In Brief Integrative Medicine & Complementary / From Research to Practice

Several botanical and biological products claim to lower blood glucose or decrease complications of diabetes, and some of these are being used by peo- ple with diabetes. Products thought to lower blood glucose include gymnema, fenugreek, bitter melon, ginseng, and nopal. Claims have also been made for aloe, bilberry, and milk thistle, but there is less evidence in support of these. Botanical products thought to decrease diabetes complications include - linolenic acid, ginkgo biloba, and garlic. A vitamin-like substance, -lipoic acid, has been used to treat neuropathic complications.

Biological Complementary Therapies: A Focus on Botanical Products in Diabetes

No one has thoroughly determined concerns about possible side effects3 how many patients with diabetes use and drug interactions.4 Patients using complementary therapies. A recent complementary therapies have experi- Laura Shane-McWhorter, PharmD, survey of diabetes educators in the enced many serious side effects; in BCPS, FASCP, CDE, BC-ADM western half of the United States1 eval- some cases, they may attribute these uated the most frequently recom- effects to another medication. Because mended and used alternative thera- patients often take medications to pies. These included physical activity, treat their diabetes, concomitant use self-help groups, lifestyle diets, laugh- of complementary therapies may also ter and humor, relaxation therapy, result in toxicity secondary to exag- prayer, imagery/visualization, medita- gerated effects or sub-therapeutic tion, massage, and music therapy. effects of their conventional medica- Although botanical products were tions. included in the survey, they were not Another concern relates to the vari- frequently recommended or used. ability of products. Botanical products Another survey of alternative treat- are available in capsules and tablets, ments used by patients with diabetes2 as well as in other forms, such as did indicate that herbal treatments are water extracts (also called decoctions used along with other modalities. In or infusions), tinctures (hydroalco- some cases, patients used these treat- holic extracts), and glycerites (glyc- ments instead of conventional medica- erin-extracted preparations that are tions, and severe complications, alcohol-free). All vary in potency. In including increased hospitalizations, addition, product quality may depend ketoacidosis, and acute hyperglycemia, on what part of the plant was used, occurred. how it was stored, how long it was Technically, an herbal product is stored, the processing technique, and made from the leaves and roots of a how the extract was prepared.3 plant, whereas a botanical product Some products are available in a includes parts or pieces from the form standardized for pharmacologi- whole plant. However, these terms are cal activity. This should guarantee often used interchangeably in discus- that there is consistency from batch to sions of complementary therapies. batch and that the active ingredients are stable.5 However, standardization CONCERNS ABOUT BIOLOGICAL is not simple because, for many OR BOTANICAL THERAPIES botanicals, the active constituents are unknown. A product may be stan- As with conventional medicines, the dardized for one or more biologically use of complementary therapies raises active compounds, but that com- 199 Diabetes Spectrum Volume 14, Number 4, 2001 pound may not be the active ingredi- tions may occur from the additive along with other legumes.9,10 The ent. Pharmacological action may effects when used concomitantly with plant grows in India, Egypt, and the come from the additive or synergistic hypoglycemic agents. Middle East.9 effects of several ingredients, none of Clinical studies. There are only a Fenugreek has also been used as a which separately has the same activity few human studies, and these do not medicinal agent to treat diabetes, con- as the whole plant.6 Furthermore, report important design details, such stipation, and hyperlipidemia.10 It has active constituents in extracts or dried as blinding or randomization. been used topically to treat inflamma- botanicals may vary secondary to geo- One study14 was conducted in type tion, and it has been used postpartum graphical or soil differences, differ- 1 diabetic patients on insulin, 27 of with a substance called jaggery to pro- ences in exposure to sunlight or rain- whom took 200-mg gymnema cap- mote lactation. Because its taste and fall, differences in the time of harvest, sules after breakfast and supper and odor resemble maple syrup, it has and differences in the methods of dry- 37 of whom took insulin only for a been used to mask the taste of medi- ing, storing, and processing. All of period of 6–30 months. After 6–8 cines.9 these variables may affect pharmaco- months, mean HbA1c decreased in the Chemical constituents of the plant logical activity.7 gymnema group from a baseline of include saponins, many of which are Other factors involve potential 12.8 to 9.5% (P < 0.001). After glycosides of diosgenin.9 The seeds misidentification, mislabeling, and 16–18 months, 22 patients remaining also contain the alkaloids trigonelline, possible addition of unnatural toxic on gymnema had a mean HbA1c of gentianine, and carpaine compounds. substances, such as adulteration with 9% (P values not given). At the end of Other components of the seeds heavy metals or steroids and contami- 26–30 months, six patients remaining include several C-glycosides. The nation with microbes, pesticides, on gymnema had a mean HbA1c of seeds contain up to 50% mucilagi- fumigants, and radioactive products.7 8.2% (P values not given). nous fiber.9 Other seed constituents Mean fasting blood glucose (FBG) include 4-hydroxyisoleucine, an BOTANICAL PRODUCTS THAT also decreased from a baseline of 232 amino acid, and fenugreekine. MAY LOWER BLOOD GLUCOSE to 177 mg/dl after 6–8 months 150 Proposed mechanism of action. LEVELS mg/dl after 16–18 months, and 152 Fenugreek is thought to delay gastric mg/dl after 20–24 months (P values emptying, slow carbohydrate absorp- Gymnema not given). The mean insulin dose tion, and inhibit glucose transport.16 It A member of the milkweed family, decreased from a baseline of 60 to 45 has been shown to increase erythro- gymnema (Gymnema sylvestra) is a units/day after 6–8 months and to 30 cyte insulin receptors and improve woody plant found in tropical forests units/day at 26–30 months (P values peripheral glucose utilization, thus of India and Africa.8–10 For more than not given). Patients on placebo had no showing potential pancreatic as well 2,000 years, people have chewed its significant changes from baseline. as extrapancreatic effects.17 leaves to treat “madhu meha” (“honey Another study15 was conducted in Various components of the seeds urine”).9 Used in Ayurvedic medicine, patients with type 2 diabetes on sul- have varying activities. For example, it is thought to destroy a person’s abil- fonylureas; 22 took 400 mg/day of the component called fenugreekine, a ity to discriminate sweet taste; hence, gymnema capsules in addition to sul- steroidal sapogenin peptide ester, may it is often called “gurmar” or “sugar fonylurea treatment, and 25 took a have hypoglycemic properties.10 destroyer.”8 placebo and sulfonylureas for a peri- Trigonelline, another component, may Chemical constituents of the plant od of 18–20 months. Mean HbA1c exert hypoglycemic effects in healthy include the gymnemic acids (gym- decreased from a baseline of 11.9 to patients without diabetes,10 but other nemosides), saponins, stigmasterol, 8.48% (P < 0.001). Mean FBG studies have shown that fenugreek has quercitol, and the amino acid deriva- decreased from 174 to 124 mg/dl no effect on fasting or postprandial tives betaine, choline, and trimethy- after 18–20 months (P < 0.001). Five blood glucose levels in nondiabetic lamine.8 Gymnema has been used for patients