Effect of Oral Corticosteroids on Chronic Warfarin Therapy

Kathleen A. Hazlewood; Susan E. Fugate, PharmD BCPS CACP; Donald L. Harrison, PhD


The Annals of Pharmacotherapy. 2006;40(12):2101-2106. 

In This Article


A review of 387 medical records was conducted, of which 32 patient encounters met the predetermined inclusion and exclusion criteria. Absence of concomitant warfarin and corticosteroid therapy was the primary reason for exclusion. Of the patients using concurrent warfarin and corticosteroids, the vast majority were excluded due to the confounding factor of concurrent antibiotic use. Overall, 24 patients were included; multiple encounters meeting study criteria occurred in 4 patients, resulting in 32 patient encounters. Characteristics of the 32 patient encounters are shown in Table 1 .

The oral corticosteroids prescribed included prednisone for 50% of the patients and methylprednisolone for the other 50% of the patients. All patients were given tapered dosing schedules, with the exception of one patient, who received prednisone 20 mg daily. Specific dosing for the prednisone taper was not documented for most patients; however, all patients prescribed methylprednisolone were receiving a 6 day dose pack. Patients were treated with corticosteroids for an average of 9 days (range 6-30 days).

The mean ± SD pre-INR value was 2.33 ± 0.37, which was measured at an average of 11.9 ± 7.1 days prior to corticosteroid initiation. The mean post-INR value was 3.57 ± 1.07 and was assessed at 6.7 ± 3.3 days following the first dose of corticosteroid. The results of the repeated-measures t-test revealed that there was a significant difference between the INR values pre- and post-initiation of corticosteroid therapy (p < 0.001). The mean difference between pre- and post-INR values was 1.24 (95% CI 0.86 to 1.62). Therefore, the null hypothesis that there is no difference between the pre- and post-INR values was rejected; a significant difference does exist between the 2 measures. The individual pre- and post-INR values are presented in Figure 1. Of note, 62.5% of patient encounters had post-INR results above their prescribed INR goal range.

Figure 1.

Individual patient encounters pre- and post-INR results. INR = international normalized ratio.

The results of the independent samples t-tests using the mean difference in pre- and post-INR values as the dependent variable and gender and ethnic groups as the independent variables revealed no statistically significant differences in INR values based on gender and ethnicity (p = 0.800). The mean differences in INR values for male versus female and white versus African American patients were not significantly different. The results of the χ2 analyses revealed no statistically significant association between the frequency of warfarin dose modifications and gender or race status (p ≥ 0.198). Therefore, no statistical association was detected between the frequency of dose modification and gender or racial status.

In addition, medical records were reviewed for other confounding factors that would explain a change in the INR such as nonadherence, health status changes, and dietary inconsistencies. No confounding variables were identified in any of the 32 patient encounters. Of the 32 patient encounters, the post-INR increased in 29 encounters, remained the same in 1 encounter, and decreased in 2 encounters. INR elevations greater than 5 were observed in 5 patients ( Table 2 ). Although a reduced anticoagulation effect was seen in 2 patients, the post-INR remained within the patients' target range in both (pre-INR 2.3 vs post-INR 2.1 and pre-INR 2.4 vs post-INR 2).

Following corticosteroid initiation, warfarin dose modifications (dose reduction and/or withheld dose) were required in 16 (50%) patient encounters. At least one dose of warfarin was withheld in 12 (37.5%) encounters, and reduction in warfarin dosing was prescribed in 11 (34.3%) encounters. The frequencies of withheld doses and dose reductions were separately compared within gender and ethnicity, and no statistically significant difference was found among groups. Although half of the patients had INR elevations requiring intervention, only one adverse event of minor epistaxis was reported. No ED visits or hospitalizations occurred as a consequence of the drug combination.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.