E77 Masterclass With Colin Schieman On Lung Nodules And Lung Cancer Screening
Listen to this podcast on SoundCloud
Chad Ball 00:12
Welcome to the Cold Steel surgical podcast with your hosts Ameer Farooq and Chad Ball. We've had the absolute privilege of chatting with some amazing Canadian as well as international guests over the past year. While the topics have been broad in range, whether clinical, social or political, our aims for the podcast continue to remain the same. We hope to inspire discussion, creativity, scholarly research and career development in all Canadian surgeons. We hope you enjoy our second season as we continue to highlight some incredible guests, deliver detailed masterclass sessions on a myriad of clinical topics and introduce some fresh new features such as debate and companion formats. We hope you relish the podcast as much as we do.
Ameer Farooq 00:55
In this episode, Dr. Colin Schieman, a thoracic surgeon at the University of Calgary gives us a masterclass on his approach to lung nodules and lung cancer screening. If you haven't heard our conversation with Dr. Schieman on direct entry surgical training and his approach to intra operative teaching, make sure to check out that episode as well. Links are in the show notes.
Chad Ball 01:34
How do you define incidental lung nodule in 2020?
Colin Schieman 01:39
So I'm grateful to talk about this topic as it's a hugely common problem. And as you say, it actually has relevance for I would guess literally all surgeons, regardless of specialty, I think trauma surgeons, vascular surgeons, joint surgeons will be responsible for managing this surprise finding at some point, I don't think that's a stretch. So it's probably the most common referral to our group would be my guess. And so, I would say, for, for the purposes of this discussion, it could be thought of as a nodule picked up unexpectedly on imaging, in an obviously in an adult patient, who has no symptoms attributable to the nodule. So you know, this isn't the person with pneumonia, you see, oh, and a pacification. On our imaging, this is the the patient that you're getting ready for their rectal cancer excision, or they have a surprise nodule. And, you know, technically are typically these are three centimeters or a size in less contained within the lung. But those cut offs exclude a large number of the surprise lumps we have to sort through. They can be single or multiple solid or as its called subsolid or round glass densities. So that's kind of how I would say, you know, any surprise nodule in the lung that isn't directly attributed to the problem that you're seeing them for, I think would fit this description in terms of how I would approach it.
Ameer Farooq 03:14
How common of a problem is it to see incidental lung nodules and, and how many of those, let's say, in the population of incidental lung nodules, end up making it to your office?
Colin Schieman 03:29
So it's actually a hugely common issue. It thankfully, the vast majority picked up incidentally are not cancerous. In some studies, up to a third of chest CT scans will detect the nodule of some variety or the other. And so it's it's extremely, you know, dependent on the radiologists interpretation, where they choose to put focus on things and, and so much to the, to the fact that in certain places like in southern Alberta, for example, Alain Tremblay and the team have worked hard to have flags triggered in the in the radiology sites, you know. This is a concerning nodule, this should generate a referral to an expert having little blurbs at that like that at the bottom instead of just giving the classic radiographic description of what the nodule looks like. So it's a super common problem here.
Ameer Farooq 04:26
Obviously, you know, the whole concept of lung cancer screening is a bit tangential to this discussion, but I did want to talk about it briefly, just because I think it dovetails nicely with the whole concept around radiology and CT scanning everybody and and sort of thinking about things from a population level. Can you talk a little bit about where we are in terms of lung cancer screening on a population level in 2020, and particularly in Canada, and if there's any evidence behind lung cancer screening?
