questionable practices

[J-Rad]

Advertisement - Members don't see this ad

Mid West.

 

[gutonc]

"Native MW family"? Honestly, what is that? Milwaukee? Otherwise, I'm not privy.

Midwest. Which means that, aside from that one time they took a Disney cruise in 1994, they haven't been further than 100 miles from the town they were born in.

EDIT: Not sure how I missed J-Rad's post.

 

[iridesingltrack]

As opposed to the manufacturer's sponsored studies that "clearly" demarcated the morbidity and mortality of the adverse effects, but treated any positive gain as a complete recovery? I mean come on the original studies are FAR more biased than Hoffman...

tPA unfortunately doesn't work well and for populations as a whole is a bad choice. There are individual patients where it may be a good choice (the 40 year old with the sudden, devastating stroke and no co-morbid conditions), but overall the data are VERY weak, especially when applied to the septo-, octo- and nono-generians that frequent our departments.

Yes. I believe that the methods of Hoffman's paper result in more bias and misunderstanding of the data than even the "original studies." The methods in THAT one paper are horrible and are not suitable for clinical descisions. My point is better pointed out by reading the editorial in the same issue of annals by Silbergleit: A Hundred (and One) Graphs Tell Only Part of the Story. Annals of Emergency Medicine, Volume 54, Issue 3, September 2009, Pages 342-343. He demonstrates that using the same technique AND outcome he can show that tPA works.

As for your further comments: I agree tPA doesn't work particularly well for some individuals...however, it clearly does ok when taken in the aggregate of all people given tPA. The cochran meta-analysis of almost 39 studies, and nearly 8,000 pts shows this to be true. But, on an individual basis: those that have more comorbidities, do poorly; some die. Question: do you think they will have better outcomes if you don't treat with tPA?

tPA is not some holy grail. Clearly, it should be respected. I guess I fall in camp that hopes something better comes along, but in the meantime I will continue to use it if indicated because it is the ONLY arrow I have in my quiver for acute ischemic stroke.

cheers,
iride

 

[deleted6669]

MEL HERBERT'S TPA RANT:
http://www.emrap.tv/index.php?option=com_content&view=article&id=2232

 

[Rendar5]

tPA is not some holy grail. Clearly, it should be respected. I guess I fall in camp that hopes something better comes along, but in the meantime I will continue to use it if indicated because it is the ONLY arrow I have in my quiver for acute ischemic stroke.

cheers,
iride

I'm actually lucky in that I work at an institution that has a neurosurgeon who endovascularly removes clots. Works further out than tPA too 😀 Unfortunately these docs are much rarer than their cardiovascular colleagues.

 

[xaelia]

I'm actually lucky in that I work at an institution that has a neurosurgeon who endovascularly removes clots. Works further out than tPA too 😀 Unfortunately these docs are much rarer than their cardiovascular colleagues.

Man, I hope it's not the MERCI device.

Stroke. 2005 Jul;36(7):1432-8. Epub 2005 Jun 16.
Safety and efficacy of mechanical embolectomy in acute ischemic stroke: results of the MERCI trial.
PMID: 15961709

AJNR Am J Neuroradiol. 2006 Nov-Dec;27(10):2048-52.
Endovascular mechanical clot retrieval in a broad ischemic stroke cohort.
PMID: 17110664

Stroke. 2008 Apr;39(4):1205-12. Epub 2008 Feb 28.
Mechanical thrombectomy for acute ischemic stroke: final results of the Multi MERCI trial.
PMID: 18309168

Stroke. 2007 Mar;38(3):967-73. Epub 2007 Feb 1.
The impact of recanalization on ischemic stroke outcome: a meta-analysis.
PMID: 17272772

The original MERCI achieved 46% recanalization rate - which compares to the stated spontaneous rate of 24.1% and IV TPA of 46.2% in the meta-analysis. 13% of the cases were complicated by device malfunctions, and 7% of their study population suffered additional harms from the device. 27% had favorable mRS at 90 days, but 43% died.

Now, they are enrolling only the most severe strokes for these trials - their mean NIHSS is ~20 +/- 6.6. I did have a little trouble finding good subgroup analyses of other stroke studies for comparison, but as a reference, ECASS III NIHSS >20 subgroup has 16% of favorable mRS with TPA and 8% with placebo. On the other hand, the mortality of that subgroup in ECASS III is 22% and 21%.

