DKA and Cerebral Edema. Are Fluids to Blame or is it Shock?

It's almost the end of the month and I am again behind schedule. But today I have something controversial and very interesting... Pediatric DKA. In particular the issue about fluid administration and risk for cerebral edema.

Pedi DKA is scary, dangerous and subject of heated debate when it comes to fluid administration. Most of the treatment protocols in pediatrics are very stingy when it comes to fluid administration. In severe cases, kids arrive in shock, hypotensive, tachycardic, severely vasoconstricted, and many of them comatose; the protocol says to give them just 10 ml/kg bolus and correct remaining deficits over 48 hrs. This doesn't make  sense to me. If you are in shock, you are not perfusing! Why would I want to drag the treatment of someone in shock? I have seen my fare share of sick pedi DKA's and the ones in the severe end of the spectrum look just like that, awful. In England you cannot give a 20 ml/kg initial bolus to a DKA kid in shock without being called to the medical director office and ordered to memorize the protocol and recite it the next day in front of everyone to hear.  The notion that rapid fluid administration is the cause of cerebral edema in severe DKA is so engraved in all the treatment protocols, but there is no solid evidence to support this idea, except for observational studies and consensus recommendations. I was really bothered by this and tried to find the science behind and found that it is not the fluid rate what is associated to the risk of cerebral edema, is the level of dehydration, acidosis with hypocapnia, hyperglycemia and brain hypoperfusion what causes all the problems. The kids who are going to develop cerebral edema are in the sickest end of the spectrum, probably arrive with cerebral edema and will become clinically obvious regardless of the fluid infusion rate. 

Last year, in the journal of Pediatrics (http://pediatrics.aappublications.org/content/131/1/e73) there was a RCT of 18 kids (I know is a small number, but this is randomized human data rather than rat studies). One group received 20 ml/kg bolus with correction of dehydration over 24 hrs. The second group received a 10 ml/kg bolus and correction of dehydration over a 48 hr period. The results showed no difference in MRI cerebral edema parameters at different treatment stages between the rapid fluid replacement approach compared with the slower infusion rates approach; and more importantly, MRI findings were consistent with vasogenic edema and were worse at the beginning of treatment compared with after-treatment in both groups. This suggests that sick DKA kids already have cerebral edema before initiation of therapy and edema improves after treatment independently of fluid infusion rates. Those who support the idea that rapid fluid infusion is the cause of cerebral edema in DKA treatment, say that it is due to rapid osmotic changes (we all have heard that), and there is probably some true about that. However, one important distinction to make here, is that osmotic cerebral edema and vasogenic cerebral edema are not the same. The TBI research (http://www.ncbi.nlm.nih.gov/pubmed/15561417) shows that vasogenic cerebral edema is due to blood-brain barrier disruption resulting in extracellular water accumulation. In the other hand, osmotic cerebral edema is caused by osmotic imbalances between blood and brain tissue. These kids who got MRI early, all had vasogenic edema. How is that different by MRI? - I have no idea, but I am sure the radiologist have.

In regards to the factors that predict the development of cerebral edema, there is a great article from the NEJM from 2001, comparing 265 children with severe DKA. This study puts it all together very nicely. (http://www.nejm.org/doi/full/10.1056/NEJM200101253440404) Some children developed cerebral edema and some did not. Analysis of other biochemical markers showed something very interesting. "Although osmotic factors and other mechanisms may play a part in the development of cerebral edema, our data lend support to the hypothesis that cerebral edema in children with diabetic ketoacidosis is related to brain ischemia. Both hypocapnia, which causes cerebral vasoconstriction, and extreme dehydration would be expected to decrease perfusion of the brain. In addition, bicarbonate therapy causes central nervous system hypoxia in laboratory animals with diabetic ketoacidosis. Hyperglycemia superimposed on an ischemic insult increases the extent of neurologic damage, blood–brain barrier dysfunction, and edema formation. This interaction might help to explain the occurrence of neurologic damage in association with minor degrees of cerebral hypoperfusion. Blood–brain barrier dysfunction and vasogenic edema may occur several hours after an ischemic insult as a result of the release of vasoactive substances and mediators of inflammation. The occurrence of cerebral edema several hours after the initiation of therapy thus correlates well with the hypothesis that the basis of this complication is ischemia. Finally, the more frequent occurrence of cerebral edema in children than in adults may be explained in part by the fact that children's brains have higher oxygen requirements than adults' brains and are thus more susceptible to ischemia".

