John Patlan, M.D.
Associate Professor
General Internal Medicine
The University of Texas MD Anderson Cancer Center
Welcome! My name is John Patlan. I am a physician here at the University of Texas MD Anderson Cancer Center in the Department of General Internal Medicine, and I am going to talk to you today about oncologic emergencies. Our talk will be divided into two parts. This will be part 1, and please stay tuned for part 2. And we are going talk about the classic oncological cancer-related emergencies.
Now, in our talk, I want to emphasize that all --- not all emergency problems are cancer-related. And I want you to be able to differentiate between cancer-related or oncologic emergencies and non-cancer-related emergency problems, which may also occur in cancer patients. But the focus of our talk will remain on oncologic or cancer-related emergencies, which are classically divided into structural problems, infectious problems, hematologic problems, and metabolic problems. And we will touch on each of those in turn, and I am going to give you an overview of how to diagnose and manage these problems in the emergency setting.
Now, I do want to emphasize that you should not assume that the symptoms that the patient presents with, or whatever acute decompensation prompts their emergency room visit, is because of the cancer or its treatment. Cancer patients have all the same comorbid medical conditions that non-cancer patients do. So they present to the emergency center with chest pain, which may not be related to lung cancer. It could be an acute coronary syndrome. They may come in with shortness of breath, which is not because of pleural effusion, a malignant pleural effusion. It may be congestive heart failure. They come in with COPD exacerbations, cardiac arrhythmias, GI bleedings, all of the same non-cancer emergency problems that other patients present with. It is important to keep in mind. The other caveat that I want you to remember is that in cancer patients, the signs and symptoms of disease may be blunted or masked. The things that are you classically taught to look for to try to rule in or rule out some diagnosis may be difficult to find in cancer patients, and this happens for a lot of reasons. One – steroid use. A lot of the patients receive corticosteroids as part of their treatment, which blunts the inflammatory response, which mediates a lot of the signs and symptoms you are taught to look for. For instance, patients with an acute abdomen may come in and have very little of its classical guarding or rebound that you may be expecting. Neutropenia similarly blunts the inflammatory response, and so patients could have fairly significant infections yet have very little objective findings to suggest that. They may have pneumonias with minimal or absent infiltrate on chest x-ray, for example. Also, cancer happens more often in elderly patients. So, all patients with advanced age may present with more subtle signs and symptoms of disease. So, one thing I ask all practitioners to remember is that you should have a higher pre-test probability of badness, and by that I mean whatever the worst diagnosis that this that could be leading to this symptom or this presentation, have a higher pre-test probability for that. So, if you think some test is indicated to evaluate for that, you might go ahead and order the test, because it is harder to figure out in cancer patients without that.
Now, the remainder of our talk will be focused on the classic oncological cancer-related emergencies. As I mentioned, the --- this topic is classically divided into four sections: structural problems, metabolic problems, hematologic problems and infectious problems. So, all of the rest of the talk at this point will be talking about emergency problems that are due to the cancer, the underlying cancer, or its treatment.
Now, structural problems are fairly easy to understand. A cancer mass lesion can compress or obstruct or invade some vital structure. The things that are typically included in the oncologic emergencies lecture are spinal cord compression, superior vena cava syndrome, and [I will] also talk about cerebral metastasis with associated edema, because that is a fairly common and urgent problem as well.
So, spinal cord compression is a fairly common problem. And this is usually occurs because patients have metastatic disease, which can really be from any primary site. But breast cancer, lung cancer because they are very common, prostate cancer because it is also common, and all of which tend to have bony metastasis, are the most common causes of this. And so what happens is that the patient develops a bony metastasis to --- usually to the vertebral body that extends directly into the epidural space and then compresses the spinal cord. If you see radiologically that you have epidural extension of tumor and indentation of the thecal sac, that is a radiologic definition of spinal cord compression whether or not the patient has associated clinical symptoms.
Now, this is going to require a radiologic study to diagnose this. The MRI is the preferred imaging mode. On this sample MRI, you can see that the arrows indicate two vertebral bodies, which are markedly different than the others. They are less dense. There is a compression fracture in the higher vertebral body. And you can see where the spinal cord is labeled C that there is some compression of the spinal cord by the tumor, which is indenting the thecal sac. So, that is radiologic spinal cord compression.