The MERCI authors concluded their trial was a negative study, stating that while the MERCI device demonstrated it was technically capable of performing recanalization, recanalization with the MERCI device was not associated with improved outcomes - which is a little interesting, since the meta-analysis states that recanalization had an OR of 4.4 of being associated with a good outcome. But in any event, they did what any manufacturer-supported group does after a negative study - more studies!

Multi-MERCI was stopped once in 2005 due to patient harms from protocol violations. The newer device showed a non-significant trend towards better recanalization (55%). Oddly enough, multi-MERCI reports their baseline NIHSS as a median of 19, interquartile range 15-23, so it's tough to compare to the severity of illness of the original MERCI's mean of 20 - but I would suggest their baseline was better. In any event, multi-MERCI had non-significant trends towards improvement in clinically significant procedural complications (5.5%), good mRS (36%) and mortality (34%).

As another comparison, the SITS-MOST review of widespread TPA administration, with 6000+ patients with median NIHSS of 12, good mRS was achieved in 55% and only 11% died.

In both MERCI and multi-MERCI, the studies are funded by the manufacturer, several authors have ownership stakes in the device, and the inventor of the device is even one of the authors. Both studies also evaluate >1000 patients for deployment before hand-picking 100-odd to enroll.

Far too many issues for me to evaluate the efficacy of this treatment and the appropriate population in which to apply it.

 

[deleted65604]

Back to the issue of strep pharyngitis. Say you find legitimate, double-blind placebo-controlled trials where antibiotics don't change outcomes, that only proves that the patients in the particular region of the country where that study was conducted aren't at risk for valvular heart disease. It doesn't say anything about MY local population. I live in a small enough community that we simply don't do much epidemiology or microbiologic studies. I will never have the study in my hand that proves that my ER patients don't need antibiotics for their strep pharyngitis.

 

[jbar]

Doesn't that go for anything? You could say about any research that since it wasn't done at your institution that the research doesn't apply to your patients. Where do you draw the line about what EBM you are willing to accept and what you are going to ignore and go with your gut?

 

[deleted65604]

Ideally, local antibiotic choices should be made on local epidemiology.

I'm looking at it from a patient satisfaction standpoint, and from a medicolegal standpoint.

The lawyers aren't going to care that the strains of a certain strep species cultured in a city thousands of miles away a decade ago didn't cause RHD "Because my client is sitting in front of you and HE did."

The parents of the kid who is febrile, and has tonsils the size of his testicles and can't drink aren't going to buy your statement that, "strep doesn't need antibiotics."

 

[Apollyon]

The parents of the kid who is febrile, and has tonsils the size of his testicles and can't drink aren't going to buy your statement that, "strep doesn't need antibiotics."

If the tonsils were the size of the testicles in a 5 year old kid, then there would be no problems.

 

[xaelia]

The lawyers aren't going to care that the strains of a certain strep species cultured in a city thousands of miles away a decade ago didn't cause RHD "Because my client is sitting in front of you and HE did."

I don't feel it's as much an issue that one of your patients may go on to develop RHD (considering they could do so even with antibiotics), and you can't be faulted for failing to treat something that, at this point, is a 1 in 100,000 or greater occurrence - unless it's currently the standard of care to treat strep with antibiotics. Whether it's appropriate and EBM or not, if your prescribing practice deviates from the standard, they will have no difficulty finding experts willing to discuss your negligence.

 

[The right Path]

I don't feel it's as much an issue that one of your patients may go on to develop RHD (considering they could do so even with antibiotics), and you can't be faulted for failing to treat something that, at this point, is a 1 in 100,000 or greater occurrence - unless it's currently the standard of care to treat strep with antibiotics. Whether it's appropriate and EBM or not, if your prescribing practice deviates from the standard, they will have no difficulty finding experts willing to discuss your negligence.

And that's exactly why we still do the Solumedrol gtt for our SCI patients, not because our neurosurgeons believe it is beneficial.

 

[Stitch]

I will never have the study in my hand that proves that my ER patients don't need antibiotics for their strep pharyngitis.

I actually agree. If you find it you should treat it. My argument is that you shouldn't be (generally) be looking for strep. That way there's nothing to treat.

The parents of the kid who is febrile, and has tonsils the size of his testicles and can't drink aren't going to buy your statement that, "strep doesn't need antibiotics."