A good while ago, Canadians also found that low CO2 and high BUN had the strongest association with cerebral edema (http://linkinghub.elsevier.com/retrieve/pii/S0022347604012168?via=sd). Again.. these are sick DKA kids. This was a case control study, after adjusting for variables, found not association between the occurrence of cerebral edema in DKA and treatment factors. The authors conclude that the presence of cerebral edema before treatment of DKA and the association with severity of illness suggest that prevention of DKA is the key to avoiding this devastating complication. 

Summarizing... Here are the strongest associations with their respective OR's, CI and P values

• Initial BUN (per increase of 9 mg/dl): 1.8 (95% CI: 1.2-2.7); p value 0.008
• Initial partial pressure of arterial carbon dioxide (per decrease of 7.8 mm Hg): 2.7 (95% CI: 1.4-5.1); p value 0.002
• Treatment with bicarbonate: 4.2 (95% CI 1.5-12.1); p value 0.008 (Don't do it!)
• Rate of increase of serum sodium concentration (per increase of 5.8 mmol/l/hr):0.6 (95% CI: 0.4 - 0.9); p value 0.01

Conclusions
- Cerebral edema in the setting of DKA is more related to the severity of the primary disease process rather than the rate of fluid treatment. The pathophysiology is complicated, but hypoperfusion leading to cerebral ischemia along with damage to the blood-brain barrier all result in vasogenic edema. 
- Pedi DKA in shock -> Treat shock !  Their brain needs perfusion. However, be cautious and don't flood this kids with fluids, just restore tissue perfusion. Some guidelines (mostly from the U.S.) suggest an initial bolus of 20 ml/kg is a reasonable start point in shocky kids, with the option to repeat if there is no improvement in hemodynamics. However, a maximum of 30-40 ml/kg in the first 4 hrs of treatment is recommended with early consideration of sepsis if no response. This is when being a good clinician is so important. 
- Don't give bicarbonate, don't give bicarbonate and don't give bicarbonate. Is that clear enough?
- Having a healthy skepticism and questioning the old dogmas may get us closer to the truth about what is the right thing to do. 

Emergency Radiology and Pregnancy

My apologies for the delay in posting something new, but this topic required a lot more reading than initially expected. There is a lot of information from very diverse sources, not just medical imaging literature but also from genetics, physics and the energy industry as well. Very interesting stuff!

The pregnant patient is, and will always be, a complex patient. The presentation of life threatening diseases is often atypical, symptoms that could be part of a normal pregnancy may also indicate serious pathology, the physiology of pregnancy is different from a non-pregnant female and to make things even more complicated, there is a little human being inside! No wonder why the sick pregnant patient in the ED scares the socks out of us. We all use our clinical acumen but that is often not enough to decide sick-or-no-sick, and we must rely on imaging technology to aid in the diagnosis. There is handful of non-pregnancy related diagnoses that create all the problems and split hairs, these include appendicitis, renal colic, ovarian torsion, hemorrhagic ovarian cysts, pulmonary embolism and trauma. It is well known that whenever possible, ultrasound is the modality of choice when it comes to imaging the pregnant patient; unfortunately, it is not always conclusive and CT scan becomes the better choice. That's when the "radiation talk" should take place. I have heard all kind of crazy things about this, mostly due to lack of information in patients and even providers. Therefore, I would like to cover some basic physics about radiation and its effects, just to put things in perspective.  It will not make you a physicist, but you can sound like one when talking to the radiologist, and that's very useful. Shall we...?

Fact number 1: Radiation comes from everywhere and everything; from natural sources like air, water, food, plants, ground and cosmic; as well as artificial sources like the electronics we use everyday, buildings, occupational exposures, nuclear and medical imaging, etc.

