Emergent Aspirin Use in Cardiovascular Disease in the Emergency Department: Oral Dosing versus Rectal Suppositories

More than 50 million people in the United States regularly take aspirin, primarily for cardiovascular disease prevention—translating into over 10 billion tablets taken yearly.1,2 In the emergency department (ED), aspirin is used emergently in two settings: 1) for the secondary prevention of noncardioembolic ischemic stroke and transient ischemic attack (TIA), aspirin at a dose of 50 to 325 mg/daily within 48 hours of an event is currently recommended3; and 2) for the management of acute coronary syndrome (ACS), treatment with 162 to 325 mg of a chewed non–enteric-coated aspirin is indicated immediately.4 Chewing is important because inhibition of platelet thromboxane A2 (TXA2) formation is detected within 3 minutes of ingestion when chewed and 7.5 minutes from ingestion when swallowed without chewing.5 At least 162 mg of aspirin chewed is needed for ACS because maximal inhibition of TXA2 occurs at 15 minutes in all subjects using this dose, compared to only some subjects using the 81-mg dose,6 albeit all subjects at both doses have maximal inhibition within 30 minutes. Enteric-coated aspirin may delay platelet TXA2 inhibition by hours and should not be used in ACS unless chewed7; chewing a 325-mg enteric-coated tablet has been shown to almost completely suppress TXA2 within 15 minutes.8 The primary mechanism of action of aspirin, also known as acetylsalicylic acid, is to bind irreversibly to the cyclooxygenase I (COX-I) enzyme through acetylation of the amino acid serine at position 529.9 COX-I enzymes are constitutively expressed in most normal cells and tissues including the gastric epithelial mucosa, but not the platelet. The platelet, being anucleate, cannot regenerate COX-I, resulting in the inability to synthesize TXA2 for the life of the platelet. TXA2, a potent platelet activator, is not synthesized and regeneration of new platelets occurs at about 10% per day, resulting in at least partial reduction in platelet aggregation for 10 days. By 72 hours, however, some COX-I activity is present due to new platelets, requiring intermittent aspirin dosing. COX-I enzymes in gastric mucosa maintain the production of cytoprotective prostaglandins. Low doses of aspirin at 30 mg/day result in 50% inhibition of gastric COX-I with maximal inhibition occurring at 1300 mg/day.10 Gastric mucosal cells regenerate COX-I enzymes within hours, preventing irreversible loss of COX-I activity in these cells. In addition, it has been suggested that aspirin at 1200 mg/day has a higher risk of gastrointestinal (GI) bleeding than aspirin at 300 mg/day.11 Multiple other studies have shown that aspirin doses ranging from 30 to 325 mg show a differential increase in GI bleeding as aspirin dose is increased.12–14 The differential GI bleeding with dose is consistent with the differential suppression of COX-I with dose. This does not include the slight increased risk of hemorrhagic stroke of 0.2 events per 1,000 patient-years of aspirin therapy.15 Therefore, the therapeutic goal has been to find the lowest dose of aspirin that would maximize therapeutic benefit and minimize bleeding. In contrast to aspirin, the other nonsteroidal anti-inflammatory medications reversibly inhibit platelet COX-I, with the transient inhibition related to the pharmacokinetics of the drugs.16 Only indomethacin reversibly inhibits platelet COX-I TXA2 synthesis greater than 95%, a degree similar to that of aspirin.16 Salicylic acid, the product of aspirin metabolism by tissue and plasma esterases, is not an inhibitor of COX-I,16,17 although it has been mistaken to be an inhibitor. In this issue of Academic Emergency Medicine, we have a study by Maalouf et al.18 comparing salicylic acid levels in normal subjects after rectal vs. oral dosing. Theoretically, the counterfactual study design is the ideal. Turning the study by Maalouf et al. into a hypothetical counterfactual example, all 24 subjects would undergo treatment with oral aspirin therapy with salicylic acid levels drawn at baseline, 30, 60, and 90 minutes. Then, going back in time, the same subjects take aspirin suppositories over the same (previous) time frame and with the same salicylic acid levels drawn. A comparison would be made and the “only difference” between treatments would be the intervention. Because this perfect design is counter-to-fact, albeit a thought experiment that helps one conceptualize theoretical weaknesses in other designs, a compromise design using patients as their own control5–7,16,17 best approximates the ideal as in the current study.18 This crossover design eliminates age, gender,19 and pharmacogenetics20 as confounders, an attractive feature in drug studies with a reduced number of participants where pharmacogenetic variation would be beyond what any small randomized study could control for.5–7,16,17 In other words, variation between subjects is eliminated