If you mean huge tonsils and you're looking at a younger kid (under 3 or 4), then I'd argue you need to expand your differential beyond strep and look for something else. Don't just jump to a rapid strep test. If you're seeing cough, hoarseness, or runny/stuffy nose, then it's most likely viral. Retropharyngeal abscess, croup, that sort of thing should be in your mind as well.

I've had the conversation about strep and rarity of RHD along with the overperscription of antibiotics. Generally parents are receptive if you give them an alternative diagnosis and treatment. A little decadron at 0.6mg/kg goes a long way to making the kid feel better, which can satisfy parents just fine.

Will I turn it into a fight if the parents are dead set on strep testing? No, but most people can be steered away from it.

Whether it's appropriate and EBM or not, if your prescribing practice deviates from the standard, they will have no difficulty finding experts willing to discuss your negligence.

True, which is why I say treat it if you've found it. However keep in mind that you're probably more likely to be sued for SJS than for rheumatic fever given the relative rarity of both entities.

 

[xaelia]

Stumbled upon another article that I thought might be an interesting piece of the TPA puzzle:
Early stroke treatment with IV t-PA associated with early recanalization.
J Neurol Sci. 2010 Aug 15;295(1-2):53-7. Epub 2010 Jun 8.
PMID: 20570289

138 consecutive stroke patients in Japan eligible for TPA, being evaluated to see if outcomes were improved if reperfusion was achieved more rapidly. Less interesting than their primary endpoint (in which they say their hypothesis was supported - and the earlier TPA was given to stroke onset, the more likely the rapid reperfusion), are two other items noted in their methods and results. First, they found on pre-treatment MRA that 33 of the patients they were planning on giving TPA to had no arterial occlusion, and thus, did not need TPA. So, they only gave TPA to 102 patients who had arterial occlusion on MRA. Second, while summarizing their outcomes, they mention only 20.6% of cohort had good outcomes (mRS 0-2) - which is worse than the placebo group in essentially every other stroke study, and far worse than the 36.1% (mRS 0-1) from the original clinical trial (industry supported) for TPA in Japan (PMID: 16763187). 37.4% had ICH on 24-hour follow-up MRI, while they say only one was symptomatic - though three patients with the worst outcomes were so bad off after TPA that they couldn't get a 24-hour MRI to evaluate for ICH.

Their conclusion is that TPA should be given as rapidly as possible to get the best results. The more interesting read that I have is - my goodness, a quarter of the patients we're planning TPA for don't have an arterial occlusion at all! And, then, once you have a stroke cohort that subtracts all the patients who were going to do well without any intervention, good outcomes in 20% with TPA doesn't compare favorably at all.

 

[WilcoWorld]

However keep in mind that you're probably more likely to be sued for SJS than for rheumatic fever given the relative rarity of both entities.

I'm just speculating here, but it seems unlikely to me that one could successfully sue for SJS unless the defendant gave a patient with a known allergy a contraindicated med.

 

[doctorFred]

I'm just speculating here, but it seems unlikely to me that one could successfully sue for SJS unless the defendant gave a patient with a known allergy a contraindicated med.

i think the supposed line of reasoning would go something like: "you prescribed a potentially DANGEROUS drug to my client, when the data shows that it was not indicated for the treatment of his condition! and now poor billy's skin is all over the kitchen floor, which we need to re-grout. and here's the bill for that too."

 

[docB]

I gotta go with Wilco on this one. Every dim bulb on the jury will know, from years of conditioning, that you're supposed to get antibiotics for sore throat. The plaintiff will have the unenviable job of preaching data to *****s.

 

[Stitch]

I'm just speculating here, but it seems unlikely to me that one could successfully sue for SJS unless the defendant gave a patient with a known allergy a contraindicated med.

You may well be right. I just meant that if we're going to worry about possible litigation, then it seemed just as likely to be sued for not testing for strep as it would to have prescribed amox. Again, I agree that you should treat it if you find it, I simply feel we should be looking for some other source in general in that age group.

The devil is in the details of course, but there's a feeling that a black box warning should be enough to tell us not to prescribe these drugs. There was one case of penicillin caused SJS. The physician was sued because he didn't test for strep but treated anyway. There's another on there that described 'excessive dosing' of amox.