Fact number 2: There are several units to measure radiation. Depending on which system is used in the country you live in, there are Rads/Grays and Rems/Sieverts. 1 Rad = 1/100 Gray and 1 Rem = 1/100 Sievert. For the purpose of this review, it is worth remembering Grays and Sieverts. A Gray (Gy) is a measure of absorbed ionizing radiation which is equal to 1 Joule of energy release in 1 kg of mater. A Sievert (Sv) is the unit of the effective dose of radiation that has a biological effect on tissue, 1 Sv is equal to effect of 1 Gy over the exposed tissues multiplied by the specific weighting factor. If this sounds too complex (and believe me, IT IS) just remember that a Gray is the amount of radiation received by the tissues and a Sievert is the unit for the effects of 1 Gray of radiation in humans. Although technically are not the same, for practical purpose they can be thought as similar units.

Fact number 3: The average dose of background radiation a human accumulates just for being temporary habitant of this planet is somewhere between 2 and 7 mSv (mili-Sieverts) depending on location and altitude. There are high radiation locations in areas of nuclear disasters, natural occurring "leaks" from the earth, mines and high altitude.

Fact number 4: From the different types or radiation, ionizing radiation is the one used in medicine. Its ability to pass through tissues of different densities makes it ideal for imaging technology and treatment of cancers. The problem is that as it goes through to the tissues, it deposits enough energy to brake molecular bonds and displace electrons from atoms creating free ions (therefore the name ionizing); this results in damaged bonds in the DNA of living cells.

Fact number 5: In industrialized countries, the most common sources of artificial source of ionizing radiation is medical imaging with an average of 3 mSv per year per person (world's average is 0.6 mSv) and air travel with 2.1 mSv per year. Of note, heavy smoking (1 pack per day) results in radiation dose of ~160 mSv per year directly into the lungs! (If you smoke, you need to stop)

Well... I think that's enough physics for one day! Now let's apply these facts to the medical imaging in humans, including unborn babies.

We just learned about the effects of ionizing radiation in living cells and that we all are exposed to radiation from multiple sources at any given time. Then why aren't we all dropping death with cancers of all types? - Well, that is because there is an extremely sophisticated and highly specialized enzymatic complex system that detects, repairs or destroys damaged cells. This awesome system corrects billions of DNA mishaps a day and maintains cellular functions. Ionizing radiation in high doses, and specially after repetitive exposure, can eventually overwhelm this mechanism and lead to various types of malignancies. The fast mitotic fetal cells are particularly vulnerable to these effects from radiation, therefore is a good idea to limit fetal exposure whenever possible.

I found this table with the average dose of radiation from different studies, their equivalent to background radiation in years and its risk for malignancy

Too much is said about the fetal radiation risks for various types of imaging technologies derived from phantom models, animal and human observational studies; all thrown in the same bowl with extrapolated nuclear bomb survivors and nuclear disasters data. The result is an estimation of risk, but let's be clear about something... No one has solid, indisputable human information with a dosimeter next to a developing fetus to accurately measure radiation doses in-utero and its effects based on randomized studies (and we never will). All of the current available recommendations are predicated on estimated risks based on less-than-perfect data. Having said that, this is all we have and it seems to be enough to draw some conclusions.

The background dose of radiation for 9 months of pregnancy is estimated at 0.5 to 1 mGy, and the threshold for increased risk of fetal anomalies or pregnancy loss is 50 mGy (5 Rads) or 50 mSv (5 Rems). Standard radiological tests produce radiation doses far below the 50 mSv threshold. The aggregate risk for spontaneous miscarriage, major malformation, mental retardation and childhood malignancy in the general population is estimated to be about 28.6%. A dose of 50 mSv of ionizing radiation will increase this risk to approximately 28.8%. Specifically about childhood cancer, defined as any cancer with onset before age 15, the most common being leukemia, the average risk of leukemia in general pediatric population is about 0.036% (3.6 per 10,000), exposure to 50 mSv will increase this risk to approximately 0.06% (6 in 10,000).  From these statistical models we can conclude that although the risk of negative effects of the cut off of 50 mSv is not zero, it is indeed, very very small. The National Council on Radiation Protection and Measurements, and the American College of Obstetricians and Gynecologists have both agreed that the potential health risks to a fetus are not significantly increased from most standard medical tests. The American College of Radiology (ACR) has also come on record saying that "No single diagnostic procedure results in a radiation dose that threatens the well-being of the developing embryo and fetus" (Hall EJ. Scientific view of low level radiation risks. Radiographics. 1991;11:509)

This table shows the fetal dose of common radiologic tests. All these give less than 50 mSv, so it is safe to say that when we need to image a pregnant patient using ionizing radiation, we can proceed knowing that any of the studies we use in the ED represent low risk.

