Abstract
Aims
Ginkgo biloba is available as an over-the-counter drug and reported to cause haemorrhage when coadministered with other antiplatelet agents. We set out to study the interactions of G. biloba with cilostazol and clopidogrel.
Methods
A randomized, open-label, crossover study of 10 healthy male volunteers. The dosage schedules were 120 mg G. biloba, 240 mg G. biloba, 100 mg cilostazol, 200 mg cilostazol, 75 mg clopidogrel, 150 mg clopidogrel, 120 mg G. biloba+ 100 mg cilostazol and 120 mg G. biloba+ 75 mg clopidogrel. Platelet aggregation, platelet count, bleeding time and clotting time were measured 0 and 6 h after drug administration. Platelet aggregation was performed using a dual channel aggregometer, by the turbimetric technique using adenosine diphosphate 5 µmol and 10 µmol, and collagen 1 µg ml−1.
Results
Platelet inhibition with the combination of G. biloba and clopidogrel or cilostazol was not statistically significant compared with individual doses of drugs, with all the three aggregants. There was significant (P< 0.05) potentiation of prolongation of bleeding time with the combination of cilostazol and G. biloba compared with individual doses of both the drugs. There was no significant change in clotting time and platelet count.
Conclusions
Coadministration of G. biloba either with cilostazol or clopidogrel did not enhance antiplatelet activity compared with individual agents. Ginkgo biloba potentiated the bleeding time prolongation effect of cilostazol. There was no significant correlation between prolongation of bleeding time and inhibition of platelet aggregation.
Keywords: bleeding time, Ginkgo biloba, healthy volunteers, pharmacodynamic interaction, platelet aggregation
Introduction
The use of herbal supplements has become increasingly popular in recent years. Herbal medicines fall into the category of alternative/complementary medicines. These are not regulated by governments with the same scrutiny as conventional drugs. In the USA their regulation by the Food and Drug Administration is restricted as a result of the Dietary Supplement Health and Education Act (DSHEA) passed by the US Congress in 1994. These products are available to consumers as over-the-counter items in various forms of preparations and doses [1]. Several factors contribute to the increased use of herbal products: (i) easy accessibility, (ii) the perception that herbs are safe, (iii) the desire for self-medication, (iv) they are less costly, and (v) most of all, enactment in 1994 of the DSHEA, which broadly defined herbal products as nutritional supplements. However, herbal products are not as safe as they are believed to be. Most studies reveal that heavy metals, particularly lead, have been a regular constituent of Indian remedies [2], thus causing serious harm to patients taking such remedies.
Concern has recently been expressed regarding the safety of these products, in particular the potential interaction of these drugs with conventional drugs. It has been documented that as many as 31% of patients use herbal supplements concurrently with the prescribed conventional drugs and 70% of them do not report the use of these products to their healthcare providers [1]. It is therefore likely that healthcare professionals will encounter patients who use herbal supplements and who may seek help concerning herb–drug interactions. In order to tackle this problem and provide a better healthcare service to such patients, practitioners should have knowledge of at least commonly occurring or anticipated interactions between herbal supplements and conventional drugs [1]. Despite the widespread use of herbal medicines, documented herb–drug interactions are sparse.
Ginkgo biloba is commonly used and one of the seven top-selling herbal medicines [3]. It is used mainly in the management of intermittent claudication and other vascular disorders to improve circulation. The antiplatelet drug clopidogrel and the newer phosphodiesterase inhibitor, cilostazol, have been recently introduced for management of peripheral arterial disease (PAD). Both these agents are known to cause prolongation of bleeding time and cause haemorrhage [4–6]. In view of the known antiplatelet activity of G. biloba[7], there is a possibility that concurrent use of G. biloba, cilostazol and clopidogrel in patients with PAD may potentiate the antiplatelet activity and cause prolongation of bleeding time, thus increasing the risk of haemorrhage. The present study was conducted to evaluate the pharmacodynamic interactions of G. biloba with clopidogrel and cilostazol, after single-dose administration.