Colin Schieman 05:01
Sure, yeah. So you're absolutely correct. Lung cancer screenings, it's actually remarkably complicated in a way. It becomes quite mathematical and scientific, fairly quickly. I don't, I certainly don't profess to be an expert on this issue. But in brief, I would say that there's a few fairly safe conclusions that that result from the body of literature that looks at this, this issue. And, and so first, I guess, just as background, as you guys know, lung cancer still remains the most commonly diagnosed cancer and is, is the leading cause of cancer deaths in Canada. To put in perspective, it's easy to forget this, but more people die of lung cancer every year than colon cancer, breast and pancreatic cancer combined. And so it's still despite impressive reductions in smoking a huge societal issue. And so because of its incidence, and because of its lethality, coupled with the fact that early stage disease has quite a high survival rate, it's very, very appealing to think of it in terms of a disease that's amenable to a screening strategy. And so it's been, I would say, it's been fairly well studied in the last decade. Unfortunately, chest X-ray is not effective as a screening modality, there's really good evidence to say that. And low-dose CT scanning is much more sensitive than chest chest X-ray. And there's there's now at least two large randomized trials with low-dose CT scan screening that demonstrate reduced lung cancer specific mortality when it's used as a screening tool. And in general, these are patients that have fairly extensive smoking history. So these are targeted screening interventions, 30 pack years, often greater than 55 years in age. And so in some ways, the science is settled it, it's probably effective. And just given the sheer numbers of people involved, it has great potential for society, such that in Canada, several societies but but probably most most notably the Canadian Task Force on Preventative Health Care advocates for lung cancer screening. And so that's all fine. Unfortunately, that bumps up with the harsh reality is that it's extremely difficult to implement. And, and so much so that in Canada, it's not an exaggeration to say there are no population-based screening programs. There are many smaller ones, piloted ones, university-based ones, small group funded one. So I don't want anybody to misinterpret that there's several programs happening in the country. But it wouldn't be a stretch to say these are largely small scale. And the challenge is, is there's many challenges, but the challenge is, when you take that high-risk cohort of patients and you screen them with low-dose CT scans, a lot of them have nodules, as I've already hinted at, a lot of CT scans in general have nodules. And about 40% of these patients will have a nodule, and then if you really trickle down through the data, 96% of those are false positives. And so you have to have a fairly expert team of clinicians and radiologists and and counselors, quite frankly, surgeons, to work through how best to evaluate these patients, because you run a very high risk of overtreatment. And so it's it's not just something that that you can do, because you buy three new CT scanners, it legitimately has to be delivered as a fairly comprehensive program, which entails, you know, secondary risk prevention and smoking cessation, and a lot of other elements to it. And not, not the least of which is it's fairly stressful for patients to go through and be told, oh, you do have a spot in your lung and, and then, as you guys know, there's other incidental findings which arise in the screening world. And so, you know, to put in perspective, it's, it's got a huge number of very expensive and fairly burdensome implications if it's rolled out. That's not to say that I'm opposed to it, by the way, the Canadian Association of Thoracic Surgeons also advocates for it, but it's we're still quite a ways away. I remember I was at a meeting a few years ago, and the the false positives and the incidental findings are so significant that one of my colleagues who was at a private setting in the United States got up and said the implementation of our screening programs is the is the best thing we've done. The downstream revenue that we've generated from this from the thyroid nodules and the aortic aneurysms and the liver nodules is massive. And, you know, it was just, it was astounding to hear somebody think about it and talk about it in that way when, when I think when you and and I would hear that you think, wow, that's sounds dangerous, you know, like, there's a lot of cost and harm and, and risk there. And so that's kind of the current status of things in the lung cancer screening world. I mean, we could, we could talk a lot about the studies and the data. But that's kind of the overview.
Ameer Farooq 10:42
Yeah, that's great. I mean, it just highlights how complex these decisions are, when you translate them from the the randomized control trial to an actual health policy platform, it becomes very complicated very quickly. So thanks for breaking that down for us. So let's say you get a patient in clinic that gets referred to you for an incidental lung nodule. You know, aside from the obvious history of smoking, what are those sorts of things are you asking them on history? And looking for on physical exam?
