* Context.--Ginsengs are widely used by the general population. These herbs interfere with serum digoxin measurement using the fluorescence polarization immunoassay.

Objective.--To assess potential interference of different ginsengs (Asian, American, and Indian, also known as Ashwagandha)

Design.--Aliquots of drug-free serum pools were supplemented with ginseng and apparent digoxin concentrations were measured using enzyme-linked chemiluminescent immunosorbent digoxin assay, turbidimetric assay, and fluorescence polarization immunoassay digoxin assays. Mice were fed with different ginseng preparations and apparent digoxin concentrations were measured 1 and 3 hours later. In a separate experiment, aliquots of serum digoxin pools were further supplemented with ginsengs and the serum digoxin concentrations were measured again.

Results.--A significant apparent digoxin concentration was observed both in vitro and in vivo using the fluorescence polarization immunoassay, but no apparent digoxin concentration was observed using enzyme-linked chemiluminescent immunosorbent digoxin assay and turbidimetric assay. No interference was observed with enzyme-linked chemiluminescent immunosorbent digoxin assay and turbidimetric assay when digoxin serum pools were further supplemented with various ginsengs.

Conclusions.--It was concluded that both enzyme-linked chemiluminescent immunosorbent and turbidimetric digoxin assays are free from ginseng interferences.

(Arch Pathol Lab Med. 2007;131:619-621)

Ginseng is a popular herbal remedy in the United States. The ginseng that grows in Manchuria is Panax ginseng, which is commonly marketed as "Asian ginseng." It is used as an antioxidant, anti-inflammatory agent, anticancer remedy as well as a cardioprotective agent in traditional Chinese medicines, but its pharmacologic properties have not been established by rigorous research. (1) Asian ginseng contains several compounds called ginsenosides. (1,2) American ginseng is the extract of Panax quinquefolius. More recently, Ashwagandha (Indian ginseng), prepared from an entirely different plant (Withania somnifera), has become commercially available in the US market. Although this product is popularly called Indian ginseng, the Farm Security and Rural Investment Act of 2002 (Farm Bill) banned commercial use of the name "ginseng" for any product not containing the herb in the genus Panax. Ashwagandha has been used in Indian ayurvedic medicine for more than 3000 years. The major constituents are steroidal alkaloids and steroidal lactones in a class of compounds termed withanolides, which have structural similarity to digoxin. (3) It has been previously reported that Asian and American ginsengs and Ashwagandha (Indian ginseng) interfere with serum digoxin measurement by the fluorescence polarization immunoassay (FPIA). (4,5)

Recently, the Bayer HealthCare Diagnostics Division (Tarrytown, NY) marketed a new enzyme-linked chemiluminescent immunosorbent digoxin assay (ECLIA-digoxin) for application on the ADVIA IMS system. We evaluated potential interference of Asian, American, and Indian ginsengs on serum digoxin measurement both in vitro and in vivo in a mouse model with the ECLIA-digoxin assay and a turbidimetric assay (TIA) on Bayer's ADVIA Chemistry systems and compared the results with the results obtained by using FPIA. To our knowledge, the potential interference of Asian, American, and Indian ginsengs with the ECLIA or the turbidimetric digoxin assays has never been studied.

MATERIALS AND METHODS

Different ginsengs used in this study were purchased from local herbal stores in Houston, Tex. Ashwagandha was manufactured by Herb Pharm (Williams, Ore) and was sold as a liquid extract. Asian ginseng was the Song Shiu Pan brand P ginseng manufactured in Shanghai, China. The North American ginseng was also a liquid extract sold by Herba Natural Products Inc (Brooklyn, NY). The fluorescence polarization immunoassay kits for digoxin were purchased from the Abbott Laboratories (Abbott Park, Ill) and the assays were run on an FLx/TDx analyzer (Abbott). The new assay, ECLIA-digoxin, was run on the ADVIA IMS 800i modular system, and the digoxin TIA was run on the ADVIA 1650 analyzer (Bayer). Swiss Webster female mice (age, 8-9 weeks) were purchased from Harlan Sprague Dawley (Houston, Tex). The experimental protocol involving mice was approved by the Animal Protection Committee of the University of Texas-Houston Medical Center. An Amicon Centrifree membrane with a 30 000 molecular weight cutoff (Millipore Corporation, Bredford, Mass) was used for the preparation of serum ultrafiltrate to determine free apparent digoxin concentration.

