Edited by: Adolfo Andrade-Cetto, National Autonomous University of Mexico, Mexico
Reviewed by: Thomas Efferth, University of Mainz, Germany; Daniel Shriner, National Human Genome Research Institute, USA
*Correspondence: Jose M. Prieto
This article was submitted to Ethnopharmacology, a section of the journal Frontiers in Pharmacology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
It is becoming increasingly evident that patients with diabetes do not rely only on prescription drugs for their disease management. The use of herbal medicines is one of the self-management practices adopted by these patients, often without the knowledge of their healthcare practitioners. This study assessed the potential for pharmacokinetic herb-drug interactions (HDIs) amongst Nigerian adult diabetic patients. This was done through a literature analysis of the pharmacokinetic profile of their herbal medicines and prescription drugs, based on information obtained from 112 patients with type-2 diabetes attending two secondary health care facilities in Nigeria. Fifty percent of the informants used herbal medicines alongside their prescription drugs. Worryingly, 60% of the patients taking herbal medicines did not know their identity, thus increasing the risk of unidentified HDIs. By comparing the pharmacokinetic profile of eight identified herbs taken by the patients for the management of diabetes against those of the prescription drugs, several scenarios of potential HDIs were identified and their clinical relevance is discussed. The lack of clinical predictors points toward cultural factors as the influence for herb use, making it more difficult to identify these patients and in turn monitor potential HDIs. In identifying these possible interactions, we have highlighted the need for healthcare professionals to promote a proactive monitoring of patients' use of herbal medicines.
Herbal medicines are becoming an increasing component of Western pharmacotherapy mostly due to the perception that “natural” means “safe” (Brantley et al.,
With the use of herbal medicine becoming more widespread, a lot of focus has been given to evaluating them for possible herb-drug interactions (HDIs), particularly pharmacokinetic interactions since pharmacodynamic interactions are often easier to identify and monitor. Efforts have also been made to collate existing information about herbal medicines as a guide to healthcare practitioners, and as a means of forestalling potentially clinically relevant HDIs (Ulbricht et al.,
The implications of clinically relevant HDIs can have far-reaching effects for most African populations, such as Nigeria, with an inadequate healthcare system (Anyika,
A fieldwork study was conducted in August and September 2012 in the Pharmacy departments of two secondary health care facilities—the General hospital in Lagos Island and Central hospital in Benin City—two urban cities in the southern part of Nigeria. The study-including the proposed questionnaire to be administered—was approved by the ethics committee of UCL (Ethics Application 4124/001); and prior consent was also obtained from the ethics committee of both hospitals.
During the study period, every prescription that was identified as having one or more oral hypoglycaemic drugs during the “prescription assessment stage” was assigned to the interviewer for drug dispensing and counseling and possible inclusion of the patient in the study. Each patient was thereafter given a brief description of the study. However, only patients who gave their verbal consent to be asked additional questions relating to their use of herbal medicines, based on a semi-structured questionnaire were included. These questions were asked while drugs were being dispensed and medication counseling was given to the patients afterwards.
Information obtained from the questionnaire include demographic background of respondents including age and sex, information on the identity of the herbs and prescription drugs used (where available) either for diabetes or other co-morbidities as well as the monthly cost of prescription. Herbal medicines as described in this study include the following: supplements containing extracts of medicinal plants, un-standardized herbal preparations from herbalists, herbal preparations made by the patients themselves using plant parts purchased in herbal markets or collected from their own gardens.
No samples of herbal medicines were collected as part of the study. The identity of the herbal medicines used, were provided by the patients either as local names or its common name in English. Various literature sources were consulted to determine the botanical names of plants from their given local names. Verification of their local names were done by Mr. Ibrahim Muazzam, an ethnobotanist and staff of the department of medicinal plant research and traditional medicine, National Institute for Pharmaceutical Research and Development (NIPRD) and by staff of the Forestry Research Institute of Nigeria (FRIN) in Ibadan and assignment of the correct botanical names was done based on
Assessment of the potential HDIs was done by analysing the prescription drugs and herbal medicines used by the patients against published information about any known pharmacokinetic effects to identify possible overlapping effects that might pose a risk.