Colin Schieman 11:14
Yeah, that's great. I would say, just before I answer that, I would say that the proper evaluation of a nodule doesn't need to be a big, scary or problematic exercise. And I think the only real error that we run is, is not having the nodule formally evaluated by somebody with a bit of experience. And so if your algorithm is simply to get a CT scan and refer the patient, I don't mean that in a way to be condescending, but that that's probably going to cover you if it isn't something you're interested in. The the evaluation in some ways can be fairly nuanced, although I must say it's often fairly straightforward. And it's largely based on the CT scan. So so to answer your question, you know, the whole, the whole evaluation is really designed to determine if the nodule is cancerous and and so, to begin, when I'm first encountered with the patient, you know, as you've already alluded to, the history is fairly targeted, you know, smoking status, history of malignancy in general, infectious symptoms, traveled to endemic areas where fungal infections can arise just in the air and evolve to form granulomas, such as in the southwestern United States, or the Mississippi River Valley. We have in Alberta, a huge number of patients which winter in Arizona, Mexico, Texas, and they come back with granulomas, and when I worked in Ontario, we had a huge number of people that got granulomas from histoplasmosis, traveling into the United States. And so the history is that those are the big things. I mean, there's, you'll also, you know, we start to transition our brains into therapy mode and try to get a sense of their functional status. And almost always as you as you remember, me, or these patients have comorbidities and you sort of start risk stratify them in your brain. The physical exam, I must confess, is fairly unhelpful, aside from just a general assessment of frailty. So there's, there's no, I don't think it's worth getting too concerned about that. Yeah, that's that's sort of how I begin.
Ameer Farooq 13:28
You know, when if you do some cursory reading about this, if you talk about using these quantitative models to actually risk stratify someone into having a higher risk of having a malignant nodule, is that something that you use routinely, in clinic? My recollection from, from being on the service was that it was not, but I'm curious if you had any thoughts about incorporating a standardized risk calculator into your decision-making process?
Colin Schieman 13:59
Yeah, so I think it's very individualized from a practitioner perspective. And I must say that I very often use the quantitative predictors to assist me. And I say that humbly because I, at points in my, in my career, I did not and I think I really incorrectly estimated risk just based on my Gestalt. And so I, there's there's been quite a few which have been developed. The most useful and accurate ones, in my opinion, are the Braak model, which is largely a Canadian initiative and the Herder model. I have those on my iPhone. App can be downloaded through a variety of interfaces, but it's sort of as the prominent nodule risk calculator. And, and I actually use them fairly often even even in patients that are unequivocally high risk. So you know, my elderly smoker with a speculated nodule, I often will punch them into the calculator just to get a sense of just how high risk they are. And I find that, in particular, it helps me to gauge my level of worry. And it's useful to show that to people. And so, you know, I would say that most experienced thoracic surgeons can look at a CT scan in a patient and decide if they're worried or not. And that's probably close to accurate enough. But if I tell somebody that their risk of malignancy is, you know, 90%, or if it's 50%, a lot of patients will interpret that very differently. So, I use them all the time, like I probably with a third or more of my knowledge of consults I see.
Chad Ball 15:39
You know, it's interesting if we take a bit of a tangent, thought to the future with the introduction of more and more artificial intelligence, you could almost see how, you know, AI in terms of screening not not just for lung cancer, but in terms of almost everything is probably the future. I mean, you look at some of this recent. Yeah, this recent AI work. And by asking a series of eight questions, they're predicting psychological diagnoses much better than DSM five, to a level of like, 95%. When you look at, you know, the stuff we've heard at MIT with diagnosing, you know, the presence of active COVID, just based on the auditory sound of a cough. I don't know what that means for lung cancer. But these are all real world things that are occurring now. And the the impact of AI, hopefully, for the good could be incredibly dramatic for screening so much of what we struggle with now.
Colin Schieman 16:40
Oh, it's, it's happening. I couldn't agree with you more Chad. Yeah, his ability to accurately measure and risk stratify a pulmonary nodule. I mean, there's, there's a lot of groups looking at that. It's, it's pretty impressive. I don't think we're going to be able to avoid that. I think it's probably going to outperform certainly guys like me, but I, I would argue most radiologists and and that's not to say to displace them. I think it's probably just to compliment as I use the nodule calculators to complement their interpretation.
Ameer Farooq 17:14
So while we're still waiting for, you know, Hal 2000 to interpret our X-rays and CTs, can you talk to us a little bit about what the important features are when you're looking at these nodules on X-ray and then, obviously, more importantly, on the CT scan? What like, what are the things that you're really practically looking for, to help you make a decision as to whether this is a malignant nodule or not?