were able to discontinue sul- subjects.18 centuries to treat diabetes.10 fonylureas. In this study, lipids also Side effects and drug interactions. Proposed mechanism of action. decreased significantly. Patients on The main side effects are flatulence and Although the exact mechanism is placebo had no significant changes in diarrhea, which subside after a few unknown, there are a variety of theo- HbA1c, FBG, or lipids. days. However, fenugreek also has rized mechanisms. Besides impairing Clinical notes. Because of the pos- uterotonic properties.10 Hypersensitivity the ability to discriminate sweet taste, sibility of hypoglycemia when gymne- reactions have also been reported,10 gymnema increases the enzyme activi- ma is used with insulin or other dia- including rhinorrhea, wheezing, and ty responsible for glucose uptake and betes agents, doses for existing dia- fainting after inhalation of the fenu- utilization.11 It may stimulate -cell betes therapies may have to be adjust- greek-seed powder. Wheezing and function, increase -cell number, ed with the addition of gymnema. facial angioedema after application of a and/or increase insulin release by This product should not be used with- topical fenugreek paste for dandruff increasing cell permeability to out medical supervision. A typical was reported in a patient with chronic insulin.8,12 In pancreatectomized ani- dose is 400 mg/day standardized to asthma.19 mals, it has no hypoglycemic effect, contain 24% gymnemic acids.8,10 Because fenugreek is a member of indicating that its effect may require the Leguminosae family, which in- some residual -cell function.13 Fenugreek cludes peanuts,9 it is theoretically pos- Side effects and drug interactions. Fenugreek (Trigonella foenum-grae- sible for someone with a peanut allergy No side effects have been reported sec- cum) has been used as a cooking spice to react to fenugreek. However, this ondary to gymnema use. Hypoglycemia and flavoring agent for centuries.9 It is reaction has never been reported. is a potential side effect;10 drug interac- a member of the Leguminosae family, All of fenugreek’s side effects may 200 Diabetes Spectrum Volume 14, Number 4, 2001 10 occur in infants of nursing mothers Clinical notes. Fenugreek has been melon has been reported. Favism Integrative Medicine & Complementary / From Research to Practice who use this substance. categorized as “generally recognized (acute hemolytic anemia character- Because of the coumarin con- as safe” in the United States.10 Dose ized by headache, fever, abdominal stituents, fenugreek may potentially recommendations vary depending on pain, and coma) has also been enhance anticoagulant activity of the source. However, all recommend- reported from one of the seed con- drugs or herbs that have antiplatelet ed doses are much lower than those stituents of bitter melon, vicine.10 activity.10 It may inhibit corticosteroid used in the clinical studies, which The red arils around bitter melon drug activity, interfere with hormone ranged from 15 g mixed in water to seeds have produced toxicity; in one therapy, and potentiate monoamine 50 g twice daily. One source recom- child, vomiting, diarrhea, and death oxidase (MAO) inhibitor activity. A mended 1–2 g of the seed or its equiv- occurred.9 Bitter melon has abortifa- theoretical interaction is decreased or alent three times daily or 1 cup of the cient effects9,10 and in animals has delayed absorption of concomitant tea, made by steeping 500 mg seed in produced hepatotoxicity.10 medications because of the high 150 ml cold water several times a Concomitant use with stimulant mucilage content. Additionally, there day.10 laxatives or other agents that decrease may be additive hypoglycemic activity potassium may increase the risk of when combined with diabetes medica- Bitter Melon potassium depletion.10 If combined tions.10 Bitter melon (Momordica charantia), with secretagogues, additive hypo- Clinical studies. There are few also known as bitter gourd, bitter glycemia may occur. This was report- human studies, and most of these are apple, bitter cucumber, karolla, and ed when a patient used both bitter short-term, involve few patients, and karela, is a vegetable cultivated in melon and chlorpropamide.23 do not adequately report study design tropical areas, including India, Asia, Clinical studies. Most studies in details. South America, and Africa. It is yel- humans involve very small numbers, In one study,20 10 patients with low-orange with a bumpy exterior are of very short duration, and have type 1 diabetes on insulin underwent resembling a gherkin and is bitter but been reported with only sketchy infor- a 10-day metabolic trial. Patients were edible.9 mation about details of study design, blinded, but it was unclear whether Bitter melon has been used to treat including blinding and randomization. the investigators were blinded. diabetes and psoriasis and for HIV Although there are several studies Patients were randomized to either supportive therapy, and it has been using bitter melon, only studies in placebo or 50 g fenugreek defatted studied as a potential contraceptive which glucose values were reported seed powder twice a day in chapati agent.9,10 It has been taken for cen- will be discussed. (unleavened bread). Mean FBG turies as a dietary treatment and more An early study of 19 patients (11 decreased from a baseline of 272 to recently has been used as an injectable with type 1 and 8 with type 2 dia- 196 mg/dl (P < 0.01). Patients also extract for research.22 betes)22 used polypeptide-P zinc chlo- demonstrated a statistically significant Bitter melon contains several chemi- ride, a momordica extract, prepared decline in serum total cholesterol (P < cal constituents, including the glyco- in the same manner as bovine insulin. 0.001), triglycerides, and LDL choles- sides mormordin and charantin. This “plant insulin” was injected sub- terol (P < 0.01 vs. placebo for triglyc- Charantin contains mixed steroids cutaneously in five patients with type erides and LDL), but no change in with hypoglycemic activity. Another 1 diabetes and six patients with type 2 HDL cholesterol. component is the peptide polypeptide- diabetes. No details of randomization The largest fenugreek study21 was P.9 Bitter melon also contains the alka- or blinding were provided. FBG was a 6-month trial in 60 patients with loid mormordicine. Its seeds contain measured at the time of injection, as inadequately controlled type 2 dia- the abortifacients -mormorcharin well as 4, 6, 8, and 12 h after injec- betes. The authors did not provide and -mormorcharin, as well as the tion. The control group, consisting of information on randomization or pyrimidine nucleoside vicine.9 six patients with type 1 and two blinding. Patients were administered a Proposed mechanism of action. patients with type 2 diabetes, did not 75-g oral glucose tolerance test The fruit and seeds of bitter melon are receive any treatment. (OGTT) to determine mean baseline thought to exert hypoglycemic effects In the patients with type 1 diabetes, parameters. Fenugreek seed powder, in normal and diabetic animal mean FBG decreased from 304 to 169 25 g/day, was administered in two models.10 The specific components mg/dl 4 h after injection (P < 0.05), equal doses with lunch and dinner for thought to contribute to its hypo- and this effect was maintained at 6 6 months. Mean FBG decreased from glycemic activity include charantin, and 8 h after injection (FBG 176 and a baseline of 151 to 112 mg/dl after polypeptide P, and vicine. Other theo- 174 mg/dl, respectively, P < 0.05 com- 24 weeks. The mean 1-h OGTT base- retical actions include extrapancreatic pared to baseline). At 12 h, FBG had line was 284 mg/dl, compared to 196 activity, such as increased tissue glu- started to increase (208 mg/dl). mg/dl after 24 weeks. Mean 2-h val- cose uptake, liver/muscle glycogen In the patients with type 2 diabetes, ues decreased from a baseline of 257 synthesis, and decreased blood glu- changes were significantly different to 171 mg/dl after 24 weeks (P < cose synthesis through depression of between the experimental and control 0.001 vs. baseline for both). Mean the enzymes glucose-6-phosphatase, groups at 1 and 6 h (P < 0.05). HbA1c decreased from a baseline of fructose-1, and 6 bisphosphatase, and However, there were no significant 9.6 to 8.4% after 8 weeks (P < enhanced glucose oxidation by differences from baseline in the exper- 0.001). Although 40 patients were on enzyme G6PDH pathway.9 imental group. In the control groups oral diabetes medications, the authors Side effects and drug interactions. of both the type 1 and type 2 diabetes did not report on whether they were Bitter melon may produce digestive interventions, blood glucose values able to decrease their doses or discon- tract discomfort.9 Hypoglycemic were not significantly changed from tinue their diabetes drugs. coma from a tea containing bitter baseline. 