The descriptive and inferential statistics to determine measures of frequency and relationship between variables were carried out using IBM SPSS statistical package (version 22, IBM U.S.A.). Chi square test was used to test the association of herb use with patients' background characteristics, such as gender, age groups, presence of co-morbidities, and cost of monthly prescription. A significant relationship between variables was identified if
A total number of 112 patients with type-2 diabetes, 37 from Central hospital Benin-city and 75 from General hospital Lagos gave their consent and were interviewed for the study. A summary of the background characteristics of the interviewed patients is given in Table
Gender | Male | 43 (38.4) | 23 (20.5) | 20 (17.9) |
Female | 69 (61.6) | 33 (29.5) | 34 (30.4) | |
Age groups (years) | ≤ 30 | 1 (0.9) | 0 (0.0) | 1 (0.9) |
1–45 | 9 (8.1) | 4 (3.6) | 5 (4.5) | |
46–60 | 39 (34.8) | 19 (16.9) | 20 (17.9) | |
61–75 | 57 (50.9)6 | 30 (26.8) | 25 (22.3) | |
>75 | (5.4) | 3 (2.7) | 3 (2.7) | |
Presence of co-morbidities | Hypertension + Arteriosclerosis + Arthritis | 1 (0.9) | 0 (0.0) | 1 (0.9) |
Hypertension + High cholesterol + Arthritis | 6 (5.4) | 4 (3.6) | 2 (1.8) | |
Hypertension + High cholesterol + Arteriosclerosis | 1 (0.9) | 1 (0.9) | 0 (0.0) | |
Hypertension + High cholesterol | 17 (15.2) | 11 (9.8) | 6 (5.4) | |
Hypertension + Arteriosclerosis | 2 (1.8) | 1 (0.9) | 0 (0.0) | |
Hypertension + Arthritis | 4 (3.6) | 2 (1.8) | 1 (0.9) | |
Hypertension + Neuropathy | 2 (1.8) | 1 (0.9) | 1 (0.9) | |
Hypertension | 46 (41.1) | 22 (19.6) | 24 (21.5) | |
High cholesterol | 1 (0.9) | 1 (0.9) | 0 (0.0) | |
None | 30 (26.8) | 13 (11.6) | 17 (15.2) | |
Monthly prescription cost (Naira) |
< 5000 | 23 (20.5) | 9 (8.0) | 12 (10.7) |
5001–10,000 | 27 (24.1) | 17 (15.2) | 10 (9.9) | |
10,001–15,000 | 10 (8.9) | 4 (3.6) | 6 (5.4) | |
>15,000 | 2 (1.8) | 0 (0.0) | 2 (1.8) |
More than 70% of the patients had one or more chronic co-morbidity alongside their diabetes, the most common of which was hypertension. The other identified co-morbidities were arthritis, arteriosclerosis, hypercholesterolemia and neuropathy (Table
Approximately 50% of all patients used herbal medicines alongside their prescription drugs, and this percentage was the same for both males and females. Almost 60% of those who used herbal medicines were older patients aged between 61 and 75 years (Table
Given that the inherent characteristics of the individual (age and sex) were not a predictor of herb use, we hypothesized that the presence of one or more chronic co-morbidity as well as a high monthly prescription cost could be an incentive for patients to use herbal medicines as evidenced by the study carried out by Nahin et al. (
In trying to understand the divergent relationship between number of co-morbidities and monthly cost of prescription with herb use, it was observed that the prescription cost was not necessarily dependent on the number of co-morbidities the patient was being treated for, but on the medications on the patient's prescription. For instance, a patient with only diabetes may have been prescribed gliclazide by his/her doctor, which is more expensive than other types of sulphonylureas. On the other hand, another patient having diabetes alongside other co-morbidities might be prescribed a cheaper sulphonylurea, such as glibenclamide as well as other non-expensive medications for their co-morbidities resulting in lower prescription costs. Thus, a patient with more co-morbidities may not necessarily have higher prescription costs. At the same time, based on the results of the study, patients with more drugs on their prescription due to co-morbidities or those with the need to spend more money on their prescription drugs were not more likely to use herbs than those without. As a result, the presence of diabetic complications wasn't identified as a predictor for herb use among patients with type-2 diabetes.
