine the effect of an intervention on quality of life type measures (eg pain, fatigue, etc).  Ask the patient what is the most troubling symptom or problem they have experienced and measure that as your outcome. Have the patient keep a diary or ask them to rate their symptoms on some meaningful scale at certain follow-up intervals. Do this while on active and placebo treatments. You will have to determine how much of a difference is clinically meaningful.  How do I interpret N-of-1 trial data? This can be a little difficult for non-statistically oriented clinicians. You could do the eyeball test and just see if there are important trends in the data. More rigorously you could calculate the differences in means scores of the placebo and active treatment periods. These would then be compared using a t test (freely available on the internet).

Back to Joy and colleagues N-of-1 trial on statins. They enrolled patients with prior statin-related myalgias. Participants were randomly assigned to get the same statin and dose that they previously didn’t tolerate or placebo. They remained on “treatment” for 3 week periods with 3 week washout periods in between. Patients weekly rated their symptoms on visual analogue scales for myalgias and specific symptoms (0-100, with 0 being no symptoms and 100 being the worst symptoms). It was felt a difference of 13 was clinically significant. What did they find? There were no statistically or clinically significant differences between statins and placebo in the myalgia score (4.37) nor on the symptom specific score (3.89). The neat thing the authors did was to determine if patients resumed taking statins after reviewing the results of their N-of-1 trial and 5 of the 8 patients resumed statins (one didn’t because a statin was no longer indicated).

So are statin related myalgias mostly in our patients’ heads? Maybe. This study is by no means definitive because it only enrolled 8 patients but it at least suggests a methodology you can use to truly test if a patient’s symptoms are statin related or not. This is important to consider because the most recent lipid treatment guidelines focus on using statins only and not substituting other agents like ezetimibe or cholestyramine. So give this methodology a try. You and your patients will likely be amazed at what you find.