201 Diabetes Spectrum Volume 14, Number 4, 2001 The largest study using bitter thought to have various hormonal of ginseng tablets for 8 weeks. At the melon24 was done in 100 type 2 dia- and central nervous system (CNS) end of the study, mean FBG values for betic patients but was conducted for effects. Some ginseng compounds the three groups were 149, 139, and only 2 days using an aqueous suspen- show contradictory effects; for exam- 133 mg/dl, respectively. (Results were sion of the vegetable pulp. Day 1 ple, ginsenoside Rg1 has hypertensive only statistically significant for the mean FBG was 152 mg/dl, and after a and CNS-stimulant effects, whereas 100-mg/day group, P < 0.05.) Baseline 75-g OGTT, the mean 2-h postpran- ginsenoside Rb1 has hypotensive and values were not reported, and average 9,25 dial glucose level was 257 mg/dl. On CNS-depressant effects. HbA1c levels at the end of the study day 2, momordica extract was given, Proposed mechanism of action. In were 6.5, 6.5, and 6.0%, respectively and blood glucose was measured 1 h diabetes, the mechanism of beneficial (P < 0.05 for 200-mg group). later. This mean value was 131 mg/dl effect is unknown. Animal research Another study compared the effects and was significantly different from has indicated that ginseng may lower of American ginseng and placebo in the FBG of 160 mg/dl (P < 0.001). blood glucose by possibly decreasing patients with and without type 2 dia- Patients then received a 75-g oral glu- the rate of carbohydrate absorption betes.40 Ten patients without diabetes cose load, and blood glucose was into the portal hepatic circulation28 and nine patients with type 2 diabetes sampled 2 h later. The mean 2-h and possibly increasing glucose trans- were given a 25-g OGTT, with and blood glucose was 222 mg/dl. One port and uptake.29 Another potential without 3 g of ginseng capsules. hour after momordica was given, 86 mechanism is modulation of insulin Patients were given ginseng 40 min patients had decreased values and an secretion.30 Some ginseng fractions before or with the glucose challenge. attenuated blood glucose response to have increased serum insulin levels Patients were blinded, but the authors the 75-g glucose load, in comparison and glucose-stimulated insulin secre- did not state whether the investigators to the previous day (222 vs. 257 tion in mice. were blinded. mg/dl, P value not significant). Side effects and drug interactions. No significant difference in post- Clinical notes. There is insufficient The most commonly reported side prandial glucose was reported when information to recommend a reliable effects include nervousness and excita- ginseng was taken with the glucose dose.10 Various forms of bitter melon tion.9 Other effects include headache, challenge in patients without diabetes. have been used in research, including hypertension, insomnia,9,10 estrogenic Taking ginseng 40 min before the powder, extract, juice, as well as the effects including mastalgia,31 vaginal OGTT resulted in a significant reduc- cooked vegetable. Some sources have bleeding,32 and cerebral arteritis.33 tion in postprandial glucose (P <0.05 stated that 50 ml fresh juice taken “Ginseng abuse syndrome” is a vs. placebo). In patients with diabetes, daily with food may be used.10 controversial adverse effect that was ginseng given either concomitantly or Tinctures are becoming available. reported in 14 of 133 long-term users 40 min before the OGTT significantly Medical supervision is always neces- of high daily doses.34 This syndrome lowered postprandial glucose (P <0.05 sary when using bitter melon. consisted of hypertension, nervousness, vs. placebo). sleeplessness, skin eruptions, increased Clinical notes. Ginseng products Ginseng libido, and morning diarrhea. have been found to be adulterated A variety of products are called “gin- Several drug interactions have been with other substances including man- seng.” The most commonly used are reported. Diuretic resistance can drake root or phenylbutazone,9 and in three different botanicals: Asian or occur when ginseng is used in con- one case even led to positive results on Korean ginseng (Panax ginseng C.A. junction with a diuretic.35 Tremors36 a “doping” test for an athlete.41 Meyer), American ginseng (Panax and hypomania37 can occur when it is Typical doses are 200–600 mg/ quinquefolius L.), or Russian or given with phenelzine, an MAO day.10 These doses are lower than in Siberian ginseng (Eleutherococcus sen- inhibitor. Insomnia, headache, and the study using American ginseng cap- ticosus Maximum). The part used is decreased warfarin effect have also sules but a bit higher than the tablets the root.25 been reported.38 Concomitant use used in the study of newly diagnosed Ginseng has been described as an with diabetes agents may cause hypo- type 2 diabetic patients. Besides cap- “adaptogen”—a drug that may glycemia. If combined with stimu- sules and tablets, ginseng comes in a increase resistance to adverse influ- lants, ginseng may potentiate stimu- variety of forms ranging from fresh ences such as infection and stress.25 lant effects.10 and dried roots to extracts, solutions, Individuals use ginseng in an attempt Clinical studies. Ginseng has been sodas, teas, and cosmetics.9 The root to enhance physical performance, psy- evaluated in many studies of varying contains at least 1.5% ginsenosides.27 chomotor performance, and cognitive quality. A recent meta-analysis that Some people use ginseng on a con- function; for immunomodulation; and evaluated its effects on physical or tinual basis, but others use ginseng for as treatment for infections and dia- psychomotor performance, cognitive a period of 3 weeks to 3 months.10,27 betes.26 Generally, length of use is up function, immunomodulation, and Evidence for efficacy of ginseng is lim- to 3 months, with possible repeated other outcomes26 drawing on only ited, at best. courses.27 In diabetes, only Korean randomized, placebo-controlled stud- and American ginseng have been stud- ies showed that ginseng has subopti- Nopal ied, so this discussion will be limited mal efficacy. Nopal (Opuntia streptacantha Le- to these two substances. Ginseng has only been studied in maire), also known as prickly pear, is Ginseng contains a family of type 2 diabetes. In a randomized, dou- a member of the cactus family.9 steroid-like compounds called gin- ble-blind study of ginseng in 36 newly Multiple species are known as senosides. Although there are many diagnosed type 2 diabetic patients,39 Opuntia, including Opuntia megacan- subtypes, ginsenosides are tetracyclic 12 people each were assigned to tha, Opuntia ficus indica, and Opuntia triterpenoid saponin glycosides placebo, 100 mg/day, or 200 mg/day fuliginosa. Research has focused on 202 Diabetes Spectrum Volume 14, Number 4, 2001 Opuntia streptacantha Lemaire to 206 mg/dl at 60, 120, and 180 min, abdominal cramps, pain, and severe Integrative Medicine & Complementary / From Research to Practice lower blood glucose. respectively (P < 0.05 vs. baseline for diarrhea, with subsequent fluid and Nopal stems are used as a food 60 and 120 min; P < 0.01 vs. baseline electrolyte disturbances. This form source in Mexico.10 The leaves con- at 180 min). may deplete potassium, predisposing tain mucopolysaccharide soluble Another trial compared 14 patients people who use it to cardiac abnor- fibers and phytochemicals. Leaves and with type 2 diabetes on sulfonylureas malities. It may also potentially inter- stems are also used for other reasons, (Group 1) to individuals without dia- act with glycosides because of the such as for diabetes control and treat- betes (Group 2).46 Details of blinding hypokalemic effects.10 Additive ment of hyperlipidemia.10 or randomization were not provided. hypokalemia may also occur when it Proposed mechanism of action. Both groups received 500 g broiled is coadministered with other drugs Studies in pancreatectomized animals nopal or 400 ml water. that deplete potassium, such as diuret- have shown that the hypoglycemic In patients with diabetes, mean ics or corticosteroids.10 Additive activity does not depend on the pres- decreases in glucose concentrations cathartic effects may occur when com- ence of insulin.42 However, the mech- were 21, 28, and 41 mg/dl at 60, 120, bined with other laxatives. Topical anism of action is unknown. The fiber and 180 min, respectively, after nopal use has not been reported to produce and pectin components are thought to administration (P < 0.005 for 60 and any problems. have hypoglycemic activity. Because 120 min vs. baseline; P < 0.001 for Aloe gel taken internally may exac- insulin concentrations decrease with 180 min vs. baseline). There were no erbate Crohn’s disease and ulcerative nopal administration, enhanced significant differences after water colitis. It may also produce additive insulin sensitivity is another theorized administration. Insulin concentrations hypoglycemia when taken concomi- mechanism of action.43 also declined significantly with nopal. tantly with secretagogues.10 Nopal’s hypoglycemic activity has In patients without diabetes, there Clinical studies. A very small, been reported to reach maximum were no significant differences. uncontrolled study was conducted in effects 3–4 h posptrandially and to Clinical notes. Nopal is given with patients with type 2 diabetes.47 Five last up to 6 h.10 The fiber content may meals in doses of 500 g broiled or patients were administered one-half affect intestinal uptake of glucose. fresh nopal stems.10 However, ideal teaspoonful, twice daily, of dried aloe Animal research indicates that the doses and optimal preparation meth- sap for 4–14 weeks. Details were very pectin component may alter hepatic ods have not been established. There sketchy, and information regarding cholesterol metabolism.10 are no known risks with nopal use, blinding was not provided. FBG fell Side effects and drug interactions. but there have been no well-designed from a mean of 273 to 151 mg/dl (P < Reported adverse effects include or long-term studies. Evidence in 0.001). Mean HbA1c decreased from increased stool volume and frequency favor of nopal is limited at this time. 10.6 to 8.2% (significance not report- and abdominal fullness.44 Dermatitis ed). has also been reported.10 In a case Aloe Clinical notes. Aloe has been used report of 8-week coadministration Aloe is a desert plant with a cactus- topically and as a cathartic. It has also with chlorpropamide, additive effects like appearance. It belongs to the fam- been used orally to boost the immune on blood glucose and insulin levels ily Liliaceae.9 There are more than system and for treatment of asthma, were found.43 Additive hypoglycemia 500 species, but the most familiar ulcers, and diabetes. with secretagogues is a theoretical form is aloe vera. It has been used Doses are variable and include concern, although there have been no since prehistoric times for burns and 50–200 mg/day of aloe leaf gel.10 There reports of this. wound healing. is insufficient evidence for the use of Clinical studies. Most trials are There are two forms of aloe vera: aloe in diabetes. Supplementation is small and have been published in dried juice from the leaf and aloe gel. not recommended. Spanish, although English abstracts Latex from pericyclic cells obtained are available. Two small trials have beneath the skin of leaves may be Bilberry been published in English. evaporated to form a sticky substance Bilberry is a plant closely related to One trial involved three groups of known as “drug aloes” or “aloe.” the American blueberry, cranberry, patients with type 2 diabetes treated This aloe juice contains the cathartic and huckleberry.10 Two forms of bil- with diet alone or in combination with anthraquinone, barbaloin, a glucoside berry are used: the dried fruit and the sulfonylureas.45 Details of blinding or of aloe-emodin, as well as other sub- leaf. The dried fruit is used to treat randomization were not provided. stances.9 diarrhea27 and to improve visual acu- After a 12-h fast, 16 patients received Aloe gel is obtained from the inner ity and night vision.10,48 The leaf is 500 g broiled nopal (Group 1); 10 portion of the leaves. It does not con- used for diabetes, arthritis, circulatory people received 400 ml water (Group tain anthraquinones but does contain disorders,9,27 cataracts,49 and varicose 2); and 6 people received 500 g a polysaccharide, glucomannan, that veins.10 In folk medicine, it is used as a broiled zucchini (Group 3). is similar to guar gum.9 Aloe gel is “blood sugar–reducing” drug, and is In Group 1, mean blood glucose used topically, but it has also been therefore a common constituent in declined from 222 mg/dl fasting to used orally for diabetes. “antidiabetic” teas.50 203, 198, and 183 mg/dl after 60, Proposed mechanism of action. Proposed mechanism of action. 120, and 180 min, respectively (P < Aloe gel taken internally may produce Anthocyanosides are bioflavonoids, 0.001 compared to baseline). There a mild reduction in mean glucose lev- chemical constituents in bilberry fruit was no significant drop in glucose in els, but the mechanism has not been thought to be responsible for some of Group 2. In Group 3, broiled zucchini elucidated. its vascular effects.9 Anthocyanosides caused a drop from a mean baseline Side effects and drug interactions. are thought to decrease vascular per- of 246 mg/dl fasting to 226, 219, and The laxative form of aloe may cause meability and redistribute microvascu- 203 Diabetes Spectrum Volume 14, Number 4, 2001 lar blood flow.10 They are similar to Recently, its use has been proposed in group of 30 received 600 mg/day of some of the agents in grape seed. diabetes to diminish insulin resis- silymarin, and the other group of 30 The mechanism in diabetes may be tance.54 received a placebo for 12 months. related to the high chromium content Chemical constituents are found in Mean FBG declined from 190 in bilberry leaf (9 parts per million), the fruit, seeds, and leaves. Milk this- mg/dl at baseline to 165 mg/dl at 12 but further research is needed to tle contains silymarin, which is com- months (P < 0.01 vs. baseline). HbA1c determine this.50 posed of three main constituents: sily- declined from 7.9% at baseline to Side effects and drug interactions. bin, silychristine, and silidianin. 