Okay, so we will, so the faculty will read the questions and go over the, just highlight some of the answers, and then there's going to be a QR code. You scan it and, you know, we'll answer anonymously, and then we will discuss the answers. So, do I? I'm not sure how. Okay, it's a new, new system. Oh, should I read the question? Yeah. Yeah, go ahead. You will be there. All right, so I'm going to go ahead and start reading the first question. I'll read the answer choices too, and then the QR code will be on the next slide. So, an 18 months old girl is admitted for acute respiratory distress and evidence of systemic inflammatory response syndrome with fever, tachypnea, tachycardia, and leukocytosis. She becomes acutely hypotensive, requires inotropic support, and progresses to respiratory failure requiring intubation and sedation. She's well oxygenated on 45% FiO2 and modest, modest ventilator settings. The attending physician prepares to insert a femoral venous central line for the purpose of inotrope, antibiotic, and fluid administration. Using ultrasound guidance and in accordance with the catheter-related bloodstream infection guidelines, the physician inserts an appropriate size double lumen central venous line that draws blood and flushes well. The physician completes the procedure and would like to begin using the catheter to continue resuscitating the patient. Which of the following steps should be taken to confirm venous access before use? A, no further steps are needed because placement was confirmed via ultrasound. B, await plain film of the abdomen to confirm appropriate location and depth. C, no further steps are needed because the color of the blood during insertion indicated that it was venous. D, transduce the newly inserted CVL to look for an arterial waveform. E, measure blood gases to confirm an appropriate venous partial pressure of oxygen. It's like a race, I can see it live. Saginaw can play the winner. That's right. Tim, you're touching colleagues. Okay, so I go to the, this one? Okay. It's like at the bottom of the dark. All right, looks like that is the end. The answer is D. Transduce the newly inserted CVL to lick from an arterial waveform. And then I go to the rationale. Okay. It's a new system, so. Okay. Well, so let's see the answers. I'll just go. As we work through this, on some of these questions, obviously you can do more than one thing. But what they're getting at here is to be sure that the positioning is venous, not arterial. So transducing it is the answer. I'm aware of another institution of a lawsuit on just this case. So you really need to prove that this is venous, not arterial. You would still get an X-ray? The key with an X-ray, if you get an X-ray, what do you need to do? Two views, right? One is for how high or where the tip is, but also a cross table to make sure it's not sitting in the lumbar plexus, which happens as well. So you would do both. But, again, what's the question really getting at is, is this really a venous line? Yeah, and I think also because it's to continue resuscitation. So that's something you can do quickly, right, and transduce right at the bedside, not having to wait for the X-ray. One of the test-taking points, there was a huge stem, and what I found on this board, on the Pete's Critical Care boards, as compared to a lot of other boards, is I was time crunched, and I'm not normally time crunched when I take them. So read the question. Like go to the very end of the question and say, what are they asking? You didn't have to read the majority of that stem. You could have gotten the answer if you'd just gotten to that point. So that's how you can save some time. Read the last part, but also read the answer choices before you go back. Know what they're asking and what they're looking for. It will save you so much time because, like Robbie, I'm always time crunched on this as well. Do you think there's little transducer things that you can screw on to your catheter at your hospital? Okay. I don't know what's going to happen. Wait, what? Okay. Yeah, poor man's CVP. Okay. Question number two. Next. I can do this. 18-month-old girl is admitted for tachycardia, decreased perfusion, mild respiratory distress. She's febrile. There are multiple recent sick contacts in the house. Vital signs are temperature 38.6, heart rate 210, blood pressure 70 over 35, respiratory rate 40 with mild retractions and oxygen saturation as measured by pulse oximetry is 92% on two liters via nasal cannula. Chest radiograph shows diffuse congestion. A right-sided infiltrate with associated pleural effusion and concomitant cardiomegaly. Subsequent echo demonstrates large pericardial effusion with right atrial collapse during diastole, but with intact cardiac function. Which of the following should be the most immediate next step in management? Start an etzolol infusion, intubate, and ventilate. Give a furosemide bolus, administer a crystalloid