Now, clinically, there can be a whole range of presentations. It is hard to imagine, but some patients can be fairly asymptomatic, yet have that same kind of radiologic presentation that I just showed you. They may have a little bit of back pain, but it may not be striking, but an asymptomatic patient with that kind of MRI is uncommon, but possible. What is more common is that patients will present with back pain. Almost all of them will present with back pain. One clue that this is different than the typical garden-variety low back pain, which is one of the most common causes for any patient to seek medical attention, is that this is typically located in thoracic vertebral column. Some will be lumbosacral. Some will be cervical. Another clue this is a different kind of back pain is that this pain may be worse with recumbency or with valsalva, which is not typical for ordinary lumbosacral sprain.
Now, as things progress, they may develop neurologic compromise. So, they can find --- they can have motor weakness below the level of compression. An even more advanced sign of cord compression would be the development of sensory deficits. It is very difficult sometimes to localize a level of compression, and your objective sensory level may be several levels below where the cord is actually compressed. A very late sign would be bowel or bladder dysfunction. The thing that you would usually find will be urinary retention.
So, once you suspect this diagnosis, because of new or worsening back pain that is atypical or the patient is developing some neurologic compromise, it is urgent that you perform a radiologic study to establish the radiologic diagnosis. And, even when the diagnosis is suspected, even before the MRI is performed, it is reasonable to initiate corticosteroids to try to preserve neurologic function. Dexamethasone is the usual steroid that we choose. The optimal dose is really not defined. There has never really been any randomized control trial to give us the optimal dose. Standard dosing, however, usually is somewhere between a 10 to 20 mg bolus. Some people have used up to 100 mg or higher in patients with severe neurologic compromise and then some scheduled divided dose after that. You will need to get an urgent MRI as I mentioned. If patients can't get an MRI because they have got a pacemaker or they have metal in their body or there is some other contraindication, you can get a CT myelogram.
Now, once you have diagnosed this, once you have gotten your radiologic study, you see that there is indentation of the thecal sac, they will require some urgent treatment. Hopefully, you have already initiated corticosteroid therapy. Most patients will be treated with radiation treatment. A neurosurgeon should always be involved for consultation, and if the patient has been previously irradiated because of known vertebral metastasis, or if their spine is unstable, they will probably require neurosurgical intervention. You will see listed in various textbooks that chemotherapy is an option for very chemosensitive tumors such as germ cell tumors or lymphomas. In real life, most patients will receive steroids and radiation treatment up front.
Now, the prognosis depends upon whatever the neurologic status is at the time of diagnosis and when treatment is initiated. If the patients are ambulatory, there is a very good chance that they will be able to walk out of the hospital if they walked into your emergency room. If they do have some weakness, some lower extremity weakness, however, only about a third of them will remain ambulatory after we finish their treatment. If they are completely paraplegic, the prognosis is not good, only 6% will regain locomotion. Now, as I have described it to you, I mentioned that there is a whole range of symptoms from very little symptoms with radiologic cord compression, to back pain, to the development of motor weakness, to the development of sensory deficits, and finally to bowel and bladder dysfunction. This process from pain to neurologic compromise develops typically over a period of weeks. In one study, patients who are ultimately diagnosed with cord compression were asked to identify when did they develop new and worsening back pain and, on average, it was probably six to seven weeks when the back pain started. Now, that is good news and bad news. The bad news is patients were developing cord compression and it took six or seven weeks to be able to figure that out. The good news is that, in general, this is a kind of a subacute process, and so, you have time to work with. So, if you have a high-risk patient, say a patient with metastatic prostate cancer, known vertebral metastasis or a patient with some other high risk cancer with known bony metastasis and they have some suspicious clinical sign or symptom, new or worsening back pain, even if they don't yet have neurologic compromise, consider getting an MRI. It is not an emergency. You don't have to do it immediately, but if you can get it within the next 24-48 hours as long as they are neurologically intact, that would be reasonable.
