Sunday, August 11, 2013

The trouble with diltiazem infusions

My general opposition to diltiazem infusions was well-known at my previous institution, and several people have asked me to write an entry explaining why. Being a native of the South, writing in the form of a three-point sermon comes only naturally, and perhaps is a fitting way to write an entry on this sunny Sunday afternoon.

"[Intravenous diltiazem] is a terrible drug. It ought to be removed from the formulary."
- Head of electrophysiology at my previous institution

As I begin, I think it is important to clarify that much of my opposition to diltiazem infusions (i.e., "dilt drips") is based on their use for acute rate control in patients with significant underlying cardiovascular disease. There are other settings (e.g., general medicine, emergency medicine), where the temporizing use of a diltiazem infusion may be an appropriate strategy, assuming the drug is properly managed from a practical standpoint. However, in patients with a history of coronary artery disease, heart failure with reduced ejection fraction (HFrEF), and other chronic cardiovascular conditions, diltiazem infusions are rarely an ideal strategy.

First, several pharmacokinetic characteristics make intravenous diltiazem a suboptimal agent to administer as a continuous infusion. Its delayed onset of action necessitates the administration of bolus doses with the initiation of a continuous infusion and with each rate increase. Unfortunately, these are often omitted, which may lead clinicians to believe an infusion is inadequate at its current rate (and often resulting in rapid dose escalation).  Additionally, its long-half life increases the risk of accumulation with prolonged use, especially with rapid dose escalation. Finally, diltiazem does not demonstrate linear pharmacokinetics, so changes in dose rarely correlate with its therapeutic effects. Minimal (if any) evidence supports the use of diltiazem infusions beyond 24 hours, giving little guidance to clinicians on its practical management [1].

Second, as a continuous infusion, there is a tendency to believe that diltiazem can be titrated like a rapidly-acting vasoactive agent; this, combined with the aforementioned pharmacokinetic characteristics, make it a commonly mismanaged agent in the acute care setting. Given its delayed onset of action, a weight-based bolus should be administered prior to the initiation of a continuous infusion. Although the package insert recommends a 0.25 mg/kg bolus, this often results in hypotension, and may explain why bolus doses are often omitted. I usually recommend starting with a 0.15 mg/kg bolus and then administering additional boluses if necessary, although some evidence suggests that pre-medicating with intravenous calcium may mitigate the risk of hypotension. As I alluded to above, when bolus doses are omitted, the tendency is to rapidly increase the rate of infusion. Although a therapeutic effect may be observed at 1-2 hours, significant accumulation occurs as the drug approaches steady state (i.e., 4-6 hours later), increasing the risk for adverse effects (e.g., precipitous changes in heart rate or blood pressure, high-degree atrioventricular block) that may persist for 10-12 hours or more. Furthermore, this delay to steady state is also why diltiazem infusions should not be titrated to effect.

Third, in many patients with significant underlying cardiovascular disease, diltiazem is rarely an ideal long-term strategy for rate control. Although the negative inotropic effects of diltiazem are probably no worse than beta blockers in an acute setting, it does not confer the same long-term benefits associated with beta blockers across several cardiovascular conditions, such as reductions in the risk of sudden cardiac death among those with prior myocardial infarction [2], or improvements in all-cause mortality in patients with HFrEF [3]. In fact, in this latter population, the use of diltiazem is actually associated with a nearly twofold increase in the risk of worsening heart failure [4].

Proponents of diltiazem infusions will often argue that its disadvantages are related to chronic administration and this should not preclude its use in an acute setting.  While I agree its acute risks are probably no worse than beta blockers, why even start down a path that does not represent a long-term solution (when reasonable alternatives exist), especially given the practical limitations associated with its use? Moreover, if diltiazem is successful in the immediate setting, will therapy be continued? If not, how quickly can a more definitive management strategy be implemented? Should a patient's length of stay be extended solely for the purposes of cross-titrating to a more appropriate long-term strategy? My point is this: if no plans for a more definitive strategy exist, and there are not plans to continue an agent in the long-term, it just seems like a step backward to even start there.

There are some scenarios where diltiazem infusions are a reasonable approach, such as a short-term infusion to maintain hemodynamic stability as a patient awaits more definitive management (e.g., ablation), or as a transition to oral therapy in a patient without structural heart disease, in whom diltiazem is a reasonable agent for long-term rate control.  However, even in these scenarios, caution should still be exerted in terms of its practical management in order to reduce the risks associated with drug accumulation.

  1. Diltiazem HCl Power for Solution [package label]. Lake Forest, IL: Hospira, Inc; 2008. 
  2. Turi ZG,Braunwald E.The use of beta-blockers after myocardial infarction. JAMA.1983 May 13;249(18):2512-6.
  3. Effect of metoprolol CR/XL in chronic heart failure: Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF). Lancet. 1999 Jun 12;353(9169):2001-7.
  4. Goldstein RE, et al; for the Adverse Experience Committee and the Multicenter Diltiazem Postinfarction Research Group. Diltiazem increases late-onset congestive heart failure in postinfarction patients with early reduction in ejection fraction. Circulation. 1991 Jan;83(1):52-60.


Allison said...

Really well written and reasoned post. Thanks for taking the time.

Vince DiGiulio said...

Thanks for this interesting perspective. I've only worked in the prehospital and ED setting where IV diltiazem is the usual mainstay of rate-control in rapid a-fib, so it's nice to hear some different thoughts on the practice. I think that at this point, at least in my settings, folks are so used to using and administering the medication (often incorrectly) that's it's going to be difficult to bring about any change in practice unless some new drug comes along that's supported by lots of Pharm-backed advertising and "education."