This next table summarizes the average dose of multiple radiologic studies and the number of studies needed to reach the aggregate dose of 5 Rads (50 mGy/mSv)

And this last table is from the ACR 2013 revision of its practice guideline on imaging the pregnant or potentially pregnant women using ionizing radiation, reaffirms what has been said by other organizations regarding the cut off of 50 mGv as safe level. 

What about contrast? - Well, I did find some useful information about that from the ACR 2013 manual on contras media. Basically it says that the water soluble iodinated low-osmolarity contrast media (the one use currently) does cross the placenta but there is no current evidence of mutagenic or teratogenic effects from it. As far as the effect on neonatal thyroid function, the document says that the amount of contrast in the fetal circulation is small and transient, and there are no reported cases of neonatal hypothyroidism in babies whose mothers received this type of contrast and the FDA has given it category B status. The ACR's recommendation about the Gadolinium-based contrast media (GBCM) used in nuclear medicine studies is not as strong. It says that although there have been no know adverse effects to human fetuses by the use of this agent, there is only one study of 26 pregnant patients who were exposed to Gadolinium during the first trimester, none had teratogenic nor mutagenic effects of the progeny. Therefore, the use of GBCM should only be used when the benefits justify the potential risk to the fetus. 

OK... are you with me so far? - Good! Let's now get practical and put all this theory where the rubber meets the road, at the bedside of the pregnant patient with a potentially serious diagnosis.

Let's start with the rule-out appendicitis case and the ultrasound comes back with something like this "Appendix not seen, acute appendicitis cannot be excluded, consider pelvic pathology... clinical correlation required". Now what? - Sure, you can try to put your pregnant patient in the MRI for 30 minutes and hope for a clear diagnosis; however, MRI is not as sensitive nor specific compared with CT, thus resulting in equivocal results, and most radiologist are far better diagnosing acute appendicitis on CT than MRI. CT with oral and IV contrast is the better choice, and the 25 mSv dose of radiation are still considered relatively safe by ACR and ACOG. Now the conversation with the patient should include the following... If this is acute appendicitis and we don't find out on time and it ruptures, there is between 6-37% chance for fetal loss, maternal morbidity and mortality range around the 5% and 1% respectively. and the risk of the radiation dose of the CT abd/pelvis for anomalies, fetal loss or childhood cancer is less than 1%. It seems clear that scanning is the best option. 

Urolithiasis with renal colic is the most common non-obstetric diagnosis requiring hospitalization, affects about 1 in 1500 pregnant patients and it is often confused with appendicitis, diverticulitis, ovarian pathology and placental abruption. Ultrasound is first line test to diagnose urolithiasis during pregnancy. When the stone is visualized that's great, but when all you see is hydronephrosis it is hard to tell if that is the hydronephrosis of pregnancy or due to a distal obstruction. The good news is that about 60-80% of stones will pass with conservative management, the bad news is that 20-40% will not and urologist use size of stone and location to determine treatment options. MRI is good to see hydronephrosis but not so much stones, so it doesn't really have significant advantage over ultrasound. Intravenous pyelogram has fallen out of favor because of the 50% higher radiation dose compared with CT scan and it only provides imaging of the urinary tract. CT scan again comes as top option for complicated cases of urolithiasis because of its high sensitivity and specificity, and ability to screen for other intra abdominal/pelvic pathology. 

The pelvic pathology including ovarian torsion, adnexal mass, hemorrhagic cyst and degenerating fibroid are best seen with ultrasound, so no surprises here. However, in late pregnancy the gravid uterus may obscure adequate view with the ultrasound. The MRI without contrast could be used in the stable patient. CT with IV contrast becomes the imaging modality of choice in the unstable patient with hemoperitoneum.