The FPIA digoxin assay is linear to a serum digoxin concentration of 5.0 ng/mL with a sensitivity of 0.20 ng/mL. The ECLIA-digoxin assay is linear to a serum digoxin concentration of 6.0 ng/mL and has sensitivity of 0.04 ng/mL. The TIA digoxin assay, a latex-enhanced immunoturbidimetric assay, is linear to a serum digoxin concentration of 5.0 ng/mL and the sensitivity of the assay is 0.10 ng/mL.

Two drug-free serum pools, and 2 additional serum pools prepared from patients receiving digoxin, were used for the studies. Serum specimens containing digoxin were combined to prepare each digoxin pool after removing patient's identity. These specimens are routinely submitted to our clinical laboratory for therapeutic drug monitoring of digoxin; only left-over discarded specimens were used after reporting other clinical results (specimens were saved for 1 week as required by existing laboratory protocols). In the first set of experiments, liquid extracts of Asian, American, or Indian ginseng were added to drug-free serum to achieve concentrations mimicking in vivo concentration range expected after recommended use and severe overdose. After supplementation, apparent digoxin concentrations were measured in triplicate using all 3 digoxin immunoassays and values were expressed as the mean [+ or -] standard deviation. To examine potential in vivo interference of the ginsengs in digoxin assays, mice were fed 300 [micro]L of Asian, American, or Indian ginseng preparations by gavage, and blood specimens were collected 1 and 3 hours after feeding using the retro-orbital bleeding technique. Apparent digoxin concentrations were measured by all 3 immunoassays.

Because cross-reactivity should be studied in the presence of the primary analyte, 2 serum pools prepared from patients receiving digoxin were further supplemented with various ginsengs, and digoxin concentrations were measured again using all 3 digoxin immunoassays. All measurements were performed in triplicate and the values expressed as the mean [+ or -] 1 SD. In another experiment, separate aliquots of another drug-free serum pool were supplemented with Asian, American, and Indian ginsengs (60 [micro]L/mL of serum) to study protein binding of digoxin-like immunoreactive components of various ginsengs. After incubation at 37[degrees]C for 1 hour, total and free apparent digoxin concentrations (in the protein-free ultrafiltrate prepared by centrifuging serum for 40 minutes at 1500g using Centrifree) were measured using the FPIA. Statistical analysis was performed using the 2-tailed Student t test. A difference was considered significant only at a 95% confidence level or higher (P < .05).

RESULTS

When the aliquots of drug-free serum pool were supplemented with Asian, American, and Indian ginsengs, significant apparent digoxin concentrations were observed using the FPIA (range, 0.35-0.71 ng/mL, digoxin equivalent), but no apparent digoxin concentrations were observed using the TIA. Very low apparent digoxin concentrations were observed using the ECLIA assay for Asian and American ginsengs, but such apparent digoxin concentrations may be related to the very low detection limit (0.04 ng/mL) of the ECLIA assay. When mice were fed with various ginsengs, apparent digoxin concentrations were observed using only the FPIA assay (Table 1). No apparent digoxin concentration was observed in the serum of any mouse using ECLIA and TIA assay, indicating that metabolites of ginsengs do not interfere with these assays. The half-life of digoxin-like immunoreactive components was relatively shorter with Indian ginseng compared with both Asian and American ginsengs, indicating that metabolites of such ginseng may also interfere with the FPIA assay.

In a different experiment, the effect of Asian, American, and Indian ginsengs on serum digoxin measurement was studied by supplementing aliquots of 2 digoxin pools with various ginsengs. We observed statistically significant increases in serum digoxin concentrations in the presence of various ginsengs when FPIA was used. In contrast, both the ECLIA and TIA assays showed no significant change in serum digoxin measurement in the presence of various ginsengs, indicating that both the ECLIA and the TIA assay are free from interferences of various ginsengs (Table 2). Digoxin-like immunoreactive components of different ginsengs are poorly protein-bound (30%-35%) in vitro in human serum.