A summary of the drugs used in the pharmacological management of diabetes in the interviewed patients is given in Table
Glibenclamide only | 1 | 1 |
Insulin only | 4 | 2 |
Insulin + Gliclazide | 1 | 0 |
Metformin + Acarbose + Glibenclamide | 2 | 1 |
Metformin + Acarbose + Gliclazide + Insulin | 1 | 1 |
Metformin + Glibenclamide | 42 | 25 |
Metformin + Gliclazide | 10 | 8 |
Metformin + Gliclazide + Pioglitazone | 3 | 3 |
Metformin + Glimepiride | 19 | 16 |
Metformin + Glimepiride + Pioglitazone | 2 | 2 |
Metformin + Insulin | 3 | 2 |
Metformin + Insulin + Glibenclamide | 2 | 1 |
Metformin + Pioglitazone | 1 | 1 |
Metformin + Vidagliptin | 1 | 1 |
Metformin only | 18 | 17 |
Less than 10% of the patients interviewed used insulin either singly or in combination with an oral hypoglycaemic agent for their diabetes management. This isn't surprising as despite the proven efficacy of insulin in blood glucose control, its level of compliance and/or acceptance amongst patients with type-2 diabetes is still low due to its relatively higher cost compared to oral hypoglycaemic drugs and unpreferred mode of administration. Additional factors which are unique to Nigeria and other developing countries include limited availability and unreliable storage facilities, given that constant power supply isn't guaranteed. Thus, despite the decreased risk of pharmacokinetic interactions with insulin, patients are still unlikely to use it as an alternative.
Out of the 56 patients who admitted to using herbal medicines alongside prescription drugs for their disease management, < 40% were aware of their identity. For these patients, they either grew the herbs themselves in their gardens or they purchased them from herbal markets. The remaining 60% were unaware of the identity of the herbal medicine used, primarily because these were unlabeled preparations (often multi-component) given to them by herbalists, whose constituents were not disclosed. A number of patients however said they were not very interested in finding out the identity as long as the preparation was effective. They considered herbal medicines to be relatively “safe” compared to conventional medicines. For this cohort of patients, the risk of HDIs would be greater as they wouldn't be able to provide any information to their physicians/healthcare practitioners that can enable them provide appropriate advice with regards to its efficacy and/or safety.
Twelve medicinal plants were identified among the patients who knew the composition of the herbal medicines which they used for their disease management (Table
13 (11.6) | Inhibits P-gp efflux activity (Oga et al., |
– | – | |
5 (4.5) | Dose dependent increase in AST and ALT levels (Ajibade et al., |
|||
3 (2.7) | Polyvalent cations in the plant form non-absorbable complexes with certain drugs (Nwafor et al., |
– | – | |
3 (2.7) | Inhibits the P-gp efflux activity (Chieli et al., |
Inhibits CYP 1A1/2 and 3A4 activities in rat liver microsomes (Rodeiro et al., |
– | |
2 (1.8) | – | Inhibits CYP 3A4/5/7 enzymes to different extents (Agbonon et al., |
– | |
1 (0.9) | – | – | – | |
1 (0.9) | Water soluble fractions inhibits metformin absorption |
– | – | |
1 (0.9) | – | – | – | |
1 (0.9) | – | – | Pharmacotoxic effects of neem oil in lungs and CNS (Gandhi et al., |
|
1 (0.9) | – | – | – | |
1 (0.9) | – | Elevated AST, ALT, and GSH levels (Kouitcheu Mabeku et al., |
Hepatotoxic effects (Fakeye et al., |
|
1 (0.9) | – | – | Hepatotoxic at high doses due to punicalagin (Lin et al., |
|
36 (31.9) | NA | NA | NA | |
14 (12.4) | NA | NA | NA |
For all of the 12 identified plants, pharmacological evidence validating their traditional use in diabetes management have been published. As shown in Table
Table
Glibenclamide | |||||||||||
Metformin | AE | B | |||||||||
Pioglitazone | |||||||||||
Sitagliptin | |||||||||||
Vidagliptin | |||||||||||
Amlodipine | |||||||||||
Indapamide | |||||||||||
Losartan | |||||||||||
Nifedipine | |||||||||||
Atorvastatin | |||||||||||
Clopidogrel | |||||||||||
Tramadol |
Concurrent use of
With the above information, a practical assessment of potential HDI is herein carried out for one of the interviewed respondents “X,” a 41 year old lady who was taking the squeezed extract of
With this knowledge of the pharmacokinetics of her herbal medicines as highlighted in Table