Pulmonary embolism is, on it self, a monster topic which becomes even more monstrous in the pregnant patient. With a mortality approaching 15% and significant morbidity of anticoagulation, we must get this right in a timely fashion. There are several protocols including trimester adjusted D-dimer + leg ultrasound in leu of pulmonary imaging. In patient with symptoms suggesting PE and (+) US for DVT, most will go ahead and treat; but when this approach is not diagnostic, pulmonary imaging becomes mandatory and the options are CT pulmonary angiography (CTPA) vs V/Q scan. The ACR rates both studies as adequate in the pregnant patient with radiation doses below the 50 mSv limit. The American Thoracic Society in its 2011 practice guidelines recommends plain chest x-ray as the initial radiation-associated step. If the CXR is normal, proceed with the perfusion phase of V/Q scan followed by the ventilation portion if abnormal. If CXR is abnormal, then CTPA is recommended. The algorithm looks like this...

The problem with this approach is that if the V/Q is inadequate or non-diagnostic (and many of them are), then you still have to proceed with CTPA. The advantages of CTPA is that it can also provide alternative diagnosis (i.e. pneumonia) and is more widely available compared with a V/Q scan. The fetal radiation doses of both studies is fairly comparable with an average of 0.2 mSv for CTPA, 0.12 for the perfusion-only V/Q and 0.2 for the ventilation portion of the V/Q scan. The remaining issue to discuss about CTPA is the radiation exposure to the hyperplastic breast tissue of the pregnant patient, which is said to increase life time risk for breast cancer in about 1%. Breast shields and timed beam techniques can significantly lessen this exposure.

The final diagnostic dilemma to review is trauma. This is the easy one because everyone agrees that when it comes to radiology studies, you just do it. Trauma is the number 1 reason for non-obstetric mortality during pregnancy, and unless you are ready to do a perimortal c-section, treating mom is the best way to treat baby. The pregnant trauma victim should be imaged just as the non-pregnant with x-rays, CT or angiography when required. Sure, you take a quick look with the US to check on the fetus and the placenta, but don't get hang on that screen while the mom is bleeding out, and remember that every pregnant trauma victim beyond 24 wks gestation once stable, should be placed in continuos cardio-tocographic, which is the most sensitive way to diagnose placental abruption. It is possible that mom may require multiple studies that may add up radiation doses of > 50 mSv and rarely > 150 mSv, and someone needs to keep track of what studies have been done and what studies are still needed. In such cases with high fetal exposure doses, therapeutic abortion should be discussed with the patient.

Wow... you are still reading! I am sorry this topic is too long, but I think it contains useful information that can be applied anytime when you pick up a chart saying "20 wk pregnant with abdominal pain". Now, it is time for final points.

Pregnancy test. The ACR says that for negligible risk examinations like CXR or extremity plain films, pregnancy test is unnecessary. Documentation; when higher risk examination is needed make sure to document clearly and completely. Tell the chart you have discussed alternative diagnostic options with the patient, mention the risks of doing and not doing the test, and what you feel is in the best interest of mother and baby. Finally, remember that avoiding radiologic tests in a potentially life threatening condition in order to avoid fetal exposure to ionizing radiation is not going to score you any points when you end up with a dead mother, so do what it right for your patients.

References:
http://pubs.rsna.org/doi/full/10.1148/radiol.2481071451
http://pubs.rsna.org/doi/full/10.1148/radiol.2312030767
http://www.acr.org/~/media/9e2ed55531fc4b4fa53ef3b6d3b25df8.pdf
http://www.sievert-system.org/WebMasters/en/mesure.html
http://hyperphysics.phy-astr.gsu.edu/hbase/nuclear/radrisk.html#c1
http://www.radiologyinfo.org/en/safety/index.cfm?pg=sfty_xray#part6
http://www.radiologyinfo.org/en/safety/index.cfm?pg=sfty_contrast#part7
http://hps.org/documents/pregnancy_fact_sheet.pdf
http://hps.org/physicians/documents/Radiation_and_Pregnancy.pdf
http://www.acr.org/~/media/ACR/Documents/PDF/QualitySafety/Resources/Contrast%20Manual/2013_Contrast_Media.pdf/#2013_Contrast_Media_Manual.indd:.27754:10189
http://www.nrc.gov/about-nrc/radiation/health-effects/radiation-basics.html
http://emedicine.medscape.com/article/455830-overview
http://www.medscape.com/viewarticle/812189_2?nlid=38923_541
http://www.acog.org/Resources_And_Publications/Committee_Opinions/Committee_on_Obstetric_Practice/Guidelines_for_Diagnostic_Imaging_During_Pregnancy
http://en.wikipedia.org/wiki/Background_radiationhttp://www.aafp.org/afp/1999/0401/p1813.html
http://www.thoracic.org/statements/resources/pvd/evaluation-of-suspected-pulmonary-embolism-in-pregnancy.pdf
http://www.ncbi.nlm.nih.gov/pubmed/17620458
http://www.aafp.org/afp/1999/0401/p1813.html