Materials and methods
The present interaction study was conducted in the Department of Clinical Pharmacology and Therapeutics of Nizam’s Institute of Medical Sciences, Hyderabad. It was a randomized, open-label, crossover study of 10 healthy male subjects. The mean age was 27 ± 4 years (24–35 years). The mean body weight was 64 ± 9 kg (50–80 kg) and the mean height was 166 ± 6 cm (156–174 cm). Exclusion criteria were subjects hypersensitive to study drugs, chronic smokers or alcoholics, and a history of gastrointestinal surgery that could interfere with absorption of study drugs. The study was approved by Institutional Review Board of the institute and written informed consent was obtained from all subjects.
The subjects were given one of the following dosage regimens orally, according to randomization after overnight fasting. The dosage schedules were a single dose of (i) 100 mg cilostazol, (ii) 200 mg cilostazol, (iii) 120 mg G. biloba, (iv) 240 mg G. biloba, (v) 75 mg clopidogrel, (vi) 150 mg clopidogrel, (vii) 120 mg G. biloba+ 100 mg cilostazol and (viii) 120 mg G. biloba+ 75 mg clopidogrel. The wash-out period was 48 h. Platelet aggregation, bleeding time, clotting time and platelet count were measured 0 and 6 h after drug administration, haemodynamic parameters blood pressure (BP), diastolic time, heart rate, pulse duration time B/A and C/A ratios were measured 0, 1, 2, 4 and 6 h by using a data acquisition system (Dicrowin, Genesis Medical Systems, Hyderabad, India). In this, digital pulse volume was recorded by connecting the photo transducer to the right-hand index finger of the subject, lying in supine position. The pulse curves were analysed by in-house developed computer software. Measurement of platelet aggregation [8–10] was done by using a dual-channel aggregometer (chronolog) by the turbimetric method, using adenosine diphosphate (ADP) 5 and 10 µmol and collagen 1 µg ml−1. Nine millilitres was collected from the antecubital vein in a plastic test tube containing 1 ml of sodium citrate. The aggregometer was switched on about 30 min before the test to allow the heating block to warm up to 37°C. Platelet-rich plasma (PRP) was prepared by centrifuging blood at 110 g for 15 min. Platelet-poor plasma (PPP) was then prepared from the remaining blood by centrifuging at 2400 g for 20 min. PRP (0.5 ml) was taken into a small plastic tube and a stir bar was added and placed in the heating block. The transmission was set to zero on the chart recorder. PRP was allowed to warm to 37°C for 2 min and then 2.5 µl containing ADP 5 µmol was added.
The change in absorbance was recorded until the response reached a plateau or for 6 min, whichever was sooner. This procedure was repeated with ADP 10 µmol and collagen 1 µg ml−1. Bleeding time was measured by Ivy’s method [9].
The primary purpose was to discover any significant change in platelet aggregation from baseline. More than 20% change in platelet aggregation was considered to be a positive response. Secondary aims were to detect changes in coagulation parameters and haemodynamic parameters. Subjects were allowed to leave the department after 6 h if there were no side-effects.
Statistical analysis
Data are presented as mean, SD, SE and 95% confidence interval. Baseline and data obtained after 6 h in all groups were compared using anova. Differences within each group were analysed using Wilcoxon’s signed rank test, as the platelet aggregation data did not follow normal distribution. For parametric data, paired and unpaired t-tests were applied. The level of significance was set at 0.05 with a power of 80%.
Results
In the present investigation, 16 healthy human volunteers were screened for inclusion in the study. Out of 16 subjects, four were excluded at screening because of abnormal laboratory investigations, two withdrew voluntarily, leaving 10 to be recruited. The mean peak platelet aggregation of eight dosage schedules at baseline were 75 ± 5%, 79 ± 4% and 86 ± 3 with ADP 5 µmol, 10 µmol and collagen 1 µg ml−1, respectively. The mean peak platelet aggregation induced with ADP 5 µmol before and after administration of each drug and their combinations is shown in Table 1. The maximum inhibition of platelet aggregation was seen with 150 mg of clopidogrel. The peak platelet aggregation with 150 mg of clopidogrel was found to be 32 ± 6% at 6 h compared with 73 ± 4% at baseline (P< 0.0001). The combination of G. biloba with either cilostazol or clopidogrel did not potentiate the inhibition of 5 µmol ADP-induced platelet aggregation.
Table 1.
