INSIGHTS FROM: SAMUEL J. KLEMPNER, MD, THE ANGELES CLINIC AND RESEARCH INSTITUTE; CEDAR SINAI MEDICAL CENTER; ALICE BEERS, RN, BSN, OCN, MEDSTAR HEALTH SYSTEM
Thursday, September 07, 2017
Samuel J. Klempner, MD: To study the prevalence of overdose is quite a difficult question, for the same reasons that toxicity is somewhat difficult to assess outside of a clinical trial. Overdose is largely related to dosing calculation errors on the part of staff and clinicians, as well as technical and mechanical failures of the pump. So, the pump may bolus in 5-FU when it’s supposed to be spread over a prolonged infusion. The rate may be accelerated. The pump itself may be programed correctly, and then malfunction.
Studying these types of questions is quite difficult because, again, they are largely underreported. And if you look at the rate of overdose, most of them are attributable to pump mechanical errors as well as dosing calculations, not so much underappreciation of clearance mechanisms. So, most clinicians and staff reliably adjust for renal and hepatic function, when needed, with capecitabine and 5-FU.
Overall, the incidence is low. It’s quite rare, and the large majority are attributive to mechanical errors and dosing miscalculations. But to give you a specific percentage of all chemotherapy patients who undergo a potential overdose, I’m not aware of specific datasets that answer that question.
The recognition of overdose is tricky, and it’s improved by hospital-wide and healthcare-wide systems such as computerized order entry, which minimizes some human error. Certainly, we rely a lot on the communication between all of the office staff—everyone ranging from nursing staff, who are programming the pumps; pharmacy staff, who are checking; and our clinicians, who are writing the orders and doing the dose calculations, in conjunction with nursing and pharmacy. So, there’s multiple steps to prevent potential overdose, and these certainly do improve the rates of overdose. Recognition of overdose when it has happened is important. This includes when patients come back in for assessment, whether it’s to have their pump removed or toxicity assessments on an oral drug—asking them how many pills they have left, looking at the pump physically and seeing how much drug is left, if it’s all been given, or if all of it has been given too quickly. Those lead to recognition of accidental overdose in many cases.
Intentional overdose is generally only appreciated when patients present with the toxicity, because that is often done when patients are trying to do self-harm. That’s a very, very rare minority of chemotherapy patients, and it certainly can be seen with oral agents a little more frequently than IV agents, because it’s harder for patients to manipulate their own infusion pumps, although certainly not impossible.
So, it largely comes down to communication and multiple assessment among providers to recognize an overdose. Certainly, early toxicity and side effects are clues to overdose. They are outside of what you would normally expect. An intentional overdose can be recognized by either toxicity, family members may call and recognize this within their own loved ones, or emergency room providers may be the first encounter for an intentional overdose.
Alice Beers, RN, BSN, OCN: A chemotherapy overdose is generally detected relatively sooner than the patient’s treatment cycle. So, if you take the scenario of a patient who is sent home from an infusion clinic with a 5-FU chemotherapy pump, we tell the patient, “This is going to be an infusion that will last you 46 hours or 96 hours.” Those are the typical infusion times. And so, if a patient is calling the following morning and saying, “My pump is alarming and display says the bag is empty,” or they make a phone call in the middle of the night to the on-call physician with the same scenario—pump is alarming, the bag is empty—you know that that infusion is not supposed to be completed for another 3 days. You will certainly have the occasion where the patient is given a bolus dose in less than the intended time. That tends to be a less common issue.
In our practice, we have a very hard and fast rule that we will use mechanized chemotherapy administration pumps. There are some devices that are available that don’t have mechanics involved in them, and we will not use those in our chemotherapy patients. We don’t feel that there is a good enough index of safety and reliability. We’ve had cases where—through the insurance, they will mandate that that’s the device it suspends to the patient—we’ve had call backs, thankfully with no ill effects. However, we’ve had patients that have called and said, “You told me that this was going to finish at 2 and now it’s 10 o’clock in the morning and it’s empty,” or “You told me it was going to finish at 2 and now it’s 6 o’clock in the evening and it’s not empty.”
So, for us, it’s critically important that we have as high of a device reliability as possible. You will certainly, unfortunately, see a situation where the pump has been programmed incorrectly and those are situations in which you may have an overdose from infusional 5-FU.
With capecitabine, the scenario can be different. We give patients very thorough instructions that they are to take 3 pills—4 pills depending upon their size—at an interval of twice a day. And we tell them very emphatically, “If you forget a dose, it’s really important that you do not double up, that you don’t take 8 tablets at a time.” No matter how many times you feel that you’re being clear to people, there are always going to be situations where it’s an understanding barrier, a literacy barrier, or a language barrier where there might be a misunderstanding and someone is taking more than they should in terms of their capecitabine dose. You could have a patient who calls you and says, “You told me that this was going to last for 14 days, but it’s day 8 and I’m out, I need a refill.” And that’s clearly a red flag that they are not taking the prescription as intended.