relative to randomized trials. The main criticism of the crossover design is whether the first study period affects the second study period (as would occur in cancer chemotherapy, where a randomized design would then be the criterion standard). In this study,18 there is no biologic reason why a 1-week washout period after single dose oral 162 mg aspirin would not be adequate before crossover. Antecedent oral aspirin therapy should not affect the results of single administration rectal therapy. Ideally, the authors could have improved on design by randomizing who would first get oral versus rectal aspirin therapy prior to crossover. Again, however, there is no obvious biologic reason why oral/rectal dosing order would have made a difference in the results of this study.18 The study by Maalouf et al. comparing single-dose oral 162-mg aspirin versus single-dose 600-mg aspirin suppositories clearly reveals that both routes of administration successfully deliver an aspirin metabolite, salicylic acid, to the blood stream within 90 minutes.18 The change from the absence of baseline salicylic acid to the presence of blood salicylic acid levels signifies GI absorption of aspirin from the dose administered. Salicylic acid is metabolized from acetylsalicylic acid (aspirin) by esterases present in the GI tract, liver, and plasma.21 Normally, systemic bioavailability of oral aspirin is about 50%21 and can approach zero, particularly with sustained release or enteric aspirin preparations.22,23 Despite reduced or absent blood levels of orally administered aspirin, salicylic acid levels are always present indicating drug absorption in spite of significant or even complete presystemic metabolism by the oral route.21–23 In addition, studies have clearly shown greater than 95% inhibition of platelet TXA2 production, even with zero aspirin levels (salicylic acid levels are present), indicating that in some instances all COX-I inhibition in the platelet occurs in the portal presystemic system.21–23 Aspirin given by rectal suppository is absorbed by the inferior and middle hemorrhoidal veins that eventually drain into the inferior vena cava, avoiding presystemic metabolism through the liver.24 This will result in early aspirin levels followed a short time later by salicylic acid levels as metabolism in the plasma and eventually passage through the liver occurs. Based on Figure 1 in the article,18 this implies aspirin presence before 30 minutes in some subjects and before 60 minutes in most subjects and suggests relatively rapid inhibition of platelet TXA2 because early inhibition can be observed simultaneous with aspirin levels at 3 minutes after oral dosing.5 Current study observations are consistent with one investigation that compared 200-mg aspirin suppositories with 200-mg oral aspirin in normal volunteers.25 Actual aspirin levels peaked at 30 minutes with rectal dosing (and may have peaked earlier if levels had been drawn earlier) and 60 minutes with oral dosing, implying faster systemic absorption with suppositories.25 Suppression of platelet aggregation was similar with both routes of administration and occurred at 30 minutes with effects present up to 24 hours.25 Two subjects in the study by Maalouf et al. appear to have had very delayed rectal aspirin absorption as manifest by very low (one) or absent (one) salicylic acid levels up to 90 minutes.18 However, it is still possible that early platelet TXA2 inhibition occurred if some aspirin was absorbed. With the oral route, two subjects did not have any detectable salicylic acid levels at 90 minutes, and they were different from the two with low/none rectal absorption. It is unclear why this occurred, short of not taking the medication, delayed absorption from enteric coated aspirin not chewed, or no aspirin absorption. Without further studies and longer follow-up of aspirin levels, salicylic acid levels, and functional studies of TXA2 inhibition, it is unclear whether there are actual aspirin failures secondary to lack of GI absorption of rectal suppositories or oral aspirin. What does this study mean? Clearly, Maalouf et al. have demonstrated successful aspirin absorption from suppositories as measured indirectly by salicylic acid levels, consistent with one other study that measured aspirin levels directly.25 Therefore, in stroke patients unable to take aspirin therapy, an initial rectal suppository can be given until some method of feeding can be worked out. In the ACS patient who is vomiting or cannot take oral medication, the rectal suppository can be used with inhibition of platelet aggregation at 30 minutes25 and possibly even earlier when studies are done to determine speed of effect. The authors should be congratulated on a clever investigation and study design that validates what some have already been doing (without evidence) and reminds others of what options are available when immediate oral therapy with aspirin is not possible and aspirin therapy is desired.