Maximum platelet aggregation percentage with adenosine diphosphate (5 µmol)
| Mean | SD | SE | 95% CI (lower) | 95% CI (upper) | ||
|---|---|---|---|---|---|---|
| Cilostazol 100 mg (n = 10) | 0 h | 74 | 19 | 6 | 60 | 89 |
| 6 h | 47* | 32 | 11 | 23 | 72 | |
| Cilostazol 200 mg (n = 10) | 0 h | 75 | 16 | 6 | 62 | 88 |
| 6 h | 51** | 26 | 9 | 31 | 72 | |
| Ginkgo 120 mg (n = 10) | 0 h | 69 | 18 | 6 | 57 | 82 |
| 6 h | 44** | 22 | 7 | 28 | 60 | |
| Ginkgo 240 mg (n = 10) | 0 h | 81 | 13 | 4 | 71 | 90 |
| 6 h | 59** | 24 | 8 | 42 | 76 | |
| Clopidogrel 75 mg (n = 10) | 0 h | 70 | 15 | 5 | 57 | 83 |
| 6 h | 52* | 26 | 9 | 30 | 73 | |
| Clopidogrel 150 mg (n = 10) | 0 h | 73 | 10 | 4 | 64 | 82 |
| 6 h | 32** | 16 | 6 | 18 | 45 | |
| Clopidogrel + Ginkgo (n = 10) | 0 h | 75 | 9 | 3 | 67 | 83 |
| 6 h | 49*† | 21 | 8 | 31 | 67 | |
| Cilostazol + Ginkgo (n = 10) | 0 h | 84 | 8 | 3 | 77 | 90 |
| 6 h | 57* | 22 | 8 | 39 | 75 |
P < 0.05 compared with baseline.
P < 0.001 compared with baseline.
P < 0.05 compared with 150 mg clopidogrel.
The mean platelet aggregation obtained with 10 µmol of ADP is shown in Table 2. It was observed that there was higher peak platelet aggregation with 10 µmol of ADP. The highest percentage inhibition was noted with 150 mg of clopidogrel (similar to ADP 5 µmol), 48 ± 7% (P< 0.05). The coadministration of G. biloba with either cilostazol or clopidogrel did not enhance antiplatelet activity compared with individual agents.
Table 2.
Maximum platelet aggregation percentage with adenosine diphosphate (10 µmol)
| Mean | SD | SE | 95% CI (lower) | 95% CI (upper) | ||
|---|---|---|---|---|---|---|
| Cilostazol 100 mg (n = 10) | 0 h | 79 | 17 | 6 | 66 | 92 |
| 6 h | 59* | 13 | 10 | 36 | 82 | |
| Cilostazol 200 mg (n = 10) | 0 h | 82 | 14 | 5 | 72 | 93 |
| 6 h | 65* | 27 | 9 | 44 | 86 | |
| Ginkgo 120 mg (n = 10) | 0 h | 75 | 11 | 3 | 67 | 82 |
| 6 h | 85* | 5 | 2 | 81 | 89 | |
| Ginkgo 240 mg (n = 10) | 0 h | 85 | 11 | 3 | 77 | 93 |
| 6 h | 66* | 21 | 7 | 51 | 82 | |
| Clopidogrel 75 mg (n = 10) | 0 h | 77 | 11 | 4 | 68 | 86 |
| 6 h | 60* | 23 | 8 | 41 | 79 | |
| Clopidogrel 150 mg (n = 10) | 0 h | 75 | 11 | 4 | 66 | 84 |
| 6 h | 48* | 19 | 7 | 32 | 64 | |
| Clopidogrel + Ginkgo (n = 10) | 0 h | 75 | 10 | 4 | 75 | 92 |
| 6 h | 56* | 19 | 7 | 61 | 92 | |
| Cilostazol + Ginkgo (n = 10) | 0 h | 83 | 8 | 3 | 68 | 81 |
| 6 h | 77 | 20 | 7 | 39 | 73 |
P < 0.05 compared with baseline.
Collagen was found to be the most potent platelet aggregating agent compared with 5 and 10 µmol of ADP. None of the individual doses of drugs or their combinations was able to produce statistically significant inhibition of collagen-induced platelet aggregation.