bolus, or start Melronone. You shouldn't have to put the QR code again. Yeah, yes. Oh, they don't have to. Okay. So, okay. Just click start. Yeah. This is called play the winner as long as you trust your colleagues. If not, you're going to have a really bad path. Okay. Okay, the correct answer. The correct answer that most people agree with is choice D, administer crystalloid bolus. And the point of this question here in the rationale, next slide, is that this is a tamponade physiology discovered with POCUS when monitored with ultrasound. So, this patient probably has viral pneumonitis, pericarditis, possibly myocarditis from a virus. The fluid accumulation around the heart, another example of a starling resistor. If the pressure is high enough, it can impair filling from the right side particularly. And in this case, the ultrasound showed a compression of the right ventricle confirming tamponade physiology. So, you certainly wouldn't want to take fluid away with furosemide. You would want to ensure that there is adequate preload with this other process going on so that atrial and ventricular filling would be ensured. And I forget what the other choices were, but let's see, intubate and mechanically ventilate. If you did that right off the bat, this patient would probably almost certainly arrest. Esmolol infusion would drive the heart rate, potentially cause some peripheral vasodilation, make it worse, decrease preload, and milrinone, the same thing. So, the correct answer, yes, crystalloid bolus, ensure preload. Go ahead. Yes. If this was not at this patient, it was somewhat marginal, but if they had some peripheral fusion and they were concerned that they were evolving cardiac tamponade and they also had pericarditis in Jesus, that's a question I've had. What would be the, like, I feel, there's a lot of different things that could be done with pericarditis and Jesus was also there, I would give a bolus and have my team ready, but it was, it was, the question I did, I was surprised that it was pericarditis. the pericardiocentesis was in this stem, would you have fit for the pericardiocentesis? It's not related, no. I mean, I think your points are all well taken. That wasn't the question, but if pericardiocentesis was, well, I mean, part of this is examsmanship, for sure. If pericardiocentesis was paired up with initial volume resuscitation, I probably would give the volume while waiting for the, who's ever gonna stick the needle in first. Yes. Yeah, I think one of the- Right, it's usually what is the best next step, right? Because some things you're doing in parallel, but nobody's ready to, you know, with the needle in the ground and ready to go. So certainly the patient, you have to resuscitate him, and at the same time, you, so read what the question says, because, you know, what's the next step? And based on that, you will decide. And I was going to comment that, for sure, there are going to be questions about cardiac tamponade, and so know how to manage, how to diagnose it. Any other comments? Yeah, the only other comment I'd make is, remember, a lot of these questions here, they're teaching questions. A lot of the practice questions, they've been vetted by, like, groups of us, but they've not been formally vetted, reviewed, and tested. So a lot of these, it's fun to argue with the questions, but the key is obviously to learn from them. Now, if the real test, you can argue with them, but you're going to lose. Here, you can argue, because we're all trying to teach each other. But if you question the question, just please bring the concerns, because, again, these are learning techniques for the exam, right, so we can discuss this openly. Okay. Okay, the rationale, Dr. Zimmerman already explained that. Why not? Okay. Okay, next question. This goes back to one of the modes you might see, APRV. This is a 12-year-old boy with ARDS, transitioned from conventional, traditional mechanical ventilation to APRV. Plateau pressure on conventional was 27 centimeters of water. APRV settings are a p-high of 25, a p-low of zero, a t-high of 2.5 seconds, a t-low of 0.2 seconds. In his most recent ABGs, a pH of 7.35, CO2 of 52, PO2 of 46, with a saturation of 84% on 100% oxygen. Which of the following interventions is most appropriate to improve oxygenation? Neuromuscular blockade, decrease p-low, decrease t-high, increase p-high, or increase t-low? So you have to get at, really, what's the concept here of oxygenation, and what does the mode do? Can I go to the next slide? Yes, read the question? Okay. Okay, so 83% of you got it correct. So aside from the mode itself, you want to improve oxygenation. So most of the times improving oxygenation is going to be what? Increasing mean lung volume, increasing mean airway pressure. So the question comes down to which one of these will increase mean airway pressure. And it's going to be to either increase P-high, which is a choice, or potentially to do what