7.2% at 12 months (P < 0.01 vs. base- Reported adverse effects have been Silybin is thought to have the most line). Mean daily insulin requirement mostly benign, including mild diges- potent biological activity.53 decreased significantly from 55 tive distress, skin rashes, and drowsi- Proposed mechanism of action. units/day at baseline to 42 units/day ness.50 There are no known drug There are several proposed mecha- at 12 months (P < 0.01 vs. baseline). interactions. Because it may affect nisms that may benefit patients with Results were significant in the group blood glucose, additive hypoglycemia hepatic disease or impairment.52 One of patients on silymarin, but not in may occur with secretagogues. If an is inhibition of hepatotoxin binding to the control group. alcoholic extract is used, there could hepatocyte membrane receptor sites, Clinical notes. Milk thistle may be a disulfiram reaction (cramping, resulting in hepatocyte stabilization. protect against hepatotoxicity by such diaphoresis, and nausea if alcohol is Another is reduction of glutathione agents as acetaminophen, antipsy- ingested).10 In animals, prolonged use oxidation, which may deplete dimin- chotics, alcohol, and halothane. The and very high doses have resulted in ished glutathione levels in the liver typical dose for liver disease is 200 mg toxicity, including initial cachexia and and intestines. Milk thistle also has three times/day. Milk thistle extract excitation and eventual death.27 antioxidant activity to protect against should be standardized to contain Clinical studies. Despite enthusi- toxic free radicals. Finally, it stimu- 70% silymarin, which then contains asm during World War II about using lates protein synthesis, which may 140 mg silymarin. Because phos- bilberry preserves for improving night lead to enhanced hepatocyte regenera- phatidylcholine enhances oral absorp- vision in Royal Air Force pilots,9 bil- tion. tion, preparations containing this berry has not had demonstrated effi- Milk thistle may benefit patients ingredient may be dosed at 100 cacy in controlled trials.48 Although its who have insulin resistance secondary mg/day.52 These doses differ from use has been suggested for lowering of to hepatic damage. One theory is that doses used in clinical studies, which blood glucose, there have been no lipoperoxidation may affect patients have ranged from 280 to 800 mg/day. human trials. In streptozotocin- with diabetes, and restoration of nor- Milk thistle’s potential use for dia- induced diabetic rats, 4 days of bilber- mal malondialdehyde concentrations betes is very preliminary. ry leaf administration caused plasma may improve diabetes.54 glucose levels to consistently decrease Side effects and drug interactions. BIOLOGICAL PRODUCTS THAT by 26%.51 Reported dose-related (>1,500 mg/ MAY ADDRESS COMPLICATIONS Clinical notes. There is potential day) adverse effects include loose OF DIABETES benefit but very limited evidence that stools as a result of increased bile flow bilberry may improve conditions and secretion.55 Patients with allergies Linolenic Acid related to blood vessel function, such to the Asteraceae or Compositae fami- linolenic acid (GLA) is an -6 fatty as varicose veins, cataracts, and other ly may exhibit cross-allergenicity with acid. Although other sources of GLA ocular diseases. milk thistle administration.10 No include black currant and borage oil, For the fruit, standard doses of the adverse drug interactions have been the main source used in nutritional dried ripe berries are 20–60 g/day. reported with milk thistle. Beneficial supplements is evening primrose oil.9 Decoctions have also been prepared interactions, however, have included GLA has been used to treat diabetic by placing 5–10 g mashed berries in reduction of hepatotoxicity associated neuropathy, hyperlipidemia, mastitis, cold water, simmering for 10 min, with acetaminophen, antipsychotics, premenstrual syndrome, eczema, then straining. The form used in stud- halothane, and alcohol.52 rheumatoid arthritis, and multiple ies has been standardized to contain Clinical studies. Several studies sclerosis.9 25% anthocyanosides.10 A tea is pre- have evaluated the impact of milk Proposed mechanism of action. In pared using 1 g finely chopped dried thistle on hepatic disease. These stud- the body, linoleic acid (LA) is convert- leaf in 150 ml boiling water and ies include acute viral hepatitis, mush- ed to GLA via regulation by the steeping for 5–10 min.10 room poisoning, drug-induced hepati- enzyme -6-desaturase (D6D).56 Two tis, chronic liver disease, and alcoholic metabolites of GLA, dihomogammali- Milk Thistle liver disease.53 However, these studies nolenic acid (DGLA) and arachidonic Milk thistle (Silybum marianum) is a have been fraught with shortcomings. acid (AA) are prostaglandin precursors member of the aster family (Asteraceae Many were open-label, involved small that regulate the balance of platelet or Compositae), which also includes numbers of patients, used different aggregation and maintain blood flow daisies and thistles.52,53 Milk thistle has doses, involved different severities of in small blood vessels.56 been used extensively for various liver disease, lacked control groups, In diabetes, conversion of LA to hepatic disorders, including hepato- and lacked well-defined study end- GLA is thought to be impaired sec- toxicity secondary to acute and chron- points. ondary to problems with D6D. ic viral hepatitis, mushroom poison- Milk thistle was evaluated in a 12- Subsequent problems may occur, such ing, and alcoholic cirrhosis. It has also month, randomized, open-label trial as difficulty maintaining nerve mem- been used to attenuate hepatotoxic in 60 type 2 diabetic patients with cir- brane structure, nerve blood flow, and effects of certain medications.52 rhosis.54 All subjects used insulin. One nerve conduction.57 An exogenous 204 Diabetes Spectrum Volume 14, Number 4, 2001 source of GLA bypasses the need for penoids, consisting of ginkgolides and Garlic Integrative Medicine & Complementary / From Research to Practice conversion of LA to GLA. bilobalides.60 Garlic, a member of the lily family,9 Side effects and drug interactions. Ginkgo biloba is one of the most has been used in cooking for thousands Most adverse effects are mild and widely used drugs in Germany. It is of years. Garlic contains the sulfur-con- include headache and mild gastroin- used for “cerebrovascular insufficien- taining chemical constituent, alliin, testinal (GI) complaints, such as bloat- cy” and dementia. In diabetes, ginkgo which must be converted to allicin (the ing and loose stools.10 There are biloba may be of use in ameliorating active form) by the enzyme allinase. reports of prolonged bleeding time peripheral circulatory problems, such When the garlic bulb is chewed or and one case report of seizures.9 as intermittent claudication.61 There is crushed, this reaction occurs.9 Ajoene Seizures may potentially occur if GLA also some evidence that it may benefit is formed by the acid-catalyzed reac- is combined with phenothiazines sexual dysfunction.62 tion of two allicin molecules. because these drugs may lower the Proposed mechanism of action. Commercial preparations usually seizure threshold.10 The flavone glycosides, including contain allin, not allicin. Conversion Clinical studies. Two main clinical quercetin, kaempferol, and isorham- requires allinase, which is unstable in trials have evaluated GLA. One was a netin, are thought to have antioxidant stomach acids. Dried garlic prepara- 6-month randomized, double-blind, activity and inhibit platelet aggrega- tions may be effective only if they are placebo-controlled trial evaluating its tion. The ginkgolides are thought to enteric-coated to prevent gastric acid effects on peripheral neuropathy in 22 improve circulation and inhibit the breakdown and permit release in the patients with type 1 or type 2 dia- platelet-activating factor. The bilob- small intestine. Fresh garlic is effec- betes.58 Twelve patients received 360 alides are thought to have neuropro- tive.9,10 mg/day GLA, and 10 patients received tective properties.9,10,60 Garlic is used to treat hyperlipi- a placebo. At the end of 6 months, Side effects and drug interactions. demia and hypertension, for cancer there was improvement in neuropathy Patients may experience transient prevention, and for other antibacterial symptom scores (P < 0.001) as well as headaches for the first 2–3 days of use. activity. Although evidence is prelimi- other parameters of neuropathy in the GI upset occurs in <1% of patients. nary, garlic