Check out our own Dr. Anand Swaminathan discussing this topic and more on ER Cast here!
1. Acetaminophen vs Ibuprofen. Which do you prefer for analgesia? For fever reduction?
Pain and fever are among the most common chief complaints in the ED. Acetaminophen and ibuprofen are two of the most widely consumed medications on the market today. The relevance of this debate cannot be overstated, and yet it is rarely discussed. As this question is especially frequent in the pediatric population, we will start there.
One of the most comprehensive studies in the pediatric literature is a 2004 meta-analysis that summarized the findings from 17 randomized, controlled trials comparing the two drugs in children <18 years of age. Three studies involved pain, 10 involved fever, and all 17 involved safety. They found no difference in pain relief provided by ibuprofen (4-10mg/kg) and acetaminophen (7-15mg/kg); however, ibuprofen (5-10mg/kg) was superior to acetaminophen (10-15mg/kg) as an anti-pyretic. This was true at 2 hours, and even more pronounced at 4 and 6 hours. At these later markers, 15% more children were likely to have reduced fever with ibuprofen compared to acetaminophen. When selecting for studies using only the 10mg/kg dose of ibuprofen, there was a doubling of the effect size in support of ibuprofen. As for safety, there was no evidence that one drug was less safe than the other or than placebo. The authors determined that this data was inconclusive and that more large studies would be needed to identify small differences in safety (Perrott, 2004).
In 2010 an updated meta-analysis was published. The authors noted that no such meta-analysis had been conducted in adults, and therefore also sought to examine studies in this population. The article reported data from 85 studies (54 pain, 35 fever, 66 safety). Qualitative review revealed that ibuprofen was more effective than acetaminophen for pain and fever reduction, and that the two were equally safe. From the studies that provided sufficient quantitative data, the authors calculated standardized mean differences or odds ratios then averaged these data points. Here they found that for pain, ibuprofen was superior in children and adults; meanwhile, for fever, ibuprofen was superior in children, but conclusions could not be made for adults due to insufficient data. For safety, ibuprofen was favored, but there was no statistically significant difference (Pierce, 2010).
What about combining or alternating acetaminophen and ibuprofen? Despite a lack of consensus guidelines endorsing this practice, it is commonly employed by providers and caregivers for the treatment of fever in children. This is likely heavily influenced by “fever phobia,” a concept originally coined to describe the fear that caregivers have for perceived dangerous sequelae when a child is febrile (Schmitt, 1980). Regardless of the motives, these strategies pose two questions: does the combination actually reduce fever more effectively, and, is it safe?
There are a limited number of efficacy studies, with widely differing methodologies that make systematic analysis difficult. In addition, many of these studies have design flaws such as improper administration schedules and dosing, or too short durations of follow-up. A 2013 article in the Annals of Emergency Medicine identified 4 studies that the author deemed high-quality and relevant to emergency practitioners. Three of the four found that the combination was more effective at reducing fever than either alone (Malya, 2013). However even these higher quality studies should be interpreted with caution as they also have limitations.
Safety data for combination or alternating therapy is even more limited, and the concern for safety somewhat theoretical. Dosing errors are not infrequent in the administration of acetaminophen and ibuprofen. Particularly for the former, this can easily lead to dangerous outcomes. Combining the two medications could magnify the potential for serious toxicity. Furthermore, alternating the medications can be confusing due to the recommended dosing of acetaminophen every 4 hours and ibuprofen every 6 hours in pediatric patients (Mayoral, 2000; Sarrell, 2006). One study that looked at alternating regimens over 24 hours found that 6-13% of parents exceeded the maximum number of recommended doses (Hay, 2008). Mechanisms have been suggested by which the two drugs could act synergistically to cause renal tubular injury; however, acetaminophen and ibuprofen have different pathways of metabolism, and adverse effects in patients taking both have only been described in rare case reports (Mayoral, 2000; Smith, 2012).
As a final piece in this question, is it acceptable to prescribe ibuprofen for pain relief in patients with fractures? While combination medications containing an opiate will often be necessary for patients with fractures, ibuprofen has anti-inflammatory properties that other medications lack, and its use may reduce the need for opiates. But your orthopedic surgery consult may recommend avoiding NSAIDs in fractures because they could suppress healing. What is the evidence?
The theory behind this stems from the fact that as cyclooxygenase (COX) inhibitors NSAIDs suppress production of prostaglandins, which are important mediators in bone repair. Theoretically this makes sense, but studies to support this have only been conducted in animal models. A number of human studies have suggested that NSAID use in patients with long bone fractures is associated with nonunion; however, these are largely uncontrolled retrospective studies that fail to demonstrate causality. The authors of a recent review in the orthopedic literature state, “We found no robust evidence to attest to a significant and appreciable patient detriment resulting from the short-term use of NSAIDs following a fracture” (Kurmis, 2012).