to T-high? Increase T-high, spend more time at the higher pressure, but that's not a choice, right? Because if that was a choice, then either could be correct. You would not want to do neuromuscular blockade because APRV, you need the spontaneous breathing around P-high to adequately ventilate in most situations. So you really need to get here is increasing P-high. The other part of the rationale, if you read through it, is generally with APRV, you set your P-high at the prior plateau pressure. In this situation, actually, they set it a couple of centimeters of water below. So increase P-high is the answer. And if there is going to be one APRV question, and the exam is going to be this one. Yeah, probably, yeah. Okay. I just did an editorial that said that APRV should only be used in the research setting. Just this last week. Yeah, many centers don't use it, right? So, okay. How many of you at your center use APRV? Oh, quite a few. Very good. Routinely. Interesting. Yeah, no, we don't do it routinely. Okay. Okay. Okay, next one. A 12-year-old girl with metastatic osteosarcoma was admitted to the pediatric ICU for acute respiratory failure. She clinically improved after being placed on low mechanical ventilator support. Pressure support ventilation 22 over 8 with FIO2 40% and is spontaneously breathing. Her most recent ABG revealed pH 7.34, PaCO2 48, PaAO2 85. Heart rate acutely increases from 95 to 130, and respiratory rate increases from 12 to 30. Arterial oxygen saturation has fallen from 95 to 97 to 91, and the end tidal CO2 reading has dropped from 35 to 15. Repeat ABG shows pH 7.42, PaCO2 39, PaO2 65, and bicarbonate 27. Her dead space fraction is now closest to... Can I move to the next one? No, I'd give them a minute. They need... You have to do math. This is one of the formulas you need to know. So if you know how to calculate VDVT, there are many things in the vignette that you will ignore, right, and just go to the data. Okay. Next. Okay, we'll see the correct answer. All right, so dead space fraction, again this is right out of Ira's lecture from this morning. So you're going to take your, your end tidal CO2, or your, sorry, you take your PACO2 minus your end tidal CO2 over your PACO2. And so in this equation, if I remember the numbers right, it was essentially 40 minus 15 over 40, which is closest to 0.62. So answer is D. I'm going to be the technical annoying person. This is not VDVT. This is alveolar dead space fraction, or AVDSF. So the way they asked the question is still appropriate, because they did not say VDVT in the equation. They said you're the dead space fraction. They will not try to confuse you, to Ira's point before. Like they're not going to put VDVT and AVDSF as both, both, you know, options within there. But you just go back to that concept of end tidal CO2, or CO2 from capnography, and the PCO2 is what you use to calculate that. Yeah. It hurt me to say that earlier, but that is what you need to do for the boards. Right. And then again, like Ira mentioned earlier, and I just mentioned that this is a long question. You know, a lot of words there, but if you go back to what they are answering you, you can just expedite the process. Okay. Next one. I can read this one. Sure. 17-year-old boy with asthma who weighs 130 kilos, ouch, was intubated in the emergency department where he arrived in hypoxemic respiratory failure. He's difficult to ventilate with abnormal arterial blood gas results consistent with a mixed acidosis, pH of 7.3, CO2 of 60, base excess of minus 6, or base deficit of minus 6. Which of the following ventilator settings on intermittent ventilation are most appropriate to improve his ventilation? All right. So there are your abbreviations up there. So volume control ventilation with a rate of 30, tidal volume of 450, plateau pressure of 29, pressure limited to 40. Pressure control ventilation, IMV at 12, PIP of 40, plateau at 28, measured tidal volume of 400. Volume control, rate of 12, tidal volume 500, plateau of 36, pressure limited to 55. PRVC, rate of 25, tidal volume of 600, plateau pressure at 30, pressure limited at 50. And pressure control, IMV of 20, peak pressure 45, plateau of 35, and measured volume of 390. Okay. I will go to the next one so you can see the options in the next one. Alright, good. So 77% of you got it right. Pressure control, IMBA 12, PIPA 40, plateau pressure of 28, and volume of 400. So conceptually, right, what are the important points here? So first, asthma, right? So asthma, long time constant, lower rate, right? So of the first options there, you're going to want to pick options that have a relatively low respiratory rate. So that sort of narrows it down to those two, right, that have the rate of 12. Generally those higher rates of 30 or 25 might be too high. The rate of 20 might be okay, right? Depends on the circumstance. Then what's next? So now it's determining what's the total, what's the pressure