Colin Schieman 17:43
Sure, yeah. So the plain film is, is, has its limitations, but it is can be very powerful. So I think the critical first question, whenever you're looking at a plain film that reveals a nodule is what does the old films look like? Because if a nodule is new, or it's growing, it's a very different scenario than one that's been there for 7 8 10 years, it immediately changes the entire conversation. But, but it doesn't get you super far. I think the CT scan is still by far and away the most valued workhorse part of the workup. And so, you know, that the standard CT scan of the thorax, is, is really the key investigation. And then to answer your question, there's, there's a number of things that that are fairly revealing on the CT scan. It shows you a ton of secondary information about the surrounding structures in the chest and things like that. But as it relates to the nodule, you know, size trumps all. The the size of the nodule is, is really its greatest predictor of malignancy, and sort of as a very crude cut off greater than 8 millimeters, it should get your attention. The tiny little 1 2 3 4 or 5 millimeter nodule, those are almost never anything major significance. And then obviously, there's a bit of a gray zone between kind of 5 and 8 millimeters. And so the standard things in here, I would say, you know, the size of the lesion, the radiographic appearance of the nodule, whether it's smooth and spherical versus, you know, bumpy and speculated, whether it's solid versus, you know, increasingly these mixed density lesions gives you an idea of the solid lesions being more common, but the mixed density lesions often being cancerous. There's a few occasionally you get lucky there's a few features which are fairly telling for it being benign, such as calcium or fat within the nodule. Obviously, none of those are perfect, but those would be fairly good. You know, that's probably a fairly good place to start. And you can get a fairly good sense, with the secondary features, the size, the speculation, the location, presence of smoking or emphysematous change, about your degree of risk just off that CT scan.
Ameer Farooq 20:14
I think a very practical question related to this. So you know, you've looked at the CT scan, you've got the history, you put them to the risk calculator, you've got, obviously a pre-test probability as to as to whether this is a malignant nodule or not. Where does biopsy fit into this? Because I think that is a question that really plagues clinicians across multiple disciplines, whether, you know, you're the general surgeon or the let's say, the colorectal surgeon is where I've seen it come up, you know, you've got a you've got a nodule, you're wondering, is this a malignant nodule? Is this second primary? Is this just an incidental thing? Where does in your mind where does a biopsy fit into that whole discussion?
Colin Schieman 21:00
Yeah, that's a good one. So I would say, there's differences in opinion on the use of biopsy, you know, within the world of thoracic surgery and respirology. But, in general, I think it's pretty fair to say that biopsy is increasingly being advocated for in published societal guidelines, and, and should be strongly considered for most lesions that are thought to be cancerous. I personally try and avoid resection for diagnostic reasons if I can avoid it, especially if it's going to be something that carries any risk, like an anatomic resection, like a lobectomy, or more than a lobectomy. And so, I'm a big fan of biopsies, I don't think, you know, to answer to your question as the nonexpert, are you putting your neck out there by thinking about biopsy or ordering a biopsy and answer is you don't have to be nervous about that. There's not a ton of downside. So, you know, the obvious downside is, you know, there are complications from the procedure pneumothoraces, and and potentially requiring chest tubes and things like that. I would regard those as inconvenient not not to trivialize them, but but they're, they're manageable problems. But, you know, the radiologists, for example, with percutaneous biopsy are so good that, that, you know, you're going to get an answer. And in well over 90% of patients if the nodules got any size and substance to it. And it's often helps like, like, all the decisions we make in medicine, they're based upon, you know, trying to piece together information, and they rarely gets you at a point of absolute knowledge about a problem. But I'm a huge fan of biopsies, whether they can be done endobronchially, if the lesions fairly central, or has a so-called airway sign where there's a little airway in it, we'll send those patients for endobronchial biopsy. But, but the others, the more peripheral ones are the ones that don't have an associated airway. We talked to the to the interventional radiologist and see if they can get percutaneous biopsies of those. So I'm a big fan of biopsies.
Ameer Farooq 23:11
I guess part of what I'm trying to get out with that question is, you know, the situation where you perhaps don't have a respirology or, you know, for example, you don't have a thoracic surgeon at your hospital, but you do have an interventional radiologist who could who says, well, I can buy a biopsy this for you. You know, do you think every lesion needs to have a, you know, a trained pair of eyes from either, let's say, experienced respirologist or an experienced thoracic surgeon to look at it before you make that decision? Because, you know, the one worry I would have is not so much about the complications, but sometimes about, you know, the rate of false negatives and things like that. Do you have any comments about that?