COMMENT

Reported toxicity of Asian ginseng and unknown toxicity of recently available Indian ginseng prevented performing any experiments involving human volunteers. However, in vivo data in mice indicate that metabolites of ginseng do not interfere with the ECLIA and TIA assays. A relatively high dose of ginseng was used for feeding mice to study a worst-possible scenario. There is a popular belief that ginseng can stimulate the heart, thus manifesting as a popular herbal remedy for use in the United States. Because of its widespread use, it is possible that an individual taking digoxin may also take ginseng. Therapeutic drug monitoring for digoxin is strongly recommended for efficacy as well as avoiding digoxin toxicity. (6) Although digoxin is a useful drug in the management of chronic heart failure, when such patients also have mild-to-moderate renal failure, digoxin therapy may also increase the risk of digoxin toxicity. Digoxin has a narrow therapeutic range and digoxin toxicity may occur with lower digoxin level if hypokalemia, hypomagnesemia, or hypothyroidism coexists. A recent clinical trial indicated a beneficial effect of digoxin on morbidity, and no excess mortality in women, when serum concentrations from 0.5 to 0.9 ng/mL were examined. However, serum concentrations at 1.2 ng/mL or more appeared harmful. (7)

Although concentration of digoxin in serum can be measured by sophisticated techniques such as high-performance liquid chromatography combined with tandem mass spectrometry, most clinical laboratories utilize immunoassay for routine digoxin monitoring because of simplicity of operation, automation, and speed. Therefore, the ideal immunoassay for monitoring digoxin in a patient also self-medicated with ginseng should be one that is free from interference of ginseng. Moreover, because of poor protein binding of digoxin-like immunoreactive components of ginseng, elimination of this interference in the FPIA assay by monitoring free digoxin in the protein-free ultrafiltrate is not possible. Our study indicates that both the ECLIA-digoxin assay and the turbidimetric digoxin assay are virtually free from such interference. One explanation for why these tests are free from the interference of ginseng is that both use a specific antidigoxin monoclonal antibody in the assay design and the FPIA uses a polyclonal rabbit antidigoxin antibody.

In conclusion, our study is the first to report that the ECLIA-digoxin assay, as well as the turbidimetric digoxin assay, is free from interference of Asian, American, and Indian ginsengs.

References

(1.) Kiefer D, Pantuso T. Panax ginseng. Am Fam Physician. 2003;68:1539-1542.

(2.) Ma XQ, Liang XM, Xu Q, Zhang XZ, Xiao HB. Identification of ginsenosides in roots of panax ginseng by HPLC-APCI/MS. Phytochem Anal. 2005;16:181-187.

(3.) Khajuria RK, Suri KA, Gupta RK, et al. Separation, identification and quantification of selected withanolides in plant extract of withania sominifera by HPLC-UV(DAD)-positive ion electrospray ionization mass spectrometry. J Sep Sci. 2004;27:541-546.

(4.) Dasgupta A, Wu S, Actor JK, Olsen M, Wells A, Datta P. Effect of Asian and Siberian ginseng on serum digoxin measurement by five digoxin immunoassays: significant variation in digoxin-like immunoreactivity among commercial ginsengs. Am J Clin Pathol. 2003;119:298-303.

(5.) Dasgupta A, Reyes M. Effect of Brazilian, Indian, Siberian, Asian and North American ginseng on serum digoxin measurement by immunoassays and binding of digoxin-like immunoreactive components of ginseng with fab fragment of antidigoxin antibody (digibind). Am J Clin Pathol. 2005;124:229-236.

(6.) El Desoky ES, Al-Ghamdi HA, Halaby FH, Al-Beshri M. Therapeutic monitoring of digoxin and antiepileptic drugs in Egypt and Saudi Arabia. Ther Drug Monit. 2003;25:211-214.

(7.) Adams KF, Patterson JH, Gattis WA, et al. Relationship of serum digoxin concentrations to mortality and morbidity in women in the digitalis investigation group trial: a retrospective study. J Am Coll Cardiol. 2005;46:497-504.

Accepted for publication September 6, 2006.