that the lung is seeing, right? And that determines the risk not only of lung injury, but also potentially the risk of hemodynamic compromise in patients with asthma in particular. So that's where we look at the plateau pressure. And with asthma, large difference between peak and plateau pressure is what's expected. So the peak pressure is not really the variable that matters when we think about the risk of injury to the lung or the hemodynamics. It's that plateau pressure. And so the plateau pressure in those circumstances, you want to keep that plateau pressure in a safe range, generally below 28 or 30. So of those options, those are the only ones that sort of met that combination of a low rate and an adequate plateau pressure. Does that make sense? Yeah. Yeah, that's right, that plateau pressure would be very high. What does that have to do with pressure versus volume? Yeah, so that's a good question, and I think there are zealots that would believe that in asthma you should use one versus the other, but rarely is that going to be, as Ira alluded to, there's no definitive data that's going to say you should use one versus the other, really in any diseased condition. So yeah, the reason is the plateau pressure there. Right now, impromptu, we can debate which mode is better, but there's no data. So that's what they're trying to get you on here is trying to get caught up in the different modes, right? But they're testing concepts, as Robbie just went through. Lowest plateau pressure, low rate. That's all they're asking. I think, just to take this to its clinical conclusion, and you said it, the real answer to this question is to get the patient to a spontaneous breathing mode with pressure support and CPAP, and take out... The patient's brain will figure out the best rate, the best volume under the circumstances. They asked me to expedite. It's going to be like the exam. We have limited time. What is the cause of reduction in right atrial filling during a supported spontaneous breath while mechanically ventilated? A, spontaneous inspiration creating negative interthoracic pressure. B, positive pressure ventilation creating positive interthoracic pressures. C, increased contractility affecting atrial transmural pressure. D, ensuring elevation of positive end-expiratory pressure. Or E, increased left ventricular wall tension. Okay. All in? Okay, so the correct answer was choice B, positive pressure ventilation creating positive intrathoracic pressure. And you can read this along with me. I think the highlight is the positive intrathoracic pressure impeding venous return into the thorax and to the right atrium. Let's see, mechanical ventilation does not really affect contrillity, contractility. Let me go back to the answers. Let's see, so spontaneous breathing creating negative intrathoracic pressure would actually augment preload into the thorax, increase contractility. We already just talked about that. That's not affected. Ensuring elevation of positive end expiratory pressure, that's like saying CPAP or PEEP. You wouldn't want to do that. And increased left ventricular wall tension, no comment. I'm not sure how that fits or doesn't. Okay. Okay, long question. So what are we going to do? Start backwards. What is the most likely diagnosis? looking for right main stem intubation, pleural effusion, endotracheal tube dislodgement, and tension pneumothorax or ventilator dysfunctions. Now you know what the aim here is. So it's a 12-kilo, two-year-old boy, influenza, superimposed staph and pneumonia, admitted to the PICU three days before, intubated in outside ED for severe respiratory distress. Sedated, paralyzed, SIMV, ventilation in the PRVC mode. Taut of arms at eight per kilo with a peak pressure of 35, PEEP's 10. Respiratory rate is 20 with an FIO2 of 50. Heart rate, 110, blood pressure 115 over 75. Pulse oximetry, 92%. End tidal, and we can get to the figures in a second, starting out at 55, and he has a low grade temp. PIP steadily increased, now up to 42. The ventilator's alarming for inadequate return tidal volume. The end tidal has increased to 75. Oxygen saturations are decreasing despite an increasing FIO2, and now the patient's acutely desaturated. 