Low Risk Chest Pain and the Accelerated Protocols

This is a monster topic and the pain of our existence in the ED. We see CP patients left and right and some can represent a challenge to even the seasoned clinician. The crushing retro-sternal CP in a 66 y/o sweaty-obese-hypertensive-diabetic-smoker is the easy one, even the intern knows that patient is not going home b/c the probability of ACS is so high, that even normal tests are likely to be falsely negative. The 21 y/o with chest pain and large bruise after being hit with a baseball is going home b/c he doesn't have ACS. Those too are easy... the hard cases are the ones in between, when the story is not totally convincing, the exams is not helpful, the EKG is not conclusive and the patient may have some risk factors or positive family history. As the gate keepers and stewards of the health care system, we must move these patients through the department efficiently and effectively, without missing a single case. Right?

Multiple risk assessment tools have been developed to help us identify low risk CP patients who are safe for discharge and can be followed up as outpatient. Some of these score systems have been validated and the basic idea is to quickly identify patients who are a very low risk for adverse events in a short period of time (usually 2 hrs) instead of the usual 8-12 hr protocols. A key concept to understand is that we do not rule out ACS in the ED, a patient with 90% stenosis can graduated from the department with 3 sets of negative cardiac markers and non-diagnostic EKG's. So we re-stratify patients to a level of risk that is acceptable (and defendable) that will guide further management.

It is not my intention to write a meta-analysis given the heterogeneity of these individual studies in their methodology, protocols and type of biomarker used. But here is a quick summary of some of the main studies evaluating accelerated re-stratification protocols (Rajeev... I hope this answers your question)

The Lancet. March 2011 (http://www.thelancet.com/journals/lancet/article/PIIS0140-6736%2811%2960310-3/fulltext) was a prospective validation study of 3582 patient in Australasia using a TIMI score of 0 (http://www.mdcalc.com/timi-risk-score-for-uanstemi/), non-ischemic EKG, negative biomarkers at 0 and 2 hr time. If all of those criteria were fulfilled, these patients were considered safe for discharge. Out of the initial pool of patients, 352 met criteria for discharge and only 3 had a cardiovascular negative outcome within 30 days defined as AMI, need for revascularization or death. This gives a 99.3% sensitivity, a 99.1% NPV and specificity of only 11%.

JAMA, December 2011 (http://www.ncbi.nlm.nih.gov/pubmed/22203537?access_num=22203537&link_type=MED&dopt=Abstract) Another prospective study, done in Germany with 1818 subjects. This study used only highly sensitive troponin I (no scores and no EKG's) at 0, 3 and 6 hrs. At 3 hr, repeat negative troponins had a sensitivity of 98.2% and PPV of 95.8% for ACS, with the 6 hr marker not adding significant extra benefit, making the point that 3 hr markers are as good as the 6 hr.

Clinical Pathways in Cardiology. March 2013. This is an European journal (http://journals.lww.com/critpathcardio/Abstract/2013/03000/HEART_Score_to_Further_Risk_Stratify_Patients_With.1.aspx) and the authors here put forward a new HEART score (http://www.heartscore.nl/score/) applied to patients with an already low TIMI score (0-1) to further identify very low risk patients who are safe for discharge from the ED. This included 8815 enrolled, from which 485 had both a TIMI of 0 and HEART score of 0. From this subset none had a negative cardiovascular outcome at 30 days for a sensitivity of 100%.

JAMA, Internal Medicine. October 2013. (http://archinte.jamanetwork.com/article.aspx?articleID=1748796) This is a randomized, parallel comparison study with blinded outcomes comparing a 2-hr accelerated diagnostic protocol (ADP) including TIMI of 0, non-ischemic EKG and negative biomarkers at 0 and 2 hr, vs standard care (initial negative biomarkers, prolonged observation usually with admission to hospital and second biomarker at 6-12 hr after onset of symptoms). The ADP had 270 pt of which 52 were discharge following the protocol and the standard care group had 272 pt and 30 discharges. At 30 days none of the discharged patients in both groups reported a negative cardiovascular event, making the conclusion that accelerated protocols are as effective and more efficient than standard prolonged observation protocol.