The increased use of herbal medicines by patients alongside their prescription drugs has raised concerns for HDIs, which may be clinically relevant. By introducing a new and different approach to the results of our field study, we have carried out an investigation into the possible risks of pharmacokinetic HDIs amongst patients with diabetes in Nigeria; as well as a practical example with one patient of how such an assessment can be carried out given available information. One of the key findings of our study was that over 50% of diabetic patients in Nigeria use herbal medicines alongside their conventional drugs for their disease management, which highlights the large number of patients at risk of HDI and the need for such assessments. More worrying is that of this high number, more than 60% are unaware of the identity of the herbal medicines being taken which in turn highlights a complete ignorance of the risks of HDIs.
None of the observed variables—age, gender, cost of disease management, and presence of co-morbidities was identified as a predictor of herbal medicine use; and therefore could not be used as a means of easily identifying patients at risk of HDIs. Based on the study carried out by Nahin et al. (
Previous studies have indeed shown that the use of herbal medicines by an individual is very closely linked to his/her cultural health beliefs (van Andel and Westers,
An analysis of the drugs commonly prescribed to patients with diabetes show that most of these are substrates or modulators of known pharmacokinetic parameters commonly implicated in drug interactions. Studies have shown that the risk of harmful drug interactions increases with the number of drugs given to a patient (Fakeye et al.,
We recently conducted an extensive review of medicinal plants used in diabetes management in Nigeria. One of the findings of the study was the paucity of information on investigations into pharmacokinetic interactions of commonly administered herbal medicines, compared to experiments validating its pharmacological activity (Ezuruike and Prieto,
We have also shown how a true assessment of the risk of HDIs can be possible with the knowledge of a patient's herbal medicine. Healthcare practitioners who are aware of their patients' use of herbal medicines and its identity can more easily monitor them for any possible interactions using available information, such as that presented in Tables
It is important to mention here that the availability of reliable data is required to adequately determine the HDI liability of herbal medicines. As with the case of GFJ and losartan above, this was validated by evidence from the clinical study. However, data obtained
Unfortunately, the high percentage of users of unidentified herbal medicines due to non-disclosure by herbalists makes the task of therapeutic monitoring of herbal medicines difficult, an issue that is unlikely to be immediately tackled. This is because traditional medicine practitioners/herbalists who prepare the herbal preparations themselves insist that it is a “family secret” of which a disclosure will rob them of their business (Ogbera et al.,
Given the increasing prevalence of herbal medicine use alongside prescription medicines for disease management, this is the first time a study has been carried out specifically looking to identify potential HDIs amongst diabetic patients in Nigeria. We identified a high percentage (50%) of patients who use herbal medicines, and by so doing are exposed to a number of pharmacokinetic interactions. The lack of clinical predictors point toward non-measurable cultural factors influencing the use of herbal medicines by Nigerian adult diabetic patients. The identified pharmacokinetic HDIs are therefore highlighted here as a template for HDI assessment as well as to facilitate a more proactive monitoring by healthcare professionals.
UE researched data and wrote the manuscript. JP contributed to the discussion and reviewed/edited the manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
UE is grateful to the Commonwealth Scholarship Commission in the UK for the award of a DFID (Department for International Development) sponsored Award number NGCS-2010-306 and to the Rick Cannell trust for the award of a travel grant to carry out the study.
The Supplementary Material for this article can be found online at:
Alanine transaminase
Aspartate transaminase
Cytochrome P450
Glutathione
Glutathione-S-transferases
Herb-drug interactions
P-glycoprotein.