55% sat despite being hand ventilated off the ventilator. Vital signs, heart rate now 180, blood pressure's down to 65, 35. End tidal's down to 17, and auscultation reveals overall decreased and unequal breath sounds. So which is it, given all that information? It's showing figures. Here's your figures. So initially, end tidal is normal. Then in the second scenario there, end tidals are increasing, and then when the patient deteriorates, end tidals are decreasing. So here were the options. Thank you. Correct. OK, so the answer, obviously, tension pneumothorax. Lots of information here. What are they trying to get? One is acute change. So pleural effusion, none of you picked. Oh, sorry. I'm sorry. Unlikely, because that's not going to be acute. So it could be a right main stem intubation, endotracheal tube dislodgement, tension pneumothorax. We know it's not ventilated dysfunction, because nothing got better with hand ventilation. So you've eliminated those two. For a situation like this, where the end-tidal CO2 is decreasing, so what's going to end up in a situation where you decrease pulmonary perfusion? So tension pneumothorax is going to lead to a decrease in pulmonary perfusion. Could it be any of the others? Right main stem, entire CO2 is going to keep going up. If you end up blowing the lung, could you end up with a tension pneumothorax? Absolutely. But the answer then becomes tension pneumothorax. So when you really see that acute change in the drop in end-tidal, the increase in dead space ventilation, it's going to be tension pneumothorax. I think the one tricky thing is a lot of people picked ET tube dislodgement. And if you look at the tracings again, you shouldn't see this stepwise drop like that. ET tube dislodgement should be a relatively abrupt change. Yeah. Six-month-old boy, born at 28 weeks gestation, successfully repaired TEF. He is transferred to the NICU, or from the NICU to the PICU after trach tube placement for transition to home mechanical ventilation. While he is undergoing an attempt to wean him from PEEP of 10 to 6 over a period of several weeks, he develops worsening cough, increasing periods of desaturation spells, particularly when agitated or excited. Worsening lung compliance is noted by increasing plateau pressures despite tidal volume being unchanged. No changes have been noted in his secretions. He has not been febrile, has not shown other signs of illness. Chest radiographs have been unchanged for the past several weeks, including one performed this morning. Which of the following interventions are most likely to be helpful? A, lowering tidal volume and increasing rate. B, obtaining viral and bacterial studies from tracheal secretions. C, increasing FiO2 while he is awake. D, restarting sedation because he's not well enough to tolerate being awake. E, performing bedside flexible bronchotelic for optimal PEEP. All right, looks like pretty much everyone went with the correct answer, which is E. So looking at the question, a couple ways to think about it. You have an infant with BPD who's got a trach. The only change that they've said is that you've been solely weaning the peep. Nothing else has changed. And they kind of went out of their way to say no infectious signs. Nothing else has changed. So diagnosis is going to be most likely tracheobronchomalacia. That is worse as you've weaned the peep. So looking at the answer choices, lowering tidal volume and increasing rate will worsen your tracheobronchomalacia and spells, so don't do that. Again, no signs of a new infection, so unlikely to be B. No one picked increasing FiO2 while he's awake, which is great. And then restarting sedation, not well enough to start being awake. Like Ira said, it works, but it's not the right answer. So it's going to be E. Okay, three-year-old with AML recently finished a course of chemo, is admitted to the PICU with the following vital signs, temp is 39.5, heart rate 180, respiratory rate 37, blood pressure 64 over 37, CBC shows a white count of 0.2, hemoglobin is 7.6, platelet count is 67. Tensivus on duty determines that an internal jugular line should be placed. Which of the following needle insertion points and approaches is most appropriate? Anterior margin of the sternocleidomastoid halfway between the mastoid process and the sternum and directed towards the ipsilateral nipple. Base of the triangle formed by the clavicle and the head of the sternocleidomastoid muscle with the needle pointing towards the ipsilateral nipple. Posterior margin of the sternocleidomastoid cephalid to the area where it intersects with the external jugular vein and just under the clavicle at the insertion of the middle and medial thirds and slowly advanced while negative pressure is applied with an attached syringe. And the answer is, all of you use ultrasound, many of us. All of us, well, I don't know, Connie, but the rest, we learn without ultrasound. That is funny, because all of us, but I've got this one correct. Right. So the answer there is A, anterior margin of the sternocleidomastoid, halfway between the mastoid process and the sternum, directed towards the ipsilateral nipple. I think the rationale is actually pretty good, the way that it's written here. Just remember that if you, although, like I say, everyone uses ultrasound, but you may be traveling somewhere and they may not have it, or the ultrasound doesn't work for any reason. So they want you to know the anatomy, it's not just looking at the picture. Yeah, the main point about that middle, it's the apex of the triangle and not the middle of the triangle, for those of you that picked that other choice. And you probably know that an old timer