Now, the next structural emergency we are going to talk about is superior vena cava syndrome. I will tell you superior vena cave syndrome, while it is always presented in the classic oncologic emergencies lecture, strictly speaking is not a real emergency. It is an urgent problem, but also a subacute problem that develops over a period of at least weeks. Now, this happens because you have, usually because you have some tumor that is causing obstruction of venous drainage from the head and upper extremities. Usually, there is a mass or lymphadenopathy in the mediastinum that causes external compression or sometimes invasion and thrombosis of the superior vena cava. And because this is a thoracic or mediastinal problem, thoracic malignancies, such as lung cancers or sometimes lymphomas, are the primary cause of this. Occasionally, we will also see it as a complication of the presence of a central venous catheter causing some intraluminal stenosis or thrombosis. Now, one problem that you will encounter is that vena cava obstruction, when patients present with this --- this is the presenting sign or symptom of their diagnosis in about 60% of the time. In other words, about 60% of patients with superior vena cava syndrome do not yet have a tissue diagnosis. This is the presenting feature of their disease, and this becomes a problem, as you will see in just a minute.
So, when they come in, they can have a variety of symptoms, they may complain of shortness of breath, more specifically, they may have some facial swelling, some cough, some edema of the upper extremities. Since this process develops over a subacute period of time or period of weeks, you develop collateral vessels over the anterior chest wall and the neck that frequently have some edema or plethora of the face. Very rarely, if the patients are very, very advanced, they may be obtunded if they have complete obstruction of the cerebral venous drainage, or, if they have airway compromise, they can have stridor, but that is very rare and you will likely never see that in clinical practice.
So, here is a picture of a man with superior vena cava syndrome. I will show you later an after picture after treatment. But I think in this picture you can appreciate that his face is kind of plethoric and ruddy, and when I show you the after picture, you will also appreciate how puffy, swollen and edematous his face is.
So, when a patient comes in, they have got these kinds of signs or symptoms, you probably have already had a chest x-ray showing some kind of mediastinal mass. What you will need to get is a chest CT with contrast so that you can understand the level and extent of the venous obstruction and see whether there is any associated thrombosis of the superior vena cava. If patients have an iodinated contrast allergy, you get an MRI with gadolinium.
Now, the problem, as I mentioned to you, is that most of these patients do not yet have a tissue diagnosis. They may have a non-small cell lung cancer. They may have a lymphoma. Those will both be treated very differently, and optimal therapy really is going to depend upon the treatment of the underlying cancer. So, you are going to need to get a tissue diagnosis. Now, when they come in and they have got this clinical problem, what was frequently done in the past was to do urgent radiation treatment. The problem with that is that, if you get radiation before you have obtained a biopsy and establish a tissue diagnosis, that can obscure your histology and really limit your ability to treat the patient effectively in the future. So, if the patient requires rapid relief of their symptoms, they have got fairly significant vena cava obstruction, you can put in an intraluminal stent to open up the vena cava and provide some venous drainage and that will not preclude any subsequent therapy that you are able to do.
This is a picture of a stent being placed by an interventional radiologist. On the left-hand picture, you can see a catheter being threaded from below and a blush of contrast is being injected to outline the superior vena cava and the venous circulation. On the right-hand side, you see the stent that has been deployed.
And here is the after picture of that same patient. So, you see on the right-hand picture, his face is much less red. He is much less puffy, and although he is not smiling in that picture, I am sure he is very, very happy that you have helped him and made him feel much better by relieving his obstruction.
So, as I mentioned, because this is a thoracic problem, it is caused by thoracic tumors, lymphomas, germ cell tumors, small cell lung cancers, those are usually very chemosensitive. And so chemotherapy with or without radiation is going to be the mainstay of treatment. Unfortunately, the majority of patients will probably have a non-small cell lung cancer. It is a poor prognostic sign when patients present with non-small cell lung cancer and superior vena cava syndrome. So, really you are going to be aimed at treating --- palliating their symptoms. Now, there have been questions raised recently about how effective radiation treatment really is at re-establishing SVC patency, but even when it does not re-establish full patency, it does provide significant relief of symptoms, so it is still frequently done.