In our study, both doses of cilostazol, G. biloba, clopidogrel and their combinations produced marked prolongation of bleeding time. The mean bleeding time obtained in all groups is given in Table 3. The highest bleeding time of 211 ± 70 s (P< 0.001) was seen with the combination of G. biloba 120 mg and cilostazol 100 mg. There was significant potentiation in prolongation of bleeding time with the combination of cilostazol and G. biloba compared with individual doses of both drugs. It was observed that there was no statistically significant correlation (r = −0.01) between prolongation of bleeding time and inhibition of platelet aggregation. There were no significant changes in clotting time or platelet count in any of the treatment groups.
Table 3.
Bleeding time (s)
| Mean | SD | SE | 95% CI (lower) | 95% CI (upper) | |
|---|---|---|---|---|---|
| Baseline, n = 80 | 107 | 33 | 4 | 99 | 115 |
| Cilostazol 100 mg (n = 10) | 150* | 42 | 14 | 118 | 182 |
| Cilostazol 200 mg (n = 10) | 138** | 30 | 10 | 115 | 161 |
| Ginkgo 120 mg (n = 10) | 144* | 28 | 9 | 124 | 164 |
| Ginkgo 240 mg (n = 10) | 133* | 30 | 10 | 110 | 157 |
| Clopidogrel 75 mg (n = 10) | 141 | 59 | 21 | 91 | 190 |
| Clopidogrel 150 mg (n = 10) | 159* | 56 | 20 | 113 | 206 |
| Clopidogrel + Ginkgo (n = 10) | 148* | 78 | 28 | 83 | 213 |
| Cilostazol + Ginkgo (n = 10) | 211*† | 70 | 25 | 153 | 270 |
P < 0.05 compared with baseline.
P < 0.001 compared with baseline.
P < 0.05 compared with 150 mg of clopidogrel.
There was a significant increase in mean heart rate from 63 ± 10 to 65 ± 10 beats min−1 (P< 0.05) and pulse duration time from 962 ± 150 to 1037 ± 182 ms (P< 0.05) with cilostazol 200 mg at 2 h after drug administration (Table 4). The pulse duration time was also increased with the combination of cilostazol 100 mg and G. biloba 120 mg from 932 ± 140 to 1036 ± 123 ms at 2 h (P< 0.01). There was no significant change in systolic BP in any of the groups, but diastolic BP decreased significantly (P< 0.001) from 74 ± 5 to 68 ± 6 mmHg with cilostazol 200 mg. No significant change was observed in diastolic time, upstroke time, B/A and C/A ratios. No significant side-effects were observed during the study period. All treatments were well tolerated.
Table 4.
Pulse duration time (ms) at 2 h
| Cilostazol 100 mg | Cilostazol 200 mg | Cilostazol + Ginkgo | ||||
|---|---|---|---|---|---|---|
| 0 h | 2 h | 0 h | 2 h | 0 h | 2 h | |
| Mean | 915 | 937 | 962 | 1037* | 932 | 1036* |
| SD | 142 | 168 | 150 | 182 | 140 | 123 |
| SE | 50 | 59 | 53 | 64 | 49 | 44 |
| 95% CI (l) | 796 | 796 | 837 | 884 | 815 | 933 |
| 95% CI (U) | 1034 | 1077 | 1087 | 1189 | 1049 | 1139 |
P < 0.05 compared with baseline.
Discussion
The increased use of herbal supplements worldwide, particularly in recent years, has contributed significantly to the information available about potential herb and drug interactions. This is particularly important for modern drugs frequently coprescribed or used by patients. There is documented theoretical evidence in the literature about interactions between herbal products and modern drugs [1]. Ginkgo biloba is one of the most popular herbal products, available in various countries and commonly used for its beneficial effects in cerebral and peripheral arterial disease (i.e. dementia and claudication) [7]. Reports have shown that coadministration of G. biloba with commonly coprescribed drugs in peripheral vascular disease causes interactions [11–15]. In the present study, we have evaluated in healthy subjects a pharmacodynamic interaction of G. biloba with the most commonly used antiplatelet agent, clopidogrel, and with the recently introduced phosphodiesterase inhibitor, cilostazol, used in the treatment of PAD. The study was conducted to determine the effect of these drugs on platelet function. Two doses of ADP 5 and 10 µmol and collagen 1 µg ml−1 were used as platelet aggregation-inducing agents. In our study, both agents produced characteristic platelet aggregation of PRP within 6 min of incubation. The onset of aggregation with 5 µmol of ADP was comparatively slow and less than with 10 µmol of ADP. Collagen produced rapid and high platelet aggregation (86 ± 3%). The maximum mean platelet aggregation with 5 µmol of ADP was 75 ± 5% (range 69–84%) compared with 10 µmol of ADP (79 ± 4%, range 75–85%). Our data are in accordance with an earlier report [8].