wrote this question. So on the American Board of Pediatrics, all of this fluff at the beginning of the question wouldn't be there. The only thing that was needed on that last question was the last sentence. You didn't need any of the other stuff. And as questions go through editing, a lot of the fluff that doesn't need to be there is taken away, so you're not wasting your time or getting confused by information that's not really relevant. A 30-kilogram, 10-year-old boy with ALL is admitted directly to the peds ICU in shock. He's appropriately resuscitated, is ultimately found to have gram-negative sepsis, and improves. A right radial arterial line had been placed for hemodynamic monitoring during treatment. The line was maintained during convalescence for a laboratory sampling. In the early evening on day 10, the waveform becomes dampened and drawing blood is difficult. The arterial line is left in place to obtain morning laboratory samples with the intention of removing it in the next day. By 7 o'clock in the morning, the waveform is now absent, and the index finger appears poorly perfused. The arterial line is immediately removed, but the index finger ultimately requires amputation. Which of the following interventions focused on the arterial line would most likely have prevented this complication? A, use of a cut-down technique rather than percutaneous insertion technique. B, remove the line once continuous hemodynamic monitoring was no longer required. Choice C, use of an Allen test to confirm collateral flow through the ulnar artery. D, remove early on day 10 once the waveform became dampened and blood drawing became difficult. Best choice. And more than half chose the correct answer, removed the line once continuous hemodynamic monitoring was no longer required. This is a basic tenet of quality improvement. If what you're doing is no longer needed, get rid of the device, then you can't have the complication. Blood can be drawn from a central line or even a peripheral line with a tourniquet for blood sampling. So that's the basic underlying idea with this question. Many of the tests I've taken, there's some version of this, if you don't have the line, you can't have the complication. If you don't have the catheter, you can't have the complication. So they're looking at get rid of the device. And the Allen test demonstrates collateral flow of the alternative artery, but it doesn't prevent the complication of occlusion. And cut down actually would increase the risk of this adverse event, not decrease it. Do we still inhibit the line, basically? Just for labs, right? Yeah, this patient had improved. It was for labs. Hemodynamically. The line was maintained during combat lessons for laboratory sampling. So, and that's something. Same thing with centralized, no? And we all say, well, it's so difficult to draw. Let's keep it an extra day, and again, fever. So, yeah. Okay, so we will have to stop at 1210. Yeah, and then we will have lunch and keep going. Okay, so last question. Three-year-old, 15 kilo, idiopathic pulmonary hypertension. This boy presents to the ED with shortness of breath and syncope. Parents report he's been coughing with a low-grade fever. There are sick contacts at home. Heart rate, 120. Respiratory rate, 34. Blood pressure, 82 over 40. And oxygen saturation, 88%. Pale, diaphoretic, 1 plus pulses. Which of the following is the most appropriate first step in management while in the emergency department? So, oxygen, nitric oxide, sildenafil, intubation, Lasix. Can I go next? Yeah. Can't open it yet until she advances. Yeah, okay. Yeah. Okay. OK, and the answer is? OK, so the key there was first step, right? First thing, ED, oxygen. Can you do some of the other things if needed? Of course, but it's always first step. What really gets hard is when, on some of the questions, you do two things at once. Those are just frustrating. You've just got to figure out what the most average common practitioner would have done first. Thank you. So these questions are actually in the last question block, but I think we got to 11 of the 25 questions in the block, and these were at the end. And I knew how disappointed the group would be for not getting to do the statistics questions, and so here we are. So there's four questions. I'll be honest, two of them are pretty hard, so don't feel bad if you don't get them right. So I'll go ahead and start. So you'll have to use your phone to scan the QR codes on the next slide, just like we did before. So in a meta-analysis evaluating the effect of dexamethasone administration on successful extubation in the 24 hours preceding extubation in the children in the pediatric ICU, the dependent variable is the what? A, is dose of dexamethasone used in each study? B, rate of re-intubation in each study? C, effect size drawn from each study? Or D, aggregate rate of extubation success across all studies? So the question is, which is the dependent