Samuel J. Klempner, MD: The management of overdose starts with immediate sensation of the drug, if it hasn’t already been completely given. So, if this is capecitabine, the oral drug, then you may recognize this in the form of early toxicity, in which case all subsequent doses should be immediately stopped. In the setting of intravenous 5-FU, if there’s any drug left, it should be stopped immediately. And this is a relatively obvious way of managing overdose. But unfortunately, the majority of overdoses are recognized when the drug has either completed too early, so there’s nothing left to stop, or it was given in a shorter duration of time than was intended, so any subsequent administration would be stopped.
Following cessation or drug discontinuation, early intervention is the best way to prevent development of life-threatening toxicities. Because initially, if the overdose is recognized immediately, the patients will largely be asymptomatic. And this explains why the label of uridine triacetate, for example, includes patients who are asymptomatic from their overdose, because you expect that they may go on to get the toxicity, if you don’t intervene. So, early intervention after drug discontinuation involves aggressive supportive care—IV hydration; potential admission to the hospital, which is generally the safest way for frequent monitoring and assessment; as well as the recognition of the potential side effects that may come, which are mainly the GI, the cytopenias, the nausea/vomiting, neural and cardiac toxicities.
We generally manage these quite conservatively, so if there is an overdose that’s recognized, whether it be from an oral agent or an infusional 5-FU, we err on the side of caution and treat these patients aggressively. So, early intervention with the now approved uridine triacetate is warranted in the majority of overdosed patients, unless you have an extreme level of confidence that the overdose was not clinically significant, which is a very fast minority. Most of these are warranting intervention immediately.
Alice Beers, RN, BSN, OCN: In the event that a patient may experience a chemotherapy overdose, we have several steps that we would go through. Obviously, if we have an issue of pump malfunction, we are looking at the displays on the pump, we’re recording what the intended dose was versus what is reading on the pump, and calculating and documenting what was the differential and the intended dose time versus the actual time delivered. We have had situations in the past where we’re concerned about whether there was a problem in the mixture of the infusional 5-FU and we will return it to the pharmacy that dispensed it. There are mechanisms that those drugs can be sent out for testing to verify whether or not there was an unintentional mistake made where the drug was sent out to us. We certainly have situations where patients will report to us that they have had an issue where the drug was supposed to be infused over a certain period of time and they’re reporting back that the bag is empty.
In a situation where a patient has experienced a chemotherapy overdose, if you’re not recognizing symptoms in a timely manner, by the time the patient presents for care—meaning that they’re needing to be admitted to the hospital—the spiral of their decline can be very quick. You have a patient who’s admitted to the hospital with dropping blood counts, and if you’ve worked in oncology, you know that the concern is that they’re going to develop sepsis. And depending upon the age of your patient, depending upon the comorbidities, that can be a cycle that progresses very rapidly to a patient who needs intensive care treatment, who ends up intubated, needs extensive life support, and pressors, in order to be treated and hopefully pulled out of the episode.
I will talk to you about situations where we know we have to discontinue a drug. We have had cases where patients will report unexpected toxicity, even when the infusion is not completed—although I think in some practice settings, there could be a management situation where they would decide to bring the patient in for some IV fluids - bring them in for some supportive management. Because we have had issues in our practice where we have seen cases of early-onset 5-FU toxicity, we don’t do that. We would rather bring the patient in, discontinue the therapy, and then take supportive measures from there. In the appropriate patient, if the toxicity is severe enough, we will be initiating Vistogard. We do have Vistogard on our formulary. It’s in our pharmacy. It’s known to be available throughout other sites throughout our health system for the appropriate patients. We really have learned that we’re not going to take a watch-and-wait approach. There is no benefit to the patient for taking that. You can administer the antidote and prevent the patient from becoming critically ill.
In the event of chemotherapy overdose for a patient who has received 5-fluorouracil or capecitabine, there really isn’t a benefit to watching the patient. If you know that the patient has experienced an overdose, really you need to administer Vistogard as soon as you’ve identified what has happened. There are cases where patients, if they’re not treated, certainly could die from the side effects of an overdose or, in a lesser circumstance, they could experience a range of toxicities that would prevent them from being able to restart chemotherapy in the timeframe that they should. There have been cases reported in the literature where people experience an overdose, and they’re so ill that it’s months before they can resume chemotherapy. And that is such a detriment to the patient. They’re getting chemotherapy because they’re trying to fight their cancer, and hopefully fight for their lives. If you’re not treating an overdose situation in a timely manner, you can take that opportunity away from them.