In our study, a single administration of 120 and 240 mg of G. biloba produced significant inhibition of platelet aggregation induced by 5 µmol of ADP only. It produced no significant inhibition against a higher dose of ADP (10 µmol) or collagen. Chung et al.[16] have reported that a Ginkgo glide mixture given to human volunteers caused significant inhibition of platelet-activating factor-induced platelet aggregation, after ingestion of single doses of 80 and 120 mg. An inhibitory effect of G. biloba on oxidative stress-induced platelet aggregation [17] has also been reported in mice. To observe a significant and consistent effect of G. biloba on platelet aggregation, it has been suggested that Ginkgo must be taken continuously for 2–3 weeks [18].
Coadministration of a single dose of G. biloba 120 mg with either 100 mg cilostazol or 75 mg clopidogrel did not significantly enhance the inhibition of platelet aggregation. There was an apparent, but not statistically significant, increase in the percentage of platelet inhibition. In one study [13] antiplatelet and antithrombotic effects of an oral combination of ticlopidine (50 mg kg−1 day−1) and G. biloba (40 mg kg−1 day−1) in rats was comparable to a large dose (200 mg kg−1 day−1) of ticlopidine in ADP-induced platelet aggregation. In another study [14] of a 33-year-old woman taking a G. biloba, administration of paracetamol was found to have induced spontaneous bilateral subdural haemotomas. In our study we have used a single dose of G. biloba (120 mg) with a single dose of cilostazol. It may be possible that chronic administration of these agents may be associated with significant change in antiplatelet activity.
We found that, in healthy subjects, the mean bleeding time was increased from baseline after administration of two doses of individual drugs as well as their combinations (Table 3). Compared with other treatment groups in our study, the combination of G. biloba with cilostazol, produced a significant prolongation of bleeding time. Bleeding has also been reported when G. biloba has been administered with aspirin [11], rofecoxib or warfarin [12]. However, there was no change in clotting time or platelet count.
Evaluation of haemodynamic parameters in the present study suggests that there is a moderate effect only with cilostazol 200 mg. No other treatment groups produced any significant change in BP, heart rate or pulse duration time.
Cilostazol 200 mg produced an increase in heart rate and pulse duration which was associated with a significant fall in diastolic BP after 2 h. We found no significant change in systolic BP. Our data are in accordance with a study of a single oral dose of 100 mg cilostazol in 20 healthy volunteers, reporting an increase in heart rate of 13%, with a decrease in diastolic BP [19]. Chronic oral administration of cilostazol in human subjects has been reported to cause a small increase in heart rate (>10 beats min−1) [20]. These effects are due to phophodiesterase inhibition. All the healthy subjects well tolerated the study drugs and their combinations.
In conclusion, coadministration of G. biloba either with cilostazol or clopidogrel did not enhance antiplatelet activity compared with individual agents. Ginkgo biloba potentiated the bleeding time prolongation effect of cilostazol. No statistically significant correlation exists between prolongation of bleeding time and inhibition of platelet aggregation. The results of our preliminary study need further confirmation of interactions of G. biloba with these drugs on chronic administration.