variable in this meta-analysis? Thank you. And then is it timed, or does it stop when I push the arrow? Yes, it will stop when you push the arrow. I was just trying to see if you were just slowly moving. I think you can just click. Then it will stop. OK. And then we'll give you the right answer. OK. Yeah, you can go ahead. All right, so this is one of the ones I thought was a pretty tough question. And here we go. So let's walk through this and why this is the correct answer. So first of all, we're talking about dependent variables. And so this is going to be, again, the outcome that we're looking at in our meta-analysis. So for choice A, dose of dexamethasone used in each study, that's going to be an explanatory variable that we track. That's not going to be the outcome that we're interested in. So then you're looking at B, C, and D. So we'll stick with the answer choices for now. Then we'll go to the rationale. So a meta-analysis is different than a randomized control trial. And so if you think about the results you've seen from a meta-analysis, where, again, you're putting multiple randomized control trials all together, you're trying to compile all the sample sizes for both the treatment and control groups. You're looking at the results of each trial. You don't care as much about the statistical significance of each individual trial, because what you're doing is you're compiling all the patients that were given the treatment decadron versus the controls that weren't given decadron. And so what you're really looking at, if you look at the forest plot that you see for every meta-analysis, is the effect size. That's the diamond that you see in the forest plot for each trial. And basically what that is showing you have one side that favors the treatment using decadron and one side that favors control not using decadron. And you'll see the effect size that was seen in each study. And then at the bottom, you'll see that overall effect size for decadron after compiling each of the individual studies. And so the dependent variable for a meta-analysis is actually the effect size from each study. That is a hard question. I would be surprised, again, if they asked you one that was this difficult on the exam. But they might ask you something about a meta-analysis. So I think things to know from this question. One, again, expandatory variables versus dependent variables, what you're looking at. Two, general, what is a meta-analysis and why do we like them? We like them because they're taking individual randomized control trials which have smaller sample sizes individually. And we're compiling it into one big study that's taking all the results for all the randomized control trials to try and come to a conclusion. So this is a hard one. Don't feel bad. Those are, I think, the things to take away from it. All right. Next one. In left skewed data, which of the following tenets is correct? A, mean is greater than median. B, mean is greater than or equal to median. C, median is greater than or equal to mean. D, median is greater than mean. Or E, mean is equal to median. All right, so the correct answer is D, median is greater than mean. So things to remember for distributions, if you have left skewed data, which is pretty uncommon, most of our ICU and healthcare data is right skewed, so left skewed data will have the tail towards the left. So the values will be lower in the tails and most of the values will be on the right. And so remember mean gets pulled by the tails, so in a left skewed distribution the mean will be pulled left, which is lower values. So that's what makes D, median is greater than the mean, the right answer. C is tricky because it says greater than or equal to the mean, but again we're trying to be concrete with these questions and so the general tenet would be that in left skewed data the median would be greater than the mean. And here's a nice distribution, so this, you can't see my pointer, but the this left one is left skewed data and you can kind of see how that left-sided tail pulls the mean that way, and so the median is greater than the mean. Okay, two more. Long one. Researchers recruiting subjects for a study of pediatric ICU survivors who received mechanical ventilation for more than one week. The IRB has deemed the study to be risk level 3, which means a slight increase over minimal risk without possibility of benefit to the individual subject, but agrees that it may provide important generalizable knowledge about the subject's illnesses. An 11 year old girl who is a potential subject is in the pulmonary clinic today accompanied by her mother. The patient has normal development and cognition, her father is currently at work, her parents are married, and all three of them live together about 20 minutes away. In order to properly consent the patient for the study, in addition to obtaining the mother's consent, the researcher needs A, child's assent only, B, father's