may be useful in dia- GLA group. There was no significant Exposure to the fruit pulp may pro- betes.67 change in HbA1c. duce cross allergenicity with poison Proposed mechanism of action. Another trial was a year-long, mul- ivy. Eating the seeds may produce loss Allicin has antibacterial and antioxi- ticenter, randomized, double-blind, of consciousness and tonic-clonic dant activity. It may possibly increase placebo-controlled trial involving 111 seizures.9 There have been case reports blood levels of catylase and glu- patients with type 1 or type 2 dia- of subdural hematoma,63 subarach- tathione peroxidase activity. Ajoene betes.59 Patients were given 480 noid hemorrhage,64 and bleeding from decreases the activity of factors need- mg/day GLA. The investigators the margin of the iris.65 ed for lipid synthesis by reducing the reported significant improvement in The main drug interaction is the thiol group in coenzyme A and HMG 13 of 16 parameters of neuropathy. potential for additive antiplatelet activ- CoA reductase and also by oxidizing HbA1c did not improve, but GLA ity when combined with antiplatelet NADPH. Ajoene has antiplatelet response was better in those patients drugs, such as warfarin or aspirin or activity and interferes with thrombox- whose initial baseline HbA1c was with herbs that also have antiplatelet ane synthesis and decreases platelet <10%. activity, such as ginger, garlic, and aggregation.9 Clinical notes. GLA derived from feverfew.10 Researchers have noted that garlic EPO may improve problems with Clinical studies. Clinical studies use may be associated with increased nerve membrane structure, impulse have been done with products contain- serum insulin and improved liver conduction, and nerve blood flow. ing 24% flavone glycosides and 6% glycogen storage.67 A constituent of 10 Since HbA1c does not improve, benefit terpene lactones. For intermittent garlic, allylpropyl disulfide, may is not secondary to blood glucose con- claudication, different trials have found reduce blood glucose and increase trol. Doses used to treat neuropathy that ginkgo increases maximum walk- insulin.10 are 360–480 mg/day. Some experts ing distance and pain-free walking dis- Side effects and drug interactions. have stated that it may take several tance. A recent meta analysis revealed Side effects include breath odor, months to see results with GLA and a significant effect on increased pain- mouth and GI burning or irritation, that, for maximal absorption, it free walking distance.61 The weighted heartburn, flatulence, and rare topical should be taken with food. mean difference was 34 m. reactions. There are cases of sponta- Although the initial data seem Ginkgo was found to improve anti- neous spinal epidural hematoma68 and promising and GLA is relatively depressant-induced sexual dysfunction excessive bleeding.69 benign, its role in treating neuropathic in an open-label trial.62 Ginkgo has Potential drug interactions may complications requires more investiga- been reported to have a beneficial occur if a patient is taking antiplatelet tion. effect on erectile dysfunction by agents, such as warfarin or aspirin.10 improving penile arterial blood flow.66 Additive anticoagulant effects may Ginkgo Biloba Clinical notes. Doses used range occur when combined with herbal Ginkgo biloba is one of the world’s from 120 to 160 mg/day for peripher- products that have antiplatelet activity, oldest living tree species, dating back al vascular disease.10 Ginkgo is admin- such as ginkgo, ginger, and feverfew. more than 200 million years.9 Extracts istered in divided doses, usually two Clinical studies. A recently pub- from dried leaves of younger trees are to three times/day. Administration for lished meta-analysis included trials used in complementary therapies. 6–8 weeks is required to determine from previously published meta analy- Active ingredients include flavonoids benefit. The role for ginkgo in dia- ses and newer trials.70 The results (ginkgo-flavone glycosides) and ter- betes care is still preliminary. revealed that garlic reduced total cho- 205 Diabetes Spectrum Volume 14, Number 4, 2001 lesterol more than did placebo (P < (DHLA).18 ALA and DHLA are both mg three times daily, or placebo for 6 0.01). The result was more modest potent antioxidants and have three months. The other group was given than in previous studies. Mean total distinct actions: 1) free radical scav- daily IV placebo for 3 weeks followed cholesterol decrease was 15.7 mg/dl. enging activity; 2) regeneration of by oral placebo for 6 months. Mean In mild hypertension, a 1994 meta- endogenous antioxidants such as vita- baseline neuropathic impairment analysis indicated that mean systolic min C, vitamin E, and glutathione; scores decreased after 19 days in the blood pressure was 7.7 mmHg lower and 3) metal chelating activity.73 two ALA groups versus placebo (P = than with placebo, and diastolic blood Side effects and drug interactions. 0.02). After 7 months, however, dif- pressure was 5 mmHg lower.71 Serious side effects have not been ferences were not significant between However, only three of the trials were reported. ALA may produce GI side the groups. The authors suggested conducted in hypertensive subjects. effects and possible allergic skin con- that lack of effect in this trial, as Seven of the eight trials in the meta- ditions.74 ALA may potentially result opposed to the earlier trials, might analysis compared garlic to placebo. in additive hypoglycemia when com- have been due to intercenter variabili- Three of the trials found a significant bined with hypoglycemic agents.75 ty in symptom scoring. reduction in systolic blood pressure, Clinical studies. ALA was studied Clinical notes. There is no infor- and four trials showed a significant in the ALADIN (Alpha Lipoic Acid in mation on the prevention of neuropa- reduction in diastolic blood pressure. Diabetic Neuropathy) trials. The first thy with ALA. The clinical signifi- Clinical notes. Although lipid low- trial74 was a randomized, double- cance of the reported improvement in ering has been shown with garlic use, blind, placebo-controlled trial in 260 different parameters of neuropathy in the results are less impressive than type 2 diabetic patients with sympto- the ALADIN studies74,76,77 may be with traditional agents. Although matic peripheral neuropathy. For 3 enhanced quality of life for patients total cholesterol is lowered, there is weeks, patients were administered who experience neuropathy. less evidence for benefit to LDL, placebo or intravenous (IV) ALA Long-term trials are necessary to triglycerides, or HDL cholesterol. (100, 600, or 1,200 mg) daily. Total determine whether ALA slows progres- Garlic may not provide the aggressive symptom scores of neuropathy sion of neuropathy or merely improves lipid lowering that is recommended decreased significantly for all three neuropathy symptoms. The Neurolog- for patients with diabetes. Likewise, ALA doses versus placebo (P < 0.05). ical Assessment of Thioctic Acid in patients with diabetes need more Burning, paresthesia, and numbness Neuropathy Study (NATHAN) is an aggressive antihypertensive effects decreased significantly in patients on ongoing multicenter trial in Europe than garlic may provide. 