References
- 1.Abebe W. Herbal medications; potential for adverse interactions with analgesic drugs. J Clin Pharm Ther. 2002;27:391–401. doi: 10.1046/j.1365-2710.2002.00444.x. [DOI] [PubMed] [Google Scholar]
- 2.Ernest E. Heavy metals in traditional Indian remedies. Eur J Clin Pharmacol. 2002;57:891–6. doi: 10.1007/s00228-001-0400-y. [DOI] [PubMed] [Google Scholar]
- 3.Izzo AA, Ernst E. Interactions between herbal medicines and prescribed drugs. Drugs. 2001;61:2163–75. doi: 10.2165/00003495-200161150-00002. [DOI] [PubMed] [Google Scholar]
- 4.Beebe HG, Lawson DL, Cutler BS, Herd JA, Strandness DE, Jr, Bortey EB, Forbes WP. A new pharmacological treatment for intermittent claudication. Arch Intern Med. 1999;159:2041–50. doi: 10.1001/archinte.159.17.2041. [DOI] [PubMed] [Google Scholar]
- 5.Thebault JJ, Kieffer G, Cariou R. Single dose pharmacodynamics of clopidogrel. Thromb Haemost. 1999;25:3–8. [PubMed] [Google Scholar]
- 6.Coukell AJ, Markhan A. Clopidogrel. Drugs. 1997;54:745–56. doi: 10.2165/00003495-199754050-00006. [DOI] [PubMed] [Google Scholar]
- 7.Folium Ginkgo. Vol. 1. Geneva: WHO; In WHO Monographs on Selected Medicinal Plants; pp. 154–67. [Google Scholar]
- 8.Laffan MA, Manning RA. Investigation of haemostasis. In: Lewis SM, Bain BJ, Bates I, editors. Dacie and Lewis Practical Hematology. 9. London: Churchill Livingstone; 2001. pp. 339–90. [Google Scholar]
- 9.BcSH haemostasis and thrombosis task force. British Society for Haematology 1988 guidelines on platelet function testing. J Clin Pathol. 1988;41:1322–30. doi: 10.1136/jcp.41.12.1322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Born GVR. Quantitative investigation into aggregation of blood platelets. J Physiol. 1962;16:68. [Google Scholar]
- 11.Rosenblat M, Mindel J. Spontaneous hyphema associated with ingestion of Ginkgo biloba extract. N Engl J Med. 1997;336:1108. doi: 10.1056/NEJM199704103361518. [DOI] [PubMed] [Google Scholar]
- 12.Matthews MK. Association of Ginkgo biloba with intracerebral haemorrhage. Neurology. 1998;50:1933. doi: 10.1212/wnl.50.6.1933. [DOI] [PubMed] [Google Scholar]
- 13.Kim YS, Pyo MK, Park KM, Park PH, Hahn BS, Wu SJ, Yun-Choi HS. Antiplatelet and antithrombotic effects of combination of ticlopidine and Ginkgo biloba extract (Egb 761) Thromb Res. 1998;91:33–8. doi: 10.1016/s0049-3848(98)00075-9. [DOI] [PubMed] [Google Scholar]
- 14.Rowin J, Lewis SL. Spontaneous bilateral subdural haematomas associated with chronic Ginkgo biloba ingestion. Neurology. 1996;46:1775–6. doi: 10.1212/wnl.46.6.1775. [DOI] [PubMed] [Google Scholar]
- 15.Galluzi S, Zanetti O, Binnetti G, Trabucchi M, Frisoni GB. Coma in a patient with Alzheimer’s disease taking low dose trazadone and Ginkgo biloba. J Neurol Neurosurg Psychiatry. 2000;68:679–80. doi: 10.1136/jnnp.68.5.679a. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chung KF, Dent G, McCusker M, Guinot P, Page CP, Barness PJ. Effect of Ginkgo glide mixture (BN 52063) in antagonizing skin and platelet responses to platelet activating factor in man. Lancet. 1987;1:248–51. doi: 10.1016/s0140-6736(87)90066-3. [DOI] [PubMed] [Google Scholar]
- 17.Akiba S, Kawauchi T, Oka T, Hashizume T, Sato T. Inhibitory effect of leaf extract of Ginkgo biloba on oxidative stress induced platelet aggregation. Biochem Mol Biol Int. 1998;46:1243–8. doi: 10.1080/15216549800204812. [DOI] [PubMed] [Google Scholar]
- 18.Braquet P. The Ginkgo glides; potent platelet activating antagonists isolated from Ginkgo biloba L; chemistry, pharmacology and clinical applications. Drugs Future. 1987;12:643–99. [Google Scholar]
- 19.Woo SK, Kang WK, Kwon KI. Pharmacokinetic and pharmacodynamic modeling of antiplatelet and cardiovascular effects of cilostazol in healthy humans. Clin Pharmacol Ther. 2002;71:246–52. doi: 10.1067/mcp.2002.122474. [DOI] [PubMed] [Google Scholar]
- 20.Murada CH, Reilly P, Mohler ER. Cilostazol. Treatment of intermittent claudication. Ann Pharmacol Ther. 2001;35:48–56. doi: 10.1345/aph.19408. [DOI] [PubMed] [Google Scholar]