Amitava Dasgupta, PhD; Edward Kang, PhD; Margaret Olsen, BS; Jeffrey K. Actor, PhD; Pradip Datta, PhD

From the Department of Pathology and Laboratory Medicine, University of Texas-Houston Medical School, Houston (Drs Dasgupta and Actor and Ms Olsen); and Bayer HealthCare, Diagnostics Division, Tarrytown, NY (Drs King and Datta).

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: Amitava Dasgupta, PhD, Department of Pathology and Laboratory Medicine, University of Texas-Houston Medical School, 6431 Fannin, MSB 2.292, Houston, TX 77030 (e-mail: Amitava.Dasgupta@uth.tmc.edu).

Table 1. Apparent Digoxin Concentrations in Mice After Feeding With
Asian, American, and Indian Ginsengs

                         Digoxin, ng/mL *

Mouse No.                1 h         3 h

Asian ginseng
  1                      0.79        0.51
  2                      0.71        0.47
  3                      0.74        0.49
American ginseng
  1                      0.68        0.49
  2                      0.58        0.55
  3                      0.61        0.44
Indian ginseng
  1                      0.60        0.30
  2                      0.62        0.25
  3                      0.51        0.20

* Measured by the fluorescence polarization immunoassay.

Table 2. Effect of Ginseng Supplementation to Digoxin Pools on Serum
Digoxin Measurement by Immunoassays

                             Digoxin Concentration, Mean (SD), mg/mL *
Specimen and Type
  of Ginseng                        FPIA              ECLIA-Digoxin

Digoxin pool 1               2.73 (0.02)               2.86 (0.07)
  +25 [micro]L/mL Asian      2.94 (0.01) ([dagger])    2.82 (0.02)
  +50 [micro]L/mL Asian      3.28 (0.06) ([dagger])    2.90 (0.03)
  +25 [micro]L/mL American   2.92 (0.01) ([dagger])    2.87 (0.01)
  +50 [micro]L/mL American   3.16 (0.08) ([dagger])    2.81 (0.05)
  +25 [micro]L/mL Indian     2.89 (0.02) ([dagger])    2.83 (0.04)
  +50 [micro]L/mL Indian     3.03 (0.02) ([dagger])    2.88 (0.05)
Digoxin pool 2               1.22 (0.02) ([dagger])    1.37 (0.03)
  +25 [micro]L/mL Asian      1.46 (0.04) ([dagger])    1.33 (0.02)
  +50 [micro]L/mL Asian      1.75 (0.06) ([dagger])    1.38 (0.01)
  +25 [micro]L/mL American   1.41 (0.03) ([dagger])    1.31 (0.06)
  +50 [micro]L/mL American   1.66 (0.07) ([dagger])    1.35 (0.05)
  +25 [micro]L/mL Indian     1.47 (0.06) ([dagger])    1.38 (0.08)
  +50 [micro]L/mL Indian     1.52 (0.02) ([dagger])    1.42 (0.11)

                             Digoxin Concentration,
                               Mean (SD), mg/mL *
Specimen and Type
  of Ginseng                    Turbidimetric

Digoxin pool 1                   2.80 (0.04)
  +25 [micro]L/mL Asian          2.89 (0.06)
  +50 [micro]L/mL Asian          2.86 (0.03)
  +25 [micro]L/mL American       2.76 (0.01)
  +50 [micro]L/mL American       2.75 (0.04)
  +25 [micro]L/mL Indian         2.88 (0.05)
  +50 [micro]L/mL Indian         2.84 (0.02)
Digoxin pool 2                   1.35 (0.03)
  +25 [micro]L/mL Asian          1.38 (0.01)
  +50 [micro]L/mL Asian          1.33 (0.02)
  +25 [micro]L/mL American       1.33 (0.04)
  +50 [micro]L/mL American       1.30 (0.04)
  +25 [micro]L/mL Indian         1.42 (0.01)
  +50 [micro]L/mL Indian         1.45 (0.02)

* FPIA indicates fluorescence polarization immunoassay; ECLIA-Digoxin,
enzyme-linked chemiluminescent immunosorbent digoxin assay.

([dagger]) Significantly greater than the corresponding value of the
digoxin pool by independent 2-tailed t test (P < .05)