So, now what do we do with this information? Are these accelerated protocols ready for clinical use in the ED? Well... sort of. The good news is that these studies were done in the ED with general population cohorts and not cardiology centers, they have excellent results which can be applied to general practice. The bad news is that the AHA and ACC have not yet come on record with an official opinion about these protocols. However, as more and more patients show up through our doors with CP's the need for more efficient and cost-effective disposition has become imperative. But don't be the odd ball of your group and start doing this alone, take this information and talk to your medical directors and come up with an agreement on how to use these protocols in your departments.

A heart ATTACK !!!

The Science of Getting an IV & Collecting Blood

The case was a young trauma victim, he wasn't that sick, but sick enough to get admitted for observation. As part of the trauma protocol we collect blood for labs, type and screen and the usual stuff. Primary survey completed, we had a nice 16 g cannula in his left AC fossa but blood had not been collected yet. The lab tech arrived and was preparing to stick the patient in the R arm when I asked him. Why don't you take the blood from the cannula? - I'm not allowed to do that. He replied. I like to avoid unnecessary pain to my patients whenever possible so I reapplied a tourniquet to the L arm, pulled 3 cc of blood to discard, then collected the blood directly into the tubes. As I was doing that, the ED chief nurse saw me from across the room and almost had a stroke. She went into a rant about how this was a violation of the protocols and yara yara yara... To my amusement, she took the tubes I had just collected from that freshly-started IV and trash them into the sharps container. Then she order the lab tech to stick the patient again in the other arm for a "correct" blood sample. This was a senior nurse and I just couldn't believe she would do that. What is the difference from that blood to the blood in the other arm?!

So I did a lit search about this topic and found some interesting data. 

- The cannulas are made with soft plastic that remains open by the positive pressure exerted by the infusing fluid, the walls tend to collapse if too much negative pressure is exerted, this causes turbulence and higher rate of hemolysis. This hemolysis is more likely to occur using the larger vacutainer tubes of 10 cc because of higher negative pressure early in the draw or when using a syringe and pulling too hard on the plunger. So if you use a syringe, pull the plunger just a little bit until you see blood flowing into the barrel and then continue gently, and for the vacutainer system, use the 5 ml tubes to minimize hemolysis. 

- And what is the deal of tapping the site to "pop" the vein into view? That doesn't work, the skin gets red but the vein doesn't change in size and it also can cause hemolysis inside the vein, so let's not do that either. 

- Already inserted cannulas can be used to collect a blood sample even when IV fluids of medications have been infused through it. The only caveat is that for glucose containing fluids it is recommended to wait 3 minutes after stopping the infusion. When using the inserted IV, "washing" the cannula with 5 times its capacity and discarding that blood will eliminate any possibility of contamination. The volume capacity of a 22g cannula is about 0.05 ml and for a 14g  0.15 ml (yes, go ahead and measure it), therefore drawing 1 ml is technically enough, but get 2 ml and no one will argue.

- If you have a prolonged vein hunt for more than 2 minutes, go ahead and get the cannula in, remove the tourniquet, elevated the arm and then bring it down again to allow "fresh" blood into the extremity, reapply the tourniquet and then collect the samples. This is particularly important if you use venous lactate  in sepsis screening to avoid false positive results. 

- Follow this order when filling up the tubes: Blood cultures -> Lactate -> Coagulation -> Serum -> Heparin -> EDTA. This will minimize the chance of contamination, false positives, hemolysis and sample clotting. 

- Don't overfill, shake too hard or drop the tubes; and if you open the tubes transfer the blood from the syringe, do it without the needle.

