How do we prevent the burden of extremely harmful and clinically nonbeneficial drug–drug interactions among chronic kidney disease patients?

Abstract

The chronic intake of different medications by chronic kidney disease (CKD) patients predisposes them to extremely harmful and clinically nonbeneficial drug–drug interactions (DDIs) which can ultimately lead to increase in morbidity, mortality, healthcare cost, and frequency and length of hospitalization. This produces a negative deteriorating and counter-efficient outcome on the health, quality of life and treatment response of these patients. This was an 18-month prospective descriptive study that reviewed the medical case records of consented adult CKD patients attending the Nephrology medical outpatient clinic of a Nigerian Tertiary Healthcare Centre from January 2015 to June 2016. The Medscape drug reference database was used to evaluate patients’ medications for extremely harmful, clinically nonbeneficial DDIs. This study involved 123 consented adult CKD patients comprising of 82 (66.67%) males and 41 (33.33%) females with a mean age of 53.81 ± 16.03 years. In this study, the prevalence of extremely harmful, clinically nonbeneficial DDIs (type D or type X interaction categories only) was 24.4%, while the overall prevalence for all the observed DDIs was 95.9%. The most frequent extremely harmful, clinically nonbeneficial DDIs in this study was between α-methyldopa and metoclopramide: 16 (0.9%) interactions in eight (6.5%) patients. Furthermore, α-methyldopa decreases the antiemetic effects of metoclopramide by pharmacodynamics antagonism at the chemoreceptor trigger zone site D₂-receptors (type X; pharmacodynamics). In addition, metoclopramide decreases the level of α-methyldopa by inhibition of gastrointestinal tract (GIT) absorption, as this applies to only oral formulations of both agents (type D; pharmacokinetic). The occurrence and burden of extremely harmful, clinically nonbeneficial DDIs is significantly high among these CKD patients. There is also a critical need to minimize the number of prescribed medications for these patients in order to optimize their care.

Keywords

Drug–drug interactions chronic kidney disease extremely harmful clinically nonbeneficial

Introduction

Drug–drug interaction (DDI) can be defined as an appreciably harmful or beneficial process whereby the pharmacological effects of a drug (or object drug) is directly or indirectly influenced and modified by the presence of another drug (or precipitant drug) which can result in either treatment failure (antagonistic interaction) or drug-induced toxicity (synergistic/additive interaction).^1,2 DDIs have also been observed to be one of the most frequently occurring challenges that may alter the pharmacokinetic and pharmacodynamics profiles of drugs. They have the potential of altering the overall therapeutic response clinically. DDIs can be pharmacokinetic, pharmacodynamics, or pharmaceutical in nature.¹

Pharmaceutical interactions occur when two drugs are given together concurrently in the same infusion solution or when a drug reacts with the infusion solution (e.g. between intravenous (IV) ceftriaxone and IV calcium-containing solutions such as Ringer’s lactate, calcium gluconate, or Hartmann’s solution). While it is necessary to be aware of this type of interaction, it is relatively uncommon in clinical practice.³

Pharmacodynamics interactions are those interactions where the effects of a drug (or object drug) are changed by the presence of another drug (or precipitant drug) at its site of action, without any alteration in the concentrations of both drugs. Sometimes one drug competes directly with another for a particular receptor, but often the reaction is more indirect and involves the interference with physiological mechanisms, making pharmacodynamics interactions more difficult to classify than pharmacokinetic interactions. In pharmacodynamics interaction, the process of interaction may involve the precipitant drug altering the sensitivity or responsiveness of tissues to the object drug by having the same (agonistic) or a blocking (antagonistic) effect; these effects usually can occur at the membrane surface receptor level but may also occur intracellularly.⁴

On the other hand, pharmacokinetic interactions may be defined as those interactions in which the disposition of a drug (or object drug) is altered by the presence of another drug (or precipitant drug). As a result, the effect of the object drug is either diminished or increased. Pharmacokinetic interactions are divided into those that affect absorption, distribution, plasma protein binding, metabolism, and excretion. It therefore changes the concentration/amount and the persistence of available drug at the receptors’ sites. Pharmacokinetic interactions inevitably alter the magnitude and duration of drugs’ effects, and these effects are either predicted based on our scientific knowledge of these individual drugs or detected by therapeutic monitoring of these drugs’ concentrations in the plasma or at the specific site of action if feasible and in addition checking out for obvious or detectable clinical signs.⁵

Extremely harmful, clinically nonbeneficial DDIs are those interactions that belong to the level of clinically significant categories of either major/serious interactions (type D interactions) or contraindication/avoid drug combination (type X interactions) in which the risk of adverse drug effects (ADEs) associated with the concomitant use of these two drugs usually outweighs their benefits in the patient after each individual patient-specific assessment has been done.