Anyway, my question for you is, what med(s) do you think should be first line for our patients with acute rapid a-fib? You mention B-blockers, but do you favor any in particular? Metoprolol, or maybe esmolol?

Christopher said...

I'm with Vince, and routinely use diltiazem for rate control of AF w/ RVR prehospitally. I've always been giving it "low and slow" up to a total dose of 0.25-0.35 mg/kg which seems to reduce the rate comfortably without bottoming out the patient.

I have access to metoprolol and procainamide as well.

Very interested in a better way if there is one.

Anonymous said...

Have you found any one particular strategy effective when transitioning from diltiazem IV infusion to metoprolol oral? Or diltiazem oral (after IV infusion) to metoprolol oral?

Anonymous said...

I favor beta-blockers IV boluses(metoprolol) or IV continuous infusions(esmolol) for the following reasons: 1) They both help to block the sympathetic outflow; 2) They can be effective even in small doses; 3) They may prolong diastolic filling; 4) By decreasing contractility, they are anti-ischemic.

For COPD or Asthmatic patients, I use small 'test' doses; I auscultate prior to the 'test' dose, then at intervals afterwards.

Brent N Reed said...

Vince and Chris,

Thank you for your comments. For the patient population I see most commonly (those with significant cardiovascular disease), the beta blockers tend to make the most sense because many patients require them for the treatment of various other comorbid conditions (e.g., heart failure, post-MI). I tend to favor metoprolol, although primarily for practical reasons; most clinicians are familiar with the dosing range, it is often available in automated dispensing cabinets on most units/wards, and it can be given without a continuous infusion. Intravenous administration sometimes leads to the over-treatment (the same can be said for diltiazem), but when IV is used, I find it is often under-dosed (5 mg of IV metoprolol only amounts to about 10-12.5 mg of oral metoprolol) or not given frequently enough, both of which I feel often leads people to abandon beta blockers prematurely. Anecdotally, I have not had great experience with esmolol; I have found it usually requires high rates to be effective, which usually results in a significant volume load for the patient. It is also less ideal from a practical standpoint, requiring more aggressive IV access (and carrying the risk of infiltration injury) and higher levels of care for monitoring purposes, at least at most institutions. While there are some published strategies for converting esmolol to an oral agent, I find this is more challenging in practice given wide patient inter-variability and the difficulty of successfully coordinating between prescribers, nurses, etc.

I am not adamantly opposed to the use of intermittent boluses of diltiazem, as they do not carry as many of the risks as continuous infusions (although they do sometimes result in over-treatment) and are often effective at controlling rate. However, patients often get committed to therapy after an initial bolus or two, which can sometimes make the downstream management more challenging if the patient needs to be transitioned to a different agent or the cumulative effects of the drug catch up with them after they arrive to the floor. As with the beta blockers, boluses of IV diltiazem are often under-dosed, although the risk of hypotension makes it a reasonable approach to split up the loading dose into 1-2 boluses to observe the patient's response.

For the remaining comments (anonymous):
With regard to converting from an IV diltiazem infusion to oral metoprolol; there may be some evidence-based strategies out there, but I usually just add up the amount of diltiazem they would get in 24 hours (assuming you have had the chance to observe their response at steady state) and classify that as being a low, medium, or high total daily dose and then convert them to a corresponding low, medium, or high total daily dose of metoprolol. I tend to split the metoprolol into q6h administration initially to observe whether the patient requires adjustment before consolidating to a more practical dosing regimen.

For patients with COPD or asthma, I feel like this is definitely a reasonable approach -- only a minority of patients are unable to tolerate beta blockers, although this may be more apparent at initiation than with long-term therapy. That being said, there is some evidence emerging from the pulmonary literature that beta blockers may improve outcomes in COPD, with one of the hypotheses being that beta blockers induce up-regulation of beta-2 receptors, allowing the beta-2 agonists to be more effective. Not as relevant for the acute management of atrial fibrillation, but definitely reassuring for their use for long-term management in patients who also have significant COPD.

tpainton said...

After 14 years of practicing purely hospital medicine, I can offer some anecdotal evidence. Beta blockers have given me far more problems with hypotension than Diltiazem.

In patients without CAD, oral diltiazem therapy has proven an excellent oral agent for rate control.

I have had years of success in simply coverting the oral dose to short acting Diltiazem.. That is, if rate control is achieved with 5mg/hr, then 120mg/day is working. I start with short acting divided doses ie 30mg qid, then prior to discharge if rate control is achieved, give them 120mg XL at discharge.

Bradycardia and Hypotension are far more prevelant with my use of any beta-blockers and in patients with no history of CAD, I never use it. This over thousands of patient encoutners.

Anonymous said...

What about doing it the EMCrit way?

Brent N Reed said...

Regarding EMCrit posts: Haven't seen the slow IV diltiazem load before (2.5 mg/hr to a total of 50 mg), but it seems like a reasonable approach to balance heart rate control and risk of hypotension. Most of the risks I outlined in the blog are a consequence of prolonged diltiazem infusions, so as long as the infusion is stopped after the slow load and the patient is transitioned to oral or intermittent IV therapy, I think I would be on board with it.

John Dudley said...

For more information on Breast implant and Breast implant Delhi he recommends you to visit