And this is the summary of all of these articles. If you have a couple of hours and there is nothing on TV, go ahead and read...

http://bmhlibrary.info/11174369.pdf
http://www.ncbi.nlm.nih.gov/pubmed/15565033
https://www.futurelabmedicine.org/pdfs/LMBP_ReducingHemolysisSummary.pdf
http://www.who.int/injection_safety/phleb_final_screen_ready.pdf

And before I go, a final advice for my fellow physicians. If you think that starting IV's and collecting blood is the nurses' job, then you are missing the critical concept of team work. As leader of the team you should know how to do this and every other procedure in the ED; from starting lines, mixing drips, work up the pumps, etc. This will not only benefit the patient, it will also earn some golden points with the rest of the team. So pull up your sleeves and get busy.  

Wellen's Syndrome... an EKG Finding Not to Miss!

Let's say you're working away in the emergency department and you see a 35 year old man complaining of chest pain for about 2 hr. He does not have a significant PMHx, he is mildly obese and smokes a pack a day, the history is not totally convincing for ACS and when he arrives to the ED the pain has resolved. Sounds familiar...? - We probably see a 10 of those per shift! The protocol says, everyone who says the words "chest" and "pain" in the same sentence gets an ECG at triage. The nurse brings it to you and this is what you see.. (sorry for the shadows, but I couldn't find the right angle)

Hmmmm... A 10 second view shows sinus rhythm, a little slow, the voltages are a bit low in the limb leads. There are some ST-T changes that don't look quite right with T wave inversions in the anterior and lateral leads, with similar milder changes in the inferior leads. The J point seems to be isoelectric and there is some "eyeballed" QT prolongation (R to R seems longer than the twice the length of the QT). The protocol also says that in the absence of contraindications everyone gets hooked to the monitor, oxygen by nasal cannula, aspirin, labs, CXR, maybe NTG. (Yes, I know, and before you start going at it, I am trying to change that oxygen mandate in my department; but it is hard to change protocols!) You do a physical exam which, as in most cases, does not reveal anything significant. Few minutes later the labs come back with negative cardiac markers, his CXR looks fine and he remains pain free. The repeat EKG looks like this... The changes practically resolved!

Now what?! What's the deal here? This dude has dynamic EKG changes, but he is pain free, markers are normal, and now everything seems to be OK. Right? - WRONG! This guy is in deep doo-doo if you don't do the right thing. And the reason you should be scared is b/c he has Wellen's syndrome. 

Wellen's syndrome is an electrocardiographic phenomenon seen in critical obstruction of the LAD. Described in the 1980's, Wellen's has 2 types, type 1 has biphasic T wave in leads V1-V4, with the type 2 having deep T waves inversions. Other characteristics are no precordial Q wave and no loss of R wave progression. (Here is a type 2, note the deep TWI in anterior leads)

So, what happened to our patient? - I called cardiology and said something like this: "Sir, I have a 35 y/o with Wellen's, currently chest pain free, with negative cardiac markers. Do you want to see him here in the ED or in the CCU?" The cardiologist seemed to be in a good mood that day and didn't give me any grief. Patient went up to CCU and booked for cath the next day.

Now, putting it all together. What you need to know about Wellen's

- Recognize the pattern of biphasic T wave or deep inverted T waves in V1-V4

- It means high degree of obstruction in the LAD

- DO NOT send this patients for stress test. It will precipitate an AMI

- These patients need admission and early revascularization

- Do not miss it. This is a "pre-infarct" phenomenon. If the patient goes home, he/she will progress into an anterior MI, and that is considered not good.

For more reading...

- de Zwann C, Bar FW, Wellens. Characteristic electrocardiographic pattern indicating a critical stenosis high in the left anterior descending coronary artery in patients admitted because of impending myocardial infarction. Am Heart J. 1982; 103:730-736

- Tandy TK, Bottomy DP, Lewis JG. Wellen's syndrome. Ann Emerg Med 1999; 33:347-351 (http://www.ncbi.nlm.nih.gov/pubmed/10036351)

- Rhinehardt J, Brady WJ, Perron AD, Mattu A. Electrocardiographic manifestations of Wellens' syndrome. Am J Emerg Med. 2002; 20:638-643. (http://www.ncbi.nlm.nih.gov/pubmed/12442245)
- http://www.consultantlive.com/cardiovascular-diseases/content/article/10162/2158916
- http://www.consultantlive.com/cardiovascular-diseases/devil-t-wave-wellens-syndrome

In honor of Dr Wellens

http://www.ficemaastricht.org/index.php?page=speakers&sid=HW