Chronic kidney disease (CKD) can be defined as a progressive and irreversible deterioration in the renal function of an individual over a period of at least 3 months regardless of the underlying aetiology.^1,6 In Nigeria and worldwide, poorly controlled chronic hypertension, poorly controlled diabetes mellitus, and human immunodeficiency virus (HIV)-associated nephropathy among others are the common causes of CKD.^3,4,7 CKD is associated with rising incidence and prevalence, high cost of treatment, and poor outcomes. There is evidence to suggest that early in the course of CKD, appropriate interventions may slow down its progression or completely halt the progression of the disease. Despite this, many patients with CKD still present late to the nephrologists, so that at the time of initial patient assessment, all that can be offered is preparation for renal replacement therapy.^{1,3

–6,8,9} This is particularly so in resource-poor setting where, among several other factors, the lack of awareness, traditional beliefs about the cause and nature of the disease, the need to pay out-of-pocket for healthcare, and shortage of specialists combine to promote inappropriate healthcare-seeking behavior and late presentation to the nephrologist. Usually, in CKD, there is a progressive loss of kidney function occurring over several months to years and is characterized by the gradual replacement of normal kidney architecture with fibrous tissue.^3,5 The individual with the end-stage renal disease (ESRD) can no longer adequately excrete waste products, regulate acid–base balance, and maintain sodium, potassium and water homeostasis. There are other associated complications such as anemia, hyperkalemia, metabolic acidosis, inability to maintain calcium balance (hypocalcemia), erectile dysfunction, hyperphosphatemia, and increased cardiovascular disorders such as hypertension, stroke, ischemic heart diseases, cardiac arrhythmias, and sudden cardiac arrest, which can lead to increase in mortality.^1,3

–9 Patients with CKD stages 1–3 (estimated glomerular filtration rate (eGFR) > 30 mL/min/1.73 m²) are generally asymptomatic. Typically, it is not until stages 4–5 (eGFR < 30 mL/min/1.73 m²) that endocrine and metabolic derangements or disturbances in water and electrolyte balance become clinically manifest. The early stages of CKD are often completely asymptomatic, despite the accumulation of numerous metabolites. Serum urea and creatinine concentrations are measured in CKD since methods for their determination are available and a rough correlation exists between serum urea and creatinine concentrations and symptoms. These substances are, however, in themselves not particularly toxic.^1,3

–6,8 The nature of the metabolites that are involved in the genesis of symptoms is unclear. Such metabolites must be products of protein catabolism (since dietary protein restriction may reverse symptoms associated with CKD) and many of them must be of relatively small molecular size (since hemodialysis employing membranes which allow through only relatively small molecules improves symptoms). Little else is known with certainty. Symptoms are common when the serum urea concentration exceeds 40 mmol/L, but many patients develop uremic symptoms at lower levels of serum urea.^1,3

–9

Hypertension and diabetes mellitus are the commonest causes of end-stage renal disease (ESRD) worldwide; therefore, control of high blood pressure and optimization of blood glucose level are essential in delaying and retarding CKD progression.^3

–7 These necessitate the use of several medications to improve the quality of life of these patients and slowing down the progression of early CKD to full-blown ESRD. Although efficacy, compliance, and economic factors are considered in selecting drugs for treating patients with CKD, these patients are usually on polypharmacy with attendant risk for drug-related problems such as DDIs and adverse drug reactions (ADRs).^4
–6 Polypharmacy can be defined as the concurrent use of more than five different medications by a patient for the treatment of a particular disease or group of diseases.^2,10 As the disease progresses, the number of drugs prescribed usually increases substantially; sometimes, up to 10 or more different drugs are ingested daily by the time a patient is at CKD stage 5 (eGFR < 15 mL/min/1.73 m²).^2,4,5,8,10 The influences of the disease on drugs’ pharmacokinetic and pharmacodynamics mechanism increase the risk for drug-related problems such as DDIs and ADRs; these further compound the problems of patients with CKD and may further increase morbidity, mortality, healthcare cost, length and frequency of hospitalization, among them.^11