600 and 1,200 mg/day versus placebo and North America to evaluate the The dose to treat hyperlipidemia (P < 0.05). Pain scores decreased sig- ability of oral ALA to slow progression and hypertension is 600–900 mg/day nificantly only in the 600 mg/day ver- of neuropathy in diabetes.78 in divided doses.10 For fresh garlic, the sus placebo group (P < 0.05). Typical oral doses of ALA are dose is one clove (4 g) taken However, both the 600- and 1,200- 600–800 mg/day.10 Although intra- once/day.10 Dried garlic powder mg doses decreased Hamburg Pain venously administered ALA has been preparations standardized to 1.3% Adjective List scores (P < 0.01 vs. shown to be effective when used in allicin content have been used in stud- placebo). The Neuropathy Disability short-term clinical trials,74 it is not a ies. These preparations should be Score decreased but was significant practical form for supplementation. enteric-coated to prevent breakdown only for the 1,200-mg group versus The American Diabetes Association by stomach acids.10 the placebo group (P = 0.030). does not sanction use of ALA. The second ALADIN trial (ALADIN -Lipoic Acid II) was a randomized, double-blind, Conclusions -lipoic acid (ALA), also known as placebo-controlled 2-year trial in 65 Although biological complementary thioctic acid, is a disulfide compound patients with type 1 or type 2 diabetes therapies have been studied in human synthesized in the liver.72 It functions and polyneuropathy symptoms.76 clinical trials, there are many problems as a cofactor in enzyme complexes Following administration of placebo, with study design, study endpoints, such as pyruvate dehydrogenase, 600 mg, or 1,200 mg/day IV for 5 numbers of patients, and study dura- where it assists in the conversion of days, patients were then randomized to tion. There is insufficient evidence to pyruvic acid to acetyl-coenzyme A in oral placebo, 600 mg, or 1,200 mg/day recommend generalized use for the oxidative metabolism of glucose.72 for 2 years. Mean sural nerve conduc- patients with diabetes. Furthermore, In vitro and animal research indi- tion velocity changes were significant these products have many side effects cates that elevated glucose levels may only for 600 and 1,200 mg versus and may potentially interact with tra- increase free radical-mediated oxida- placebo (P < 0.05). Sural sensory nerve ditional diabetes medications. tion, which in turn is strongly impli- action potential scores decreased signif- cated in the pathogenesis of diabetes- icantly only for 600 mg versus placebo References 73 (P < 0.05). Tibial motor nerve conduc- related neuropathy. Because ALA 1 tion velocity changes were significant Sabo CE, Rush MS, Temple LL: The use of may decrease oxidative stress (in part alternative therapies by diabetes educators. caused by increased blood glucose lev- only for 1,200 mg versus placebo (P < Diabetes Educ 25:945–946;949–950;952–954, els), it may help minimize diabetes 0.05). 956, 1999 ALADIN III77 was a randomized, complications. ALA has been used in 2Gill GV, Redmond S, Garratt F, Paisey R: Germany for decades to treat periph- double-blind, placebo-controlled trial Diabetes and alternative medicine: cause for con- eral neuropathy. in 503 patients with type 2 diabetes. cern. Diabet Med 11:210–213, 1994 One group was given 600 mg IV/day Proposed mechanism of action. 3Huxtable RJ: The harmful potential of herbal ALA is readily converted to its of ALA for 3 weeks, and then subjects and other plant products. Drug Safety 5 (Suppl. reduced form, dihydrolipoic acid were randomized to oral ALA, 600 1):126–136, 1990 206 Diabetes Spectrum Volume 14, Number 4, 2001 4 21 39 Fugh-Berman A: Herb-drug interactions. Lancet Sharma RD, Sarkar A, Hazra DK, Miehra B, Sotaniemi EA, Haapakoski E, Rautio A: Integrative Medicine & Complementary / From Research to Practice 355:134–138, 2000 Singh JB, Sharma SK, Maheshwari BB, Ginseng therapy in non-insulin dependent diabet- Maheshwari PK: Use of fenugreek seed powder ic patients. Diabetes Care 18:1373–1375, 1995 5Hamburger M, Hostettmann K: Analytical in the management of non-insulin-dependent dia- 40 aspects of drugs of natural origin. J Pharm betes mellitus. Nutr Res 16:1331–1339, 1996 Vuksan V, Sievenpiper JL, Vernon YY, Koo Biomed Anal 7:1337–1349, 1989 MSC, Francis T, Beljan-Zdravkovic U, Xu Z, 22Khanna P, Jain SC, Panagariya A, Dixit VP: Vidgen E: American ginseng (Panax quinque- 6 Bonati A: How and why should we standardize Hypoglycemic activity of polypeptide-p from a folius L) reduces postprandial glycemia in nondi- phytopharmaceutical drugs for clinical valida- plant source. J Nat Prod 44:648–655, 1981 abetic subjects and subjects with type 2 diabetes tion? J Ethnopharmacol 32:195–197, 1991 mellitus. Arch Intern Med 160:1009–1013, 2000 23Aslam M, Stockley IH: Interaction between 7 Grant KL: Patient education and herbal dietary curry ingredient (karela) and drug (chlor- 41Cui J, Garle M, Eneroth P, Bjorkhem I: What supplements. Am J Health Syst Pharm propamide). Lancet 1:607, 1979 do commercial ginseng preparations contain? 57:1997–2003, 2000 (Letter). Lancet 344:134, 1994 24Ahmad N, Hassan MR, Halder H, Bennoor 8 Anonymous: Gymnema sylvestre. Altern Med KS: Effect of momordica charantia (karolla) 42Ibanez-Camacho R, Roman-Ramos R: Rev 4:46–47, 1999 extracts on fasting and postprandial serum glu- Hypoglycemic effect of opuntia cactus. Arch 9Facts and Comparisons: The Review of Natural cose levels in NIDDM patients. Bangladesh Med Invest Med 10:223–230, 1979 Res Counc Bull 25:11–13, 1999 Products. St Louis, Mo., Wolters Kluwer, 1999 43Meckes-Lozyoa M, Roman-Ramos R: Opuntia 25 10Jellin JM, Batz F, Hitchens K: Pharmacist’s Raman A, Houston P: Herbal products: gin- streptacantha: a coajutor in the treatment of dia- Letter/Prescribers Letter Natural Medicines seng. Pharm J 255:150–152, 1995 betes mellitus. Am J Chin Med 14:116–118, 1986 Comprehensive Database. Stockton, Calif., 26Vogler BK, Pittler MH, Ernst E: The efficacy of Therapeutic Research Faculty, 1999. Available ginseng, a systematic review of randomized clini- 44Frati A: 2000 medical implications of prickly from www.NaturalDatabase.com cal trials. Eur J Clin Pharmacol 55:567–575, pear cactus. http//www.tamuk.edu/webuser/cac- 1999 tus/cac_med.html (accessed 14 February 2001). 11Shanmugasundaram ER, Panneerselvam C, Samudram P, Shanmugasundaram ERB: Enzyme 27Blumenthal M, Busse WR, Goldberg A, 45Frati-Munari AC, Gordillo BE, Altamirano P, changes and glucose utilisation in diabetic rab- Gruenwald J, Hall T, Riggins CW, Rister RS Ariza CR: Hypoglycemic effect of opuntia strep- bits: the effect of gymnema sylvestre. J (Eds.): The Complete German Commission E tacantha lemaire in NIDDM. Diabetes Care Ethnopharmacol 7:205–234, 1983 Monographs: Therapeutic Guide to Herbal 11:63–66, 1988 Medicines. Klein S., translator. Boston, Mass. 46 12Persaud SJ, Al-Majed H, Raman A, Jones PM: American Botanical Council, 1998 Frati AC, Gordillo BE, Altamirano P, Ariza Gymnema sylvestre stimulates insulin release in CR, Cortes-Franco R, Chavez-Negrete A: Acute 28 vitro by increased membrane permeability. J Yuan CS, Wu JA, Lowell T, Gu M: Gut and hypoglycemic effect of opuntia streptacantha Endocrinol 163:207–212, 1999 brain effects of American ginseng root on brain- lemaire in NIDDM (Letter). Diabetes Care stem neuronal activities in rats. Am J Chin Med 13:455–456, 1990 13Shanmugasundaram ERB, Gopinath KL, 26:47–55, 1998 47 Radha Shanmugasundaram KR, Rajendran VM: Ghannam N: The antidiabetic activity of aloes: 29 Possible regeneration of the islets of Langerhans Ohnishi Y, Takagi S, Miura T, Usami M, Kako preliminary clinical and experimental observa- in streptozotocin-diabetic rats given Gymnema M, Ishihara E, Yano H, Tanigana K, Suno Y: tions. Horm Res 24:288–294, 1986 Effect of ginseng radix on GLUT2 protein con- sylvestre leaf extracts. J Ethnopharmacol 48Muth ER, Laurent JM, Jasper P: The effect of 30:265–279, 1990 tent in mouse liver in normal and epinephrine- induced hyperglycemic mice. Biol Pharm Bull bilberry nutritional supplementation on night 14Shanmugasundaram ERB, Rajeswari G, 19:1238–1240, 1996 visual acuity and contrast sensitivity. Altern Med Baskaran K, Rajesh Kumar BR, Radha Rev 5:164-173, 2000 30Kimura M, Waki I, Chujo T, Kikuchi T, Shanmugasundaram K, Kizar Ahmath B: Use of 49Head K: Natural therapies for ocular disorders, Gymnema sylvestre leaf extract in the control of Hiyama C, Yamazaki K, Tanaka O: Effects of part two: cataracts and glaucoma. Altern Med blood glucose in insulin-dependent diabetes mel- hypoglycemic components in ginseng radix on Rev 6:141–166, 2001 litus. J Ethnopharmacol 30:281–294, 1990 blood