2. PPIs vs H2 Blockers. Which is your first line choice for gastritis/GERD?
Gastritis and gastroesophageal reflux disease (GERD) are pervasive medical problems. Treatment of these diseases was revolutionized in 1979 by the introduction of the first H2 receptor antagonist (H2RA), cimetidine, then again in 1989 by the introduction of the first proton pump inhibitor (PPI), omeprazole (Sachs, 2010). These two drug classes now form the cornerstone of treatment of gastritis and GERD. But which one works better? More specifically, which will provide symptomatic relief more quickly in the ED setting, and which one should we prescribe patients upon discharge?
Answering these questions requires a brief review of the pharmacodynamics of these drugs. PPIs suppress acid secretion by binding to the H+/K+-ATPase in the parietal cells of the stomach. There are a few important aspects to this process that affect the onset and duration of action of PPIs. First, PPIs are prodrugs. Before they are able to bind to the proton pump they must diffuse into the parietal cell and be protonated to form the active drug. As a result, PPIs have a somewhat delayed onset of action. Second, binding to the proton pump is irreversible. Therefore the duration of effect is not related to the plasma drug concentration but rather the turnover rate of the proton pump. So, despite a short half-life, PPIs can be effective for up to 3-5 days. Finally, because PPIs inhibit the last step in acid production, their effect is independent of any downstream factors.
H2RAs work as competitive inhibitors at the histamine receptor, preventing histamine from binding and stimulating acid production. Onset of action is rapid (<1hr), with peak serum concentrations reached in 1-3hrs. Unlike the PPIs, binding is reversible, and duration of action is much shorter, approximately 12hrs. Because H2RAs do not block the final secretory step, some acid is still produced. H2RAs are less potent than PPIs, reducing daily acid production by about 70%, as compared to 80-95% for PPIs. One other important thing to be aware of concerning H2RAs is that they are known to demonstrate tachyphylaxis. Tolerance may be exhibited within 3 days of use (Wallace in Goodman & Gilman, 2011).
Gastritis describes a spectrum of pathologies ranging from known peptic ulcer disease to functional (endoscopy-negative) dyspepsia. Regardless of the exact entity, most disease is attributable to H. pylori, aspirin/NSAID use, or alcohol. Treatment for H. pylori is well-established and includes PPIs, as these agents have been shown to heal ulcers faster than H2RAs and also to contribute to the eradication of H. pylori. For patients on chronic NSAID therapy, PPIs have also been shown to be more effective than H2RAs in healing ulcers (Boparai, 2008). One study demonstrated 8-week healing rates of 80% for 20mg omeprazole daily versus 63% for 150mg ranitidine twice daily (Yeomans, 1998). A similar trial substituting esomeprazole showed healing rates of 88% and 74%, respectively (Goldstein, 2005).
In 2005 the Agency for Healthcare Research and Quality wrote a Comparative Effectiveness Review for the management of GERD. They identified 3 well-conducted meta-analyses of PPIs and H2RAs, and concluded that PPIs were superior to ranitidine for symptom resolution at 4wks (Ip, 2005). One of these meta-analyses examined 11 randomized controlled trials (1575 patients total) comparing a PPI to ranitidine. At 8 weeks each of the 4 PPIs included had a higher rate of healing than ranitidine. For omeprazole, healing was 1.6 times more likely than for ranitidine (Caro, 2001). In 2011 the AHRQ updated their review. They analyzed 39 additional primary studies and did not alter their previous conclusions (Ip, 2011). One of the largest of these studies (1902 patients) found that esomeprazole 20mg once daily “significantly improved all symptoms” in 80% of patients, compared to 47% in those taking ranitidine 150mg twice daily (Hansen, 2006). A recent Cochrane Review based on 7 trials similarly found that PPIs were significantly more effective than H2RAs for remission of symptoms (RR 0.66) (van Pinxteren, 2010).
Current practice guidelines in the gastroenterology literature advocate that the therapy of choice for GERD is an 8-week course of PPIs, initiated once daily before the first meal of the day. In incomplete responders, dosing may be increased to twice daily, or, in the absence of erosive disease, H2RAs may be substituted or added at bedtime for nocturnal breakthrough symptoms (Katz, 2013).
Unfortunately, there is a lack of comparative studies or guidelines for treating gastritis or GERD in the ED setting; however, based on other studies and the known pharmacologic properties of these drugs such as onset, duration, and tolerance, it is fair to say that PPIs are the preferable first-line treatment for all patients long-term, but in the acute setting H2RAs may provide symptomatic relief more quickly.
