 |
 |
 |
 |
 |
 |
 |
|
|
|
|||||||||
|
|||||||||||
Assistant Professor of Pharmacy Practice School of Pharmacy and Health Professions Creighton University 2500 California Plaza Omaha, Nebraska 68178 fax 402/280-1268 chock{at}creighton.edu
Assistant Professor of Pharmacy Practice School of Pharmacy and Health Professions Creighton University
Associate Professor of Pharmacy Practice School of Pharmacy and Health Professions Creighton University
Published Online, October 20, 2009. www.theannals.com, DOI 10.1345/aph.1L413a
The patient was classified as "relatively stable" on warfarin. The American College of Chest Physicians (ACCP) recommends that stable international normalized ratios (INRs) be monitored every 4 weeks, with those not stable to be followed more frequently.2 During the 6 months documented, this was not a stable record due to variable and multiple out-of-range INRs.
The patient's use of tobacco, alcohol, and over-the-counter products were ruled out as contributing factors. However, nutritional information was not provided regarding potentially interacting foods or beverages such as grapefruit, cranberries, or vitamin K–containing nutrients.3
This patient had low hemoglobin and hematocrit readings 1 year prior to the event, with a further decrease in both indices 1 month prior to his hemorrhage. Whether the decreases were due to progression of iron-deficiency anemia, anemia of the elderly, or to an underlying chronic bleed should be considered, especially in light of lower gastrointestinal bleeding documented 2 days prior to admission. A chronic gastrointestinal bleed seems possible, given the patient's history of gastroesophageal reflux disease, concomitant use of celecoxib, chronically low hemoglobin and hematocrit levels, and rectal bleeding prior to admission. Laboratory results indicated low serum protein and albumin by the time the INR was 15. These had been normal 1 month earlier; whether this signaled an undiagnosed illness is also unknown.
Carroll and Carroll1 outlined previous reports of potential interactions between warfarin and influenza vaccination, but little critical evaluation was provided. Our brief review of these reports found no INR elevations as high as that observed in their case. Among the case reports cited by Carroll and Carroll, we found a consistent failure of the original reports to rule out confounding factors. In addition, a 2005 systematic review on warfarin interactions identified only 3 warfarin–influenza vaccine interaction studies that met their quality criteria: 2 reported no significant interaction, and 1 reported inhibition of warfarin effect but only in an elderly subgroup, based on post hoc analysis. The review specifically noted a need for analysis based on large administrative databases examining this and other possible drug interactions with warfarin.4 The recent large database study listed by Carroll and Carroll (5167 subjects receiving warfarin and influenza and/or other vaccines),5 as well as 2 other more recent cited reports with relatively rigorous study designs,6,7 all reported no significant effect of influenza vaccine on INRs, nor were major bleeding episodes reported.
Past influenza vaccination in this patient was not associated with this reaction, giving negative answers on questions 5 and 6 of the Horn Drug Interaction Probability Scale (DIPS).8 There are several other alternative contributing causes for this event, so question 7 of the DIPS is also negative. Without further screening to rule out these confounding factors, it seems premature to isolate influenza vaccination as the culprit. This report may influence influenza vaccination decisions for this high-risk population. It is important to rule out potential changes in the patient's nutritional status and a possible undiagnosed underlying concomitant illness as alternative contributing factors.
Footnotes
Financial disclosure: None reported
References
| ||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
 |
 |
 |
 |