Colin Schieman 23:51
Oh, for sure. Yeah, no, I think, yeah, no, I think that's a good way of tempering my comment. I think the decision to order the biopsy does need to be done in the context of of the whole picture of what the nodule looks like, and where do you think the story is headed. And so. Because there is there is a fair amount of subtlety to the interpretation. And as you as you suggest, the nondiagnostic biopsy or the inconclusive biopsy, how do you interpret that? And if you really read the detailed report of the, the path report is, is there other information there that suggests this as a benign lesion, or things of that nature. So I guess what I would say is probably a good strategy to speak to the thoracic surgeon or the respirologist. See if you're contemplating ordering the biopsy, not because it's inappropriate, but it may just not be needed at that moment in time or there there could be other options.
Ameer Farooq 24:55
I guess the next logical question to ask you is when do you embark on surgery for these these nodules? And when when, you know, increasingly it seems like wedge resections are are being used? Is there ever a situation where you know off the hop and again, this might be too big of a question for for our podcast, but are there scenarios where you know that, you know, this person is going to need a lobectomy off the hop? How do you sort of think about that in your mind as to who gets surgery and what type of surgery?
Colin Schieman 25:29
Yeah, I mean, you're right, it sort of is getting closer to the heart of how we make our decisions in the office. I would say that, in general terms, typically, surgery is used for obviously resection of a biopsy proven cancer, which is, I would say, the most common scenario. Rarely for a nondiagnostic biopsy, as you've just alluded to. Occasionally, if there's just such significant patient concern that observation is just too unappealing, you know, and the patient's just not prepared to watch and come back in a few months. They just can't, you know, live a normal quality of life with this nodule, it has them concerned. Or occasionally, you know, the nodule is just either in too challenging of a place to biopsy, which is fairly rare, or it's just such a compelling story based on the CT scan and the PET scan the patient's history, that, that you're comfortable enough to just embark upon surgical resection. So to your point, you know, you're absolutely right there, there is a trend towards sublobar resections or wedge resections or segmentectomy, that seems to have absolutely happened in thoracic surgery, despite really no randomized evidence to support that. And so it, you know, occasionally, the diagnostic wedge resections are absolutely appropriate procedure to perform to somebody for both diagnostic and therapeutic purposes. I think even a large proportion of those could potentially be avoided with a preoperative biopsy. But certainly, if you're looking at at a more dangerous operation, like a lobectomy, or something like that, I personally feel a little bit more comfortable that I know what the pathologies that I'm contending with. So that's, I think that's probably a little bit more of my personal take on it.
Chad Ball 27:34
You know, Colin it's been very interesting listening to your your masterclass here on on pulmonary surgery and nodules. There's a lot of overlap and a lot of similarities as I'm learning today from you, between your your world and my world with regard to liver surgery. You can you can really, you know, cheat a lot of your comments over to the liver side of things as well. I'm yeah, I don't know if that's good or bad. But it's, it's interesting. You know, I was particularly curious if you would touch on, which patients do you generally assess as good candidates for MIS techniques versus traditional open techniques, just in general?