The prognosis is really related to the underlying malignancy. If it is a good prognosis cancer, such as a curable lymphoma or a germ cell tumor, then the prognosis may be very good. If it is an advanced but incurable disease, like a metastatic or locally advanced non-small cell lung cancer, prognosis is not so good. One of the take-home messages that I want you to leave with though is that, although it is always discussed in the oncologic emergencies lecture, this is not really an emergency problem. It is usually not immediately life-threatening unless there is tracheal obstruction or severe cerebral edema with mental status changes. But as I mentioned, that is very rare, so again a subacute problem, you have time to do your job.
The next structural problem we are going to talk about is cerebral metastasis, unfortunately, a very common problem and sometimes not easy to diagnose. Patients can present with a whole range of symptoms: headaches, sometimes focal neurologic symptoms, sometimes intermittent or transient neurologic symptoms. Patients may have a suspected transient ischemic attack when really it is a tumor that is causing this problem. They may have seizures, occasionally sometimes just idiopathic nausea and vomiting, which is likely related to cerebral edema. So, if patient --- if a patient with a cancer comes in and they have got nausea or vomiting and there is no good reason for it, they are several weeks out from chemotherapy, they are not hypercalcemic, consider cerebral metastasis is a cause. Now, a lot of this --- in general, the symptoms that are produced by the tumors are in large part mediated by the vasogenic edema of the tumor, so relief of that edema is going to help you with symptom management. When you have a patient and you suspect that this may be the problem, you are going to need to get an imaging study. If you don't have much time to work, you can get a CT scan. A more definitive study would be an MRI, and it will be important to initiate corticosteroids to reduce edema to try to get control of their symptoms.
So, this is an MRI showing a patient with multifocal brain metastasis. On the upper left-hand view, this is --- that is non-contrast enhanced lesions, and I think even the non-radiologists among us can appreciate there is some asymmetry between the two sides. There is some hypodensity on one side that is not seen on the other, and on the other three images, which are enhanced with gadolinium contrast, you see the arrows pointing to the multifocal enhancing lesions. So this is unfortunately a --- unfortunately a patient with multifocal brain metastasis.
Now, what you are going to do is, in the immediate emergency phase, you are going to start them on corticosteroids to try to reduce the edema, improve their symptoms. They can work very dramatically and very quickly, just within a few hours patients can feel much, much better. We generally use dexamethasone because of its lack of mineralocorticoid activity. If patients have cerebral edema and some increased intracranial pressure, you would not want to give them a steroid that could potentially further raise their blood pressure intracranial pressure. And again, as in spinal cord compression, the optimal dosing is not really defined, but, in general, a 10 to 20 mg dose as a bolus followed by a divided dose over a period of time.
Now, because your patient with cerebral metastasis is going to remain on corticosteroids for some period of time, it is going to be important for you to try to remember to try to prevent the complications of corticosteroid therapy. Prevent ulcer formation with some acid suppression. You are going to be aware and look for hyperglycemia should it occur and it occurs very commonly. And you should be aware that corticosteroids can have less commonly recognized side effects, mental status changes. Sometimes patients just feel sort of wired and excited. Sometimes they can be frankly psychotic. Over a long period of time, they can develop some steroid myopathy. Unfortunately, the only treatment for that is withdrawal of steroids, and, if patients are maintained on steroids for a longer period of time, they have an increased risk of opportunistic infections such as pneumocystis carinii, so they may require some prophylaxis for that. So, because of these problems, we will try to minimize or lower the dose as we can to try to control their edema, yet minimize these side effects. And, if in the emergency setting, patients have very significantly increased intracranial pressure, they are developing some focal neurologic findings, their mental status is declining, you are going to need more urgent measures to control their intracranial pressures such as mannitol, fluid restrictions, and neurosurgical consultation.
So, the prognosis is not good. This is a bad place to have your cancer, so it really depends upon the patient's overall status, their age, their performance status, and how likely it is that their treating oncologist is going to be able to get control of their extracranial disease. And in general, patients are divided into favorable or poor prognosis. If they have favorable prognosis, based on those factors that I mentioned, survival is about seven, a little over seven months. Poor prognosis, they have more advanced age, poor performance status, very widespread systemic disease, then survival is only about two months.