–14 ADRs can be defined as an undesirable effect of a drug beyond its anticipated therapeutic action occurring during clinical use or any noxious and unintended response to a drug that occurs at doses used in humans for prophylaxis, diagnosis, or therapy of disease.^15,16 ADRs have been classified as mild (laboratory abnormality or symptom not requiring treatment), moderate (laboratory abnormality or symptom requiring treatment or admission to hospital or resulting in nonpermanent disability), severe (laboratory abnormality or symptom that was life-threatening or resulted in permanent disability), and fatal (when it resulted in patient death).^16,17,26

DDIs are major clinical problems, accounting for 2–6% of all hospital admissions with estimated annual cost to the National Health Service of about £500 million in the United Kingdom.¹⁵ In a meta-analysis of 39 prospective studies from hospitals in the United States, it was shown that DDIs ranked fourth to sixth as a leading cause of death.¹⁶ Recent studies in the United States have also indicated that DDIs also increase the cost of treatment, length and frequency of hospitalization.^16,17 The prevalence of DDIs in CKD patients from previous reports ranged between 56.9% and 89.1%.^18

–23 This relatively high prevalence is related to the polypharmacy that is involved in the management of this condition. It is associated with increased morbidity, mortality, healthcare cost, and length and frequency of hospitalization.^2,9,10,24,25 These added burdens of extremely harmful, clinically nonbeneficial DDIs and their consequent hazardous adverse effects in CKD patients are preventable because of their predictable nature. Various software are available that can detect DDIs such as British National Formulary, Drug Interaction Facts, Medscape, Epocrates, Lexi-Interact, Harmavista and Stockley’s Drug Interaction Checker.^27,29

CKD is a major public health problem due to its increasing incidence, prevalence, and associated high burden. The global prevalence of CKD is estimated to be 11–13%.^15
–17,26 The prevalence of CKD in Nigeria varied between 11.4% and 18.8% from both community- and hospital-based studies.^1,3,5,6,8 Cardiovascular disease burdens in CKD patients are high and associated with increased frequency of hospitalization, morbidity, mortality, length of hospital stay, and healthcare cost.^7,18
–20 Cardiovascular risk factors such as hypertension, diabetes mellitus, anemia, calcium–phosphate abnormalities, hyperuricemia, and left ventricular hypertrophy are highly prevalent in CKD.^7,21
–23 These are largely responsible for cardiovascular disease burden and some complications in CKD patients. Management of these cardiovascular diseases and risk factors is important in retarding the progression of CKD and reducing its associated mortality.^9,24,25 This involves the use of multiple drugs’ combination therapy in the management of CKD and its attendant complications; hence, polypharmacy is often practiced.¹⁰ The consequences of polypharmacy include poor patients’ medication adherence due to high pill burden, increased cost of healthcare, and most importantly, drug-related problems such as DDIs and ADRs which can have deleterious effects on patients’ health.^2,10,27 DDIs may result from a number of processes, which are related to either pharmacokinetic or pharmacodynamics properties of the drugs. The consequences of DDIs include increase risk for toxicity or decrease in the efficacy of treatment. The clinical pharmacologists and clinical pharmacists also have major role to play in preventing drug-related problems such as DDIs and ADRs associated with polypharmacy by thoroughly and regularly evaluating prescriptions for CKD patients before rational dispensing.^7,28,29 Therefore, integrated professional interaction should be encouraged between nephrologists, clinical pharmacologists, clinical pharmacists and nephrology nurses in order to optimize CKD patients’ care. Vigilance by healthcare workers such as nephrologists, clinical pharmacologists, clinical pharmacists and nephrology nurses in detecting, diagnosing, and reporting drug-related problems particularly in at-risk individuals such as CKD patients is vital for continued drug safety monitoring.^30

–33

This study was designed to unravel the pattern of extremely harmful and clinically nonbeneficial DDIs among CKD patients attending the nephrology clinic of a Nigerian Tertiary Healthcare Centre. To highlight the importance of prevention, early identification, correction, and monitoring of drug interactions with prompt treatment of any associated ADRs that may occur. This will create awareness concerning the occurrence and burden of extremely harmful, clinically nonbeneficial DDIs among CKD patients in renal practice and will bring to limelight the need to regularly evaluate the prescriptions of CKD patients for medication-related problems such as DDIs and ADRs. It will also help to manage any associated complications and comorbidities appropriately.