The last question to ask is whether there are any adverse effects that may be pertinent when choosing an agent. Both drugs are very safe with few side effects. PPIs are metabolized by the liver whereas H2RAs are excreted by the kidneys, therefore care should be taken when prescribing these medications for patients with hepatic or renal insufficiency, respectively. Additionally, PPIs are metabolized by cytochrome P450 enzymes. As a result, they have the potential to interfere with elimination of other drugs such as warfarin and clopidogrel that are cleared by the same pathway.
It has been suggested that chronic PPI use may be associated with increased risk of fractures and certain infections, such as pneumonia and clostridium difficile. To address these concerns, the most recent consensus guidelines in the American Journal of Gastroenterology state the following:
- PPIs may be prescribed for patients with osteoporosis and should not be a concern unless a patient has other risk factors for fracture;
- PPI use can be a risk factor for clostridium difficile and should be used with caution in patients at risk;
- Short-term PPI use may be associated with increased risk of community-acquired pneumonia, but this is not seen with long-term use;
- PPIs can be continued in patients taking clopidogrel (Katz, 2013)
3. Meclizine vs Benzodiazepines. Which do you prescribe for vertigo?
For peripheral vertigo (labyrinthitis, vestibular neuritis and benign paroxysmal positional vertigo (BPPV)), vestibular suppressants, and to a lesser extent antiemetics, comprise the arsenal of pharmacologic treatment. The use of these drugs most applies to labyrinthitis and vestibular neuritis as BPPV is short lived and can often be corrected by positioning.
Vestibular suppressants include 3 major classes (Hain, 2003):
- Antihistamines (meclizine, diphenhydramine, dimenhydrinate)
- Benzodiazepines (diazepam, lorazepam)
- Anticholinergics (scopolamine)
There have been very few studies examining the efficacy of these drugs, and even fewer head-to-head trials. Moreover, the majority of these limited studies are decades old. In 1972, Cohen and deJong demonstrated that meclizine was superior to placebo in reducing vertigo symptoms and the frequency and severity of attacks; however, they used a sample size of only 31. Conversely, in 1980 McClure and Willet found that benzodiazepines were not superior to placebo. Again the sample size was small, with 25 patients randomized to diazepam, lorazepam, or placebo.
In a somewhat more recent study from the EM literature, dimenhydrinate was compared to IV lorazepam for the treatment of vertigo in the ED. They found that at 2 hours, dimenydrinate was more effective in relieving symptoms and less sedating than lorazepam. This study had a sample size of 74 (Marill, 2000).
Whether based on the few studies or on anecdotal evidence, most sources seem to have a slight preference for antihistamines over benzodiazepines. But further review of the literature brings to light a more important question: whether there is a place for medication at all in the treatment of vertigo.
In 2008, the American Academy of Neurology and the American Academy of Otolaryngology both published evidence-based practice guidelines for the treatment of BPPV. The neurology recommendations state: “There is no evidence to support a recommendation of any medication in the routine treatment for BPPV” (Fife, 2008).
The ENT guidelines state, “Vestibular suppressant medications are not recommended for the treatment of BPPV, other than for the short-term management of vegetative symptoms such as nausea or vomiting in a severely symptomatic patient.” The authors justify their recommendation based on the lack of evidence for these medications, but also the potential harm associated with them. Side effects of the vestibular suppressants include drowsiness, cognitive impairment, gastrointestinal motility disorders, urinary retention, dry mouth, and visual disturbances. Some of these medications are alone significant risk factors for falls and other accidents, but become even more dangerous in patients already experiencing dizziness (Bhattacharyya, 2008).
Both academies advocate the use of particle repositioning maneuvers (PRMs), such as the Epley and Sermont, as opposed to medications. This is supported by a number of studies, one of which showed improvement rates of 79-93% for medication plus PRMs, versus 31% for medication alone (Itaya, 1997).
By definition, BPPV consists of brief episodes of vertigo triggered by movement; therefore it makes sense that medications would not be a useful management strategy. They will not prevent episodes and they should not be needed to abort very short-lived symptoms. Bhattacharyya points out that although some studies have shown improvement after vestibular suppressants, patients are followed for a duration in which symptoms should be expected to resolve spontaneously.
Many sources also point out another downside to vestibular suppressants: they delay central compensation. Compensation is an adaptive response to any vestibular stimulus, whether related to normal motion or to disease. This process is key to the recovery from vestibular diseases in the sub-acute phase. All the vestibular suppressants are thought to slow compensation, although the support for this claim comes primarily from animal studies. Nonetheless, most authors consider this further evidence for use of these medications only in the acute period and not after the initial 48 hours (Hain, 2003).