So, next we are going to talk about metabolic problems. We see a lot of metabolic problems, both in and out of the hospital in cancer patients. Common problems include hypercalcemia, hyperuricemia, or tumor lysis syndrome, hyponatremia, which is probably the most common electrolyte disturbance of…in any hospital and hyperkalemia, which can happen for variety of reasons. We are really going to focus our discussion on hypercalcemia and tumor lysis syndrome.
So, hypercalcemia is a very, very common problem. It occurs in up to 20% of all cancer patients and can occur across a whole range of tumor types: leukemia patients, lymphoma patients, a variety of solid tumor patients such as breast cancer, lung cancer patients. And it can happen for variety of reasons. Intuitively, it can happen because the patient may have osteolytic metastasis and then release of calcium that has been locked up in the bones there. But more commonly it happens through paraneoplastic means, because the tumor can produce a parathyroid-related protein, which is sort of a false hormone, which then causes release of calcium from the bone into the circulation.
The symptoms early on are kind of nonspecific and more --- as the hypercalcemia progresses they become more pronounced. In general, patients complain of kind of generalized weakness, they are lethargic, they are fatigued and they may have some nausea, vomiting, constipation, and anorexia. This is all very nonspecific, and this is probably the typical symptom profile of any cancer patient that is getting active treatment. As things get more advanced and the serum calcium gets higher and higher, they may start to develop some mental status changes, may be very sedated, sleeping, you know, 12 to 18 hours a day, and may have some cognitive dysfunctions, difficulty concentrating, saying things that don't make sense, maybe hallucinating. And because hypercalcemia impairs your ability to repair, to reabsorb free water, patients develop polyuria and polydipsia and virtually all patients who have clinically significant hypercalcemia are significantly volume-depleted and dehydrated by the time they present to you. In various textbooks, you will also see hypertension as a manifestation of hypercalcemia and shortened QT interval and a predisposition to arrhythmias as a risk. Although, in clinical practice, this is virtually never seen.
Now, what you should do when the patient comes in, you have identified them as having significant hypercalcemia for which they are symptomatic. If they have kidneys that work, you are going to need to increase their urinary calcium excretion. As I mentioned, virtually all of these patients are volume-depleted, so they are going to require a lot of volume replacement. So just give them lots and lots of isotonic saline. This will facilitate some calcium uresis. Now, traditionally, we have also used loop diuretics in conjunction with saline. This is one of the very few indications for concomitant administration of both saline and diuretics. There is actually very little evidence for that. There are really no randomized controlled trials. The use of this is based on some old studies from the 1960s. We still do it, but if you do do it, be sure to try to replace other electrolytes which may be depleted such as potassium or magnesium. Now, saline calciuresis will help you quickly and in the short term, but it is not a very potent treatment. So, it will bring down your calcium, but it will not bring you down close to normal just yet.
You may require some other treatment, which is to --- which is really going to try to turn off the problem at its source. As I mentioned to you, the problem is usually paraneoplastic, because you have excessive bone reabsorption and release of calcium from the bone where most of it is stored. The most potent treatment you have available to you is at the bottom of the page here, which is bisphosphonate therapy. A variety of bisphosphonates are available. The one that is probably most commonly used these days is zoledronic acid or Zometa®. The reason it is popular is because it can be used…it can be infused quickly over about 15 minutes or so as opposed to pamidronate, which is one of the older bisphosphonates, which required a much longer infusion up to 24 hours. The problem with all the bisphosphonates though is they have a slow onset of action. You are really not going to get any peak onset for several days, but they will last for a long time, in general for about 3 to 4 weeks. So, saline diuresis gives you a quick treatment, but not very potent. Bisphosphonates give you a potent treatment, but not very quick. What you can use, sort of in between, is calcitonin. This is delivered subcutaneously and has an onset of action of just a few hours. It will only last for maybe 24 to 48 hours. After that, you will develop --- the patients will develop tachyphylaxis and it will no longer work. And, in the patients with very severe hypercalcemia, you will probably require all three of these modes of therapy.
Now, the next metabolic problem we are going to talk about is really only seen in cancer patients, and this is tumor lysis syndrome. And by the name syndrome, obviously we are referring to a group of problems that tend to occur together. The basic underlying pathophysiology is that you have some bulky tumor with a high tumor burden and you have a rapid cell turnover and release of intracellular contents. So, you will have release of potassium, which is a primarily intracellular ion, release of phosphate, which is also primarily intracellular, and if patients have significant hyperphosphatemia that can then precipitate with calcium and so you have a secondary hypocalcemia. You have release of nucleic acids and purines, and that is metabolized to uric acid, so you develop hyperuricemia, and ultimately this can produce acute renal failure.