insulin level in alloxan diabetic mice and 50Wichtl MW: Herbal Drugs and 15Baskaran K, Kizar B, Ahamath K, Radma on insulin release from perfused rat pancreas. J Pharmacobiodyn 4:410–417, 1981 Phytopharmaceuticals. Bisset NG, Ed. Stuttgart, Shanmugasundaram K, Shanmugasundaram Germany, Medpharm Scientific Publishers, 1994 ERB: Antidiabetic effect of a leaf extract from 31Palmer BV, Montgomery AC, Monteiro JC: Gymnema sylvestre in non-insulin-dependent 51Cignarella A, Nastasi M, Cavalli E, Puglisi L: diabetes mellitus patients. J Ethnopharmacol Ginseng and mastalgia (Letter). Br Med J Novel lipid-lowering properties of vaccinium 30:295–306, 1990 1:1284, 1978 myrtillus L. leaves, a traditional antidiabetic 32 treatment, in several models of rat dyslipidaemia: 16Madar Z: Fenugreek (trigonella foenum- Hopkins M, Androff I, Benninghoff AS: Ginseng face cream and unexplained vaginal a comparison with ciprofibrate. Thromb Res graceum) as a means of reducing postprandial 84:311–322, 1996 glucose levels in diabetic rats. Nutr Rep Int bleeding. Am J Obstet Gynecol 159:1121–1122, 29:1267–1273, 1984 1988 52Pepping J: Alternative therapies: milk thistle: 33 17 Ryu S-J, Chien Y-Y: Ginseng-associated cere- silybum marianum. Am J Health Syst Pharm Raghuram TC, Sharma RD, Sivakumar, Sahay 56:1195–1197, 1999 BK: Effect of fenugreek seeds on intravenous glu- bral arteritis. Neurology 45:829–830, 1995 53 cose disposition in non-insulin dependent diabet- 34Siegel RK: Ginseng abuse syndrome: problems Flora K, Hahn M, Rosen H, Benner K: Milk ic patients. Phytother Res 8:83–86, 1994 with the panacea. JAMA 241:1614–1615, 1979 thistle (silybum marinum) for the therapy of liver 18 disease. Am J Gastroenterol 93:139–143, 1998 Bordia A, Verma SK, Srivastava KC: Effect of 35 Becker BN, Greene J, Evansan J, Chidsey G, 54 ginger (zingiber officinale rosc.) and fenugreek Stone WJ: Ginseng-induced diuretic resistance. Velussi M, Cernigoi AM, De Monte A, Dapas (trigonella foenumgraecum l) on blood lipids, JAMA 276:606–607, 1996 F, Caffau C, Zilli M: Long-term (12 months) blood sugar and platelet aggregation with coro- treatment with an anti-oxidant drug (silymarin) nary artery disease. Prostaglandins Leukot 36Shader RI, Greenblatt DJ: Phenylzine and the is effective on hyperinsulinemia, exogenous Essent Fatty Acids 56:379–384, 1997 dream machine: ramblings and reflections. J Clin insulin need and malondialdehyde levels in cir-

19 Psychopharmacol 5:65, 1985 rhotic diabetic patients. J Hepatol 26:871–879, Patil SP, Niphadkar PV, Bapat MM: Allergy to 1997 fenugreek (trigonella foenum graecum). Ann 37Jones BD, Runikis AM: Interaction of ginseng Allergy Asthma Immunol 78:297–300, 1997 with phenelzine. J Clin Psychopharmacol 55Luper S: A review of plants used in the treat-

20 7:201–202, 1987 ment of liver disease: part 1. Altern Med Rev Sharma RD, Raghuram TC, Sudhakar Rao N: 3:410–421, 1998 Effect of fenugreek seeds on blood glucose and 38Janetzky K, Morreale AP: Probable interaction serum lipids in type 1 diabetes. Eur J Clin Nutr between warfarin and ginseng. Am J Health Syst 56Jamal GA: The use of gamma linolenic acid in 44:301–306, 1990 Pharm 54:692–693, 1997 the prevention and treatment of diabetic neu- 207 Diabetes Spectrum Volume 14, Number 4, 2001 ropathy. Diabet Med 11:145–149, 1994 with associated platelet dysfunction from exces- 77Ziegler D, Hanefeld M, Ruhnau K-J, Hasche sive garlic ingestion: a case report. Neurosurgery H, Lobisch M, Schutte K, Kerum G, Malessa R, 57 Horrobin DF: Essential fatty acids in the man- 2:880–882, 1990 The ALADIN III Study Group: Treatment of agement of impaired nerve function in diabetes. 69 symptomatic diabetic polyneuropathy with the Diabetes 46 (Suppl. 2):S90–S93, 1997 Burnham BE: Garlic as a possible risk for post- antioxidant -lipoic acid: a 7-month multicenter operative bleeding (Letter). Plast Reconstr Surg 58 randomized controlled trial (ALADIN III Study). Jamal GA, Carmichael H: The effect of - 95:213, 1995 linolenic acid on human diabetic peripheral neu- Diabetes Care 22:1296–1301, 1999 70 ropathy: a double-blind placebo-controlled trial. Stevinson C, Pittler MH, Ernst E: Garlic for 78Ziegler D, Reljanovic M, Mehnert H, Greis FA: Diabet Med 7:319–323, 1990 treating hypercholesterolemia, a meta-analysis of Alpha-lipoic acid in the treatment of diabetic randomized clinical trials. Ann Intern Med 59 polyneuropathy in Germany: current evidence The Alpha-Linolenic Acid Multicenter Trial 133:420–429, 2000 Group, Keen H, Payan J, Allawi J, Walker J, from clinical trials. Exp Clin Endocrinol Jamal GA, Weir AI, Bissessar EA, Watkins PJ, 71Silagy CA, Neil HA: A meta-analysis of the Diabetes 107:421–430, 1999 Sampson M: Treatment of diabetic neuropathy effect of garlic on blood pressure. J Hypertens with alpha-linolenic acid. Diabetes Care 12:463–468, 1994 16:8–15, 1993 72Evans JL, Goldfine ID: Alpha-lipoic acid: a Suggested Readings/Websites 60Kleijnen J, Knipschild P: Ginkgo biloba. Lancet multi-functional antioxidant that improves Tyler VE: Herbs of Choice: The Therapeutic Use 340:1136–1139, 1992 insulin sensitivity in patients with type 2 dia- of Phytomedicinals. Binghamton, N.Y., betes. Diabetes Technol Ther 2:401–413, 2000 Pharmacaeutical Products Press, 1994 61Pittler MH, Ernst E: Ginkgo biloba extract for 73 the treatment of intermittent claudication: a meta Giugliano D, Ceriello A, Paolisso G: Oxidative Gruenwald S, Brenoller T, Jaenicke L (Eds.): analysis of randomized trials. Am J Med stress and diabetic vascular complications. PDR for Herbal Medicine. Montvale, N.J., Diabetes Care 19:257–267, 1996 108:276–281, 2000 Medical Economics, 1999 62Cohen AJ, Bartlik B: Ginkgo biloba for antide- 74Ziegler D, Hanefeld M, Ruhnau K-J, Meibner Leung AY, Foster S: Encyclopedia of Common pressant-induced sexual dysfunction. J Sex HP, Lobisch M, Schutte K, Gries FA, The Marital Ther 24:139–143, 1998 ALADIN Study Group: Treatment of sympto- Natural Ingredients Used in Foods, Drugs, and matic diabetic peripheral neuropathy with the Cosmetics. 2nd ed. New York, Wiley, 1995 63 Rowin J, Lewis SL: Spontaneous bilateral sub- anti-oxidant -lipoic acid: a 3-week multicentre dural hematomas with chronic ginkgo biloba randomized controlled trial (ALADIN Study). Newall CA, Anderson LA, Phillipson JD: Herbal ingestion. Neurology 46:1775–1776, 1996 Diabetologia 38:1425–1433, 1995 Medicines: A Guide for Health-Care Professionals. London, Pharmaceutical Press, 64Vale S: Subarachnoid haemorrhage associated 75 Jacob S, Ruus P, Hermann R, Trittschler HJ, 1996 with ginkgo biloba (Letter). Lancet 352:36, 1998 Maerker E, Renn W, Augustin HJ, Dietze GJ, Bisset NG (Ed.): A Handbook for Practice on a 65Rosenblatt M, Mindel T: Spontaneous hyphe- Rett K: Oral administration of RAC- -lipoic acid Scientific Basis. Boca Raton, Fla., Medpharm ma associated with ingestion of ginkgo biloba modulates insulin sensitivity in patients with Scientific Publications CRC, 1996 extract (Letter). N Engl J Med 336:1108, 1997 type-2 diabetes mellitus: a placebo-controlled pilot trial. Free Radic Biol Med 27:309–314, www.nal.usda.gov/fnic/IBIDS 66Shon M, Sikora R: Ginkgo biloba extract in the 1999

therapy of erectile dysfunction. J Sex Ed Ther 76 17:53–61, 1991 Reljanovic M, Reichel G, Rett K, Lobisch M, Schuette K, Moller W, Tritschler HJ, Mehnert H, Laura Shane-McWhorter, PharmD, 67Pareddy SR, Rosenberg JM: Does garlic have The ALADIN II Study Group: Treatment of dia- BCPS, FASCP, CDE, BC-ADM, is an useful medicinal purposes? Hosp Pharmacist Rep betic polyneuropathy with the antioxidant thioc- associate clinical professor in the 8:27, 1993 tic acid (alpha-lipoic acid): a two year multicen- Department of Pharmacy Practice at ter randomized double-blind placebo controlled 68 Rose KD, Croissant PD, Parliament CF, Levin trial (ALADIN II). Free Radic Biol Med the University of Utah in Salt Lake MB: Spontaneous spinal epidural hematoma 31:171–179, 1999 City.

208 Diabetes Spectrum Volume 14, Number 4, 2001