Finally, it is important to note that in the ED setting it is often difficult to diagnose BPPV with certainty. A 2009 article in the EM literature highlights the frequency with which providers misdiagnose and mistreat BPPV and acute peripheral vertigo (APV), an umbrella term describing labyrinthitis and vestibular neuritis/neuronitis. The authors caution that while BPV is more prevalent in the general population, APV is actually the more common disorder among patients presenting to EDs. They go on to explain that the distinction is important because BPPV and APV have different treatments. Despite the pervasive use of meclizine to treat BPV, it is not indicated for this diagnosis. Meanwhile, APV should be treated with steroids (Newman-Toker, 2009).
In summary, for the undifferentiated patient with symptoms of vertigo, vestibular suppressants can be used for the acute management of severe symptoms, regardless of the patient’s diagnosis, and meclizine is generally the preferred drug. However, it is important to attempt to make the correct diagnosis and guide further treatment accordingly, whether it is with PRMs, steroids, or another strategy.
4. Calcium Channel Blockers vs Beta Blockers. Which is your first line choice for rate control in atrial fibrillation?
Atrial fibrillation (AF) is the most common dysrhythmia seen in the ED. For years, management of this disease has been rife with controversy, such as whether to anticoagulate, whether to rate control versus rhythm control, and which agents to use in each of these management strategies. This discussion, however, is limited to the most common medications used for rate control: non-dihydropyridine calcium channel blockers (CCBs) and beta blockers (BBs). Please see our previous topic on recent-onset atrial fibrillation for discussion of the other controversies: http://emlyceum.com/2011/08/29/acute-onset-atrial-fibrillation-answers/
CCBs block voltage-gated calcium channels in the heart and blood vessels. BBs competitively inhibit catecholamine binding at beta-1 and beta-2 receptors in the heart and vascular smooth muscle. Both slow conduction through the AV node and lengthen its refractory period during high rates of conduction (Demircan, 2005).
Digoxin was previously the mainstay of treatment for stable, rapid AF until the introduction and FDA approval of IV diltiazem for AF in 1992 (Schreck, 1997). A recent survey study investigated prescribing preferences in new-onset AF among nearly 2000 emergency physicians in multiple English-speaking countries. They found that in the U.S. and Canada, IV diltiazem was the most commonly preferred drug for rate control (95% and 65% of respondents, respectively). In the U.K. and Australasia, IV metoprolol was most commonly preferred (68% and 66% of respondents, respectively) (Rogenstein, 2012).
Possibly the most famous study conducted on AF is the AFFIRM trial. One arm of the analysis focused on approaches to rate control. The investigators found that 59% of patients randomized to a BB alone achieved rate control, versus 38% of those randomized to a CCB alone. They also found that more patients were switched from a CCB to a BB than vice versa (Olshansky, 2004). Of note, alteration in regimen was at the discretion of the treating cardiologist. Additionally, average follow-up in the study was 3.5 years, making the results much less applicable to the ED setting. While this study is often cited as providing the crux of available data on CCBs versus BBs for AF, there actually have been some additional small studies, including a number conducted in ED patients (!).
A Turkish study of 40 ED patients compared the efficacy of IV diltiazem versus IV metoprolol. At 20 minutes, rate control was achieved in 90% of patients randomized to diltiazem versus 80% of those randomized to metoprolol (Demircan, 2005). Another study of 52 ED patients found that diltiazem was more likely to achieve rate control at 30 minutes (Fromm, 2011). A recent, larger study compared not only efficacy but also the safety of CCBs and BBs. The primary outcome was proportion of patients requiring hospital admission: 31% for the CCB group versus 27% for the BB group. There were no significant differences in the secondary outcomes of ED length of stay, adverse events, 7- and 30-day ED revisits, stroke, and death. The authors concluded that while diltiazem has been observed to reduce heart rates more quickly than metoprolol, the two drugs are associated with similar overall outcomes (Scheuermeyer, 2013).
Are there any practice guidelines for rate control of rapid AF? ACEP has not published any, but the AHA has. They state: “In the absence of preexcitation, intravenous administration of beta blockers (esmolol, metoprolol, or propranolol) or nondihydropyridine calcium channel antagonists (verapamil, diltiazem) is recommended to slow the ventricular response to AF in the acute setting, exercising caution in patients with hypotension or heart failure” (Fuster, 2006).
The mention of heart failure is notable. All of the above drugs should be avoided or used with caution in patients with decompensated heart failure; however, in patients with compensated heart failure or left ventricular dysfunction, BBs are the drug of choice. The same is true for patients with acute coronary syndromes or thyrotoxicosis. Conversely, CCBs are preferred in patients with obstructive pulmonary disease, which is a relative contraindication to BBs (Oishi, 2013).