verbal consent over the phone with the assent given to the child to read, C, the father's verbal consent over to the mother over the phone and the child's assent, D, the father to meet face-to-face with study personnel to sign the consent with the assent given to the child to read, E, the father's consent over the phone with a faxed copy of the signed consent sent to the research team, and the child signing the assent. So almost there, there we go. All right, let's go ahead and stop it. It is E. So, this is another pretty tough one. Again, I don't think that they would go quite this in-depth in general for the boards, but there's some important things to take out of it. So I think the first thing to know is what assent is versus consent. And so the definition is in the rationale, but basically they set the age of about seven. So if a child seven or older, essentially a child who is capable of having a conversation with and understanding to some degree what's going on, you have to get their assent. Second thing to know is what, you don't really have to know what risk level three means, but you need to know that it's a slight increase over minimal risk, and it doesn't have possibility to benefit the individual subjects. So basically this is a child, and you're going to do this study that is more than minimal risk and can't benefit the child. So that should just make you think, I need all the permissions that I can get because it's more than minimal risk and might not benefit the child and unlikely to benefit them. So, the way that that, what that means is that mother and father is going to have to provide written signed consent, and then the child needs to have assent in some form, and because it's more than minimal risk and the child's old enough to write, they have to give written assent. So again, I think this is harder than the ones that you will likely see on the test, but things to know what assent is, and then just if there's more than minimal risk in a study that you typically need, both parents written consent. All right, last one. Sixteen-year-old girl was driving, and she was broadsided by a log truck at an intersection. She was extricated from the vehicle two hours... No, sorry. This is the one that looks like a biostats question, but it's not. Here we go. Okay. Group of investigators plans a double-blind randomized controlled trial with a two-by-two factorial design to compare drug A to placebo and children admitted to the PICU with septic shock. Primary outcome measure is death from any cause within 28 days of randomization. Secondary outcome includes number of days alive and vasopressor free days and mechanical ventilation free days. Using data from previous studies, the investigators anticipate that the mortality rate in children with septic shock would be 25%. Investigators would like to estimate the number of children required to be included in each arm of the study to achieve 80% power using a two-sided hypothesis test and critical level of significance of 0.05. What additional information is required in order to calculate the sample size? A, number of children anticipated to discontinue drug A or placebo before the end of the study period. B, number of vasopressor free days and ventilator free days anticipated in children with septic shock. C, the statistically significant difference in mortality rates between the two groups. D, the smallest improvement needed to be seen for drug A to be considered clinically effective and superior to placebo or clinically ineffective and inferior to placebo. All right, we'll go ahead and stop it there, all right, nice job guys. So D is the right answer. So sample size calculation, essentially you need a couple different things for the sample size calculation. So you'll need to know your type one error rate, your type two error rate, which is converted to power. And then you'll need to know, based on previous studies, usually retrospective studies, what you think the effect size of your intervention is going to be. And that's essentially what D was describing as what an effect size is, which is the smallest improvement needed to be seen for drug A to be considered clinically effective and superior to placebo. So a lot of times this is calculated using something called Cohen's D, which is the effect size. And with those pieces, you can calculate a estimated sample size for a study. So those are all the stats questions. Again, I think these are a little bit tougher than what you would see on the actual test. I think the principles that I went into in my lecture are where most of your money is going to be. But again, some good principles to take out of this one. Do we have time for, I don't know if there's any questions, or if we have time for questions, but. Awesome. Thank you, guys.

pediatric intensive care

clinical scenarios

QR codes

answer discussions

exam strategies

ARDS

asthma management

statistics

interactive session