 

[xaelia]

One of the things that Hoffman was rambling on about in the recent EMA was his frustration with the definition/clinical significance of "Serious Bacterial Illness" in children, as well as the zealotry with which we pursue UTIs in our well-appearing fever without a source infant patients. I think he made a statement to the effect of "where are all the infants sick with pyelonephritis from their missed UTIs if this is such a big deal".

My little literature search turned up a few articles in Pediatrics saying that prophylactic antibiotics in children with recurrent UTIs doesn't prevent renal cortical scarring, and that early antibiotic administration for children with pyelonephritis doesn't prevent renal cortical scarring, and a bunch of articles discussing the false positive rates of urine dip and micro - but I couldn't find much about the natural history of pediatric UTI, since I presume there isn't a randomized controlled trial in which UTIs were left untreated and followed.

So, I guess my question, is there any evidence our there that it's a reasonable strategy not to aggressively cath/bag every well-appearing febrile infant (greater than, say, 3 months of age)?

 

[Stitch]

One of the things that Hoffman was rambling on about in the recent EMA was his frustration with the definition/clinical significance of "Serious Bacterial Illness" in children, as well as the zealotry with which we pursue UTIs in our well-appearing fever without a source infant patients. I think he made a statement to the effect of "where are all the infants sick with pyelonephritis from their missed UTIs if this is such a big deal".

My little literature search turned up a few articles in Pediatrics saying that prophylactic antibiotics in children with recurrent UTIs doesn't prevent renal cortical scarring, and that early antibiotic administration for children with pyelonephritis doesn't prevent renal cortical scarring, and a bunch of articles discussing the false positive rates of urine dip and micro - but I couldn't find much about the natural history of pediatric UTI, since I presume there isn't a randomized controlled trial in which UTIs were left untreated and followed.

So, I guess my question, is there any evidence our there that it's a reasonable strategy not to aggressively cath/bag every well-appearing febrile infant (greater than, say, 3 months of age)?

I posted on this a while ago and left some evidence; I'll see if I can dig it up.

There are definitely mixed feelings on this issue. The renal guys tend to think they kidney is always infected, and there's some evidence to support that. Several studies have looked at UTI as being a concominant infection in RSV and non RSV bronchiolitis as well as URI with fever. That said, the criteria were (if I'm remembering correctly) fever greater than 39 for at least 48 hours. In those studies/settings the rate of UTI, regardless of what else was going on, was somewhere around 5-6%.

You're absolutelly right about the prophylactic antibiotics not preventing renal scarring. In fact some urologists strongly feel it just selects out for resistant bugs and makes things worse. However many nephrologists disagree and are prescribing bactrim or amox in low doses. I'm glad I don't have to get into that fight.

The main issue is the feeling that small children/infants cannot opsonize well. If they have an infection in one place it can more easily spread to other regions. That's why we get blood, urine and CSF on anyone who shows up at our door. 😛

The real question I guess is why not wait for the culture and decide from there instead of treating based on a UA. Is there harm in waiting? I don't know of any data that's looked at that specifically, but I'd imagine your practice could depend upon your setting. If it's private practice, or a kid that's coming in anyway, waiting to treat is probably ok. In the ED setting, you may never see the kid again, and if he comes back a week later near dead in shock from sepsis, I promise someone will point the finger at you for 'missing' the dirty urine when they saw you last.

EDIT: I do think it's reasonable to have this discussion with a reasonable family. If you can get the kid seen the next day by their PCP and make sure the kid looks ok, you could do nothing. You can offer a urine and let the family know that it may not be high yield. Just document well.

Also: don't get bag urines. You'll hate yourself when it looks dirty or grows mixed flora. Just cath the kid. Does anyone still do suprapubic taps?

 

[Stitch]

Okay, found it. None address your specific question, but it's food for thought. Discuss.

Here's a literature review on why we should care.

Here's a CHOP study looking at presence of UTI in younger febrile children presenting to the ED.
And two articles on finding UTIs in the presence of RSV here and here . Note the in the presence of RSV serious bacterial infection, including meningitis appears to be less likely.
Another review suggesting UA be checked in fever without source.
And yet another review noting that UA is recommended in febrile children under 36 months.
Prevalence of SBI and UTI in febrile childen with non RSV 'clinical' bronchiolitis.
Australian study documenting presence of UTI in febrile children 3-36 months. Estimated around 5%.