Colin Schieman 28:17
Sure. I would say that things have really shifted in a fairly rapid period of time towards the vast majority of lung resections being performed minimally invasively these days, across most of North America. You know, so, you know, the question often is, why can't we do this minimally invasively that we wrestle up against, and, and I still think there's, you know, I certainly have pushed too far in at times for I've done cases, which I wish I would have done open. There's other ones where I've done them open, I thought, jeez, I probably could have done that laparoscopically or I'm sorry, thoracoscopically. So for me personally, at this moment in time, I, I still think an open resection is appropriate for the larger tumors, mostly just because they're challenged to extrude out of the chest wall at the end of the case. And there's just for me, personally, there's a fairly close correlation between the larger the tumor the more difficult the resection is even remote from the tumor. So the bigger tumors tend to have more of this desmoplastic change in the hilum. They're harder to expose, the retraction forces required are just that much greater and therefore the pull on the arteries is a little bit harder. So so the size of the tumor is probably the big one for me. It's centrality, like if I think it's going to be fairly tight on the adjacent lobes, or the central airways despite really superb visualization thoracoscopically, it's still sometimes challenging to know how close the nodule or the mass is to the ongoing pulmonary artery to the remaining lung or to the ongoing airways. And so, so for the more central lesions, I think they just get done much tidier and safe, more safely and, and with better margins open. And so I'm fairly liberal at doing those open these days, even though I absolutely I think a lot of them could be done that. But I don't always think that's the best, you know, results for the patient. So there's that kind of trade off to not so much with what's possible, but what's best. And then, and then lastly, just if I if I think there's, you know, a risk of doing more than a lobectomy, for example. There's a lot of places that have published on minimum invasive pneumonectomies or minimum invasive bilobectomies. I've even done some of them. And I actually think, technically, they're not a huge stretch beyond a standard, minimally invasive operation. But I think, again, just touching on what I just hinted out, you know, there comes a limitation where I think it's, it's not necessarily appropriate, like I, I think if you're, you're doing a pneumonectomy in 2020, you probably have a fairly angry looking large central tumor. And, and that has implications for, you know, the safety of resection, and it certainly has implications for what kind of margins you're going to possibly be able to obtain. And so even if I can do the dissection and put the staplers in periscopically, I don't think that's best done. That's for for me right now. But but there's others in the world who would argue with that. So those are, those are my kind of criteria at the moment.
Chad Ball 31:46
You know, maybe I'd like to bring around to livers again, as a bias. You know, one of the one of the messages that I think we've been propagating for almost 10 years now, certainly in Canada, is that really anybody that has a colorectal liver metastasis should probably be assessed by a liver surgeon in a multidisciplinary manner, but there's no question that there's there's liver mets out there that could be resected, and could be treated more aggressively, for sure, a significant patient outcome, in terms of improvement, that are probably getting to a lot of these centers. And I think that that landscape is getting better, you know, by the year for sure. For our community general surgeons that don't have the privilege of working with you and your group and state quaternary facilities, who shouldn't they refer? Is it a similar discussion sort of send anybody? Or or is there a short list of take home pearls that, that they can safely not refer to a service like yours, or maybe an interventional pulmonology service ?
Colin Schieman 32:54
Yeah, I feel that same shift towards multidisciplinary and expert level evaluation for an increasing number of the things we see. Esophageal cancers, for sure. Lung cancers is more and more. And it's a little bit of a self-fulfilling prophecy, because I think the more that we do the trickier it is for the the community met surgeons, for example, to weigh in on things because they don't just see it as much as they used to, perhaps but. So so I would say, particularly in the era of telemedicine phone consults, and what we've gotten more comfortable with in COVID, there's not much that shouldn't be referred. And I mean that because it doesn't even necessarily mean the patient has to drive into town. I would say, you know, the, the patient that has really poor performance status, often, you know, the specific diagnosis and evaluation isn't quite as relevant, you know, then then the really high functioning patient. I would say, to our comment conversation earlier that these tiny little pulmonary nodules that are getting picked up on you know, every third scan, I think, do create a burden of potentially unhelpful work. I just, the challenge has, I don't have a nice algorithm for saying, don't worry about this one. Don't worry about that one. I would say that if they're less than 5 millimeters in size, they probably don't need a lot of worry. And I think we've protocolized the radiology well enough that we've done a good job of triaging the ones that that need to be seen. But I'm, I'm struggling to come up with a list for you of things that don't make a good referral. Like, it's pretty rare that I sit there and think this is a terrible referral I shouldn't have. Like just even if it's if it's simple and straightforward for me, it's pretty hard to come up with a list of things that I think we don't need to see.
Ameer Farooq 35:10
You've been listening to Cold steel, the official podcast of the Canadian Journal of Surgery. If you've liked what you've been listening to, please leave us a review on iTunes. We'd love to hear your comments and feedback. So feel free to email us at [email protected] or connect with us on Twitter @CanJSurg. Thanks again.