So, as I mentioned, this is common in patients with high tumor burden. We typically see this in poorly differentiated lymphomas, like Burkitt's and some leukemia patients with very high tumor burden, more commonly in acute lymphoblastic lymphoma, then acute myeloblastic lymphoma. Usually, this is something that occurs after treatment. This is usually a post-treatment problem. Rarely, but possibly, it can occur spontaneously. The classic would be a Burkitt's patient. And what is happening is that you have this rapid cell turnover, release of intracellular contents, the purines get metabolized to uric acid, the uric acid is insoluble in acidic urine. And normal urine is acidic, so these uric acid crystals crystallize in the renal tubules and then start to obstruct the tubules and cause renal failure.
Now, because all of these parameters, serum uric acid, potassium, phosphorous, calcium, they all exist on a continuum, we have criteria for diagnosing tumor lysis syndrome. The criteria that I am showing you here is the Cairo-Bishop definition, which is the most commonly used definition for tumor lysis syndrome. Patients are said to have tumor lysis, laboratory tumor lysis, if they satisfy any two of these criteria: the uric acid is above 8, potassium above 6, phosphorous greater than 4.5, or a calcium less than 7. Now, they are said to have clinical tumor lysis if they have any of those two plus the development of renal failure, and the usual criteria is creatinine is greater than 1.5 times the upper limit of normal, or because of hypocalcemia, they are developing arrhythmias or seizures.
Now, the treatment of tumor lysis is difficult. If patients don't present, don't begin to receive treatment until after renal failure has occurred, then this is very difficult. You can try to hydrate them, give them some diuresis or some saline to try to wash out the obstructing uric acid crystals. This is another case where frequently we will give intravenous fluids and loop diuretics to try to facilitate washing out these crystals. You can also use a recombinant uricase, for instance Rasburicase, to try to catalyze the destruction of the insoluble uric acid. And it will then form a water-soluble substance called allantoin. What is not helpful at this stage is administration of sodium bicarbonate. Recall that I mentioned to you that the uric acid crystals are insoluble in the normal acidic urine. What is intuitive is that well, maybe if we alkalinize the urine by giving the patients sodium bicarbonate, that can help prevent the formation of these uric acid crystals and treat this problem. Unfortunately, once renal failure has also --- is already occurred, this is not helpful and just buys you more metabolic problems. So, I would not recommend it at this phase. If patients have very advanced disease and are not responding to the treatment that you have instituted so far, sometimes they will require hemodialysis. So really, the best treatment of tumor lysis is prevention.
So, for high-risk patients like the ones that I have mentioned to you, leukemias, lymphomas, Burkitt's patients, they generally get Allopurinol before they begin treatment. This is a xanthine oxidase inhibitor. So, what will happen is that it will shunt the purines to a different pathway to reduce production of uric acid. Patients who are very high risk may be candidates for prophylactic administration of Rasburicase, which is this recombinant uricase enzyme that I mentioned. They will also reduce the uric acid level. Patients who are coming in for treatment will --- who are considered to be high risk, will receive a lot of hydration, sometimes with mannitol to increase a very brisk urine output, at least 2 to 2.5 liters per day. The faster your urine flow is, the less likely it is that uric acid crystals will form. And, at this point, if you want to use bicarbonate to alkalinize the urine, this is the point when --- where it may be helpful to try to keep the urine pH above 7.0. To be honest with you though, there is no proven benefit for alkalinization versus hydration alone.
So, in summary, the oncologic emergencies include structural problems, which we have discussed, metabolic problems, which we have discussed, and, in Part 2 of this lecture, we will talk about hematologic problems and infectious problems. I thank you for your attention.
Oncologic Emergencies: Part I video
©2010 The University of Texas MD Anderson Cancer Center
1515 Holcombe Blvd, Houston, TX 77030
1-800-392-1611 (USA) / 1-713-792-6161