Methods

This was a prospective, descriptive study carried out at the Nephrology Clinic of a Tertiary Kidney Care Centre, University of Medical Sciences, Ondo City, Ondo State, Nigeria. The hospital receives referral from within and outside the state. One hundred and twenty-three (123) consented adult CKD patients who were being managed at the center over an 18-month period between January 2015 and June 2016 were recruited for the study. Patients below the age of 18 years, those being managed for acute kidney injury, and adult CKD patients who did not grant their informed consent were excluded from the study. The medical case records of all the adult CKD patients were retrieved after a verbal informed consent had been obtained from each of them, and the following information was extracted using a pro forma: sociodemographic data; stage of CKD; number and list of medications at the time of last clinic visit for outpatients and at the time of discharge for those who received inpatient care; number and list of comorbidities such as hypertension, diabetes mellitus, obesity, heart failure, HIV infection, and stroke. The serum creatinine level was used to calculate the eGFR using CKD Epidemiology Collaboration formula, and CKD staging was done using eGFR as follows: stage 1 (eGFR of ≥90 mL/min with evidence of kidney damage), stage 2 (eGFR of 60–89 mL/min with or without evidence of kidney damage), stage 3 (eGFR of 30–59 mL/min with or without evidence of kidney damage), stage 4 (eGFR of 15–29 mL/min with or without evidence of kidney damage), and stage 5 (eGFR <15 mL/min with or without evidence of kidney damage).²⁸ The Medscape drug reference database was used to evaluate patients’ medication regimen for extremely harmful, clinically nonbeneficial DDIs. Medscape drug reference database has been reported to show the best results in precision analysis.²⁹ The Medscape drug reference database system provides accurate information about the risk, type, mechanism, and pattern of DDIs. It also gives recommendation(s) on how to prevent and manage DDIs if they occur. The software identifies and classifies DDIs according to their level of clinical significance in categories. These categories are types A, B, C, D, and X:

Type A: no known interaction

Type B: minor or mild interaction

Type C: moderate or significant interaction

Type D: major or serious interaction

Type X: contraindication or avoid drug combination

Extremely harmful, clinically nonbeneficial DDIs are those interactions that belong to the level of clinically significant categories of either major/serious interactions (type D interactions) or contraindication/avoid drug combination (type X interactions) in which the risk of ADEs associated with the concomitant use of these two drugs usually outweighs their benefits in the patient after each individual patient-specific assessment has been done. The classification of DDIs according to their level of clinical significance in categories is shown in Table 1.

Table 1.

Classification of drug–drug interactions according to their level of clinical significance.

Risk rating	Action	Description
A	No known interaction	Data have demonstrated neither pharmacodynamics nor pharmacokinetic interactions between the specified agents.
B	Minor/mild interaction—no action needed	Data demonstrate that the specified agents may interact with each other, but there is little or no evidence of clinical concern resulting from their concomitant use.
C	Moderate/significant interaction—monitor therapy closely	Data demonstrate that the specified agents may interact with each other in a clinically significant manner. The benefits of concomitant use of these two medications usually outweigh the risks. An appropriate monitoring plan should be implemented to identify potential negative effects. Dosage adjustments of one or both agents may be needed in a minority of patients.
D	Major/serious interaction—consider therapy modification or use alternatives	Data demonstrate that the two medications may interact with each other in a clinically significant manner. A patient-specific assessment must be conducted to determine whether the benefits of concomitant therapy outweigh the risks. Specific actions must be taken in order to realize the benefits and/or minimize the toxicity resulting from concomitant use of the agents. These actions may include aggressive monitoring, empiric dosage changes, or choosing alternative agents.
X	Contraindication—avoid drug combination	Data demonstrate that the specified agents may interact with each other in a clinically significant manner. The risks associated with concomitant use of these agents usually outweigh the benefits. These agents are generally considered contraindicated.

The mean for all DDIs observed per prescription was obtained by dividing the total number of all DDIs observed by the total number of all prescriptions issued.

The mean for all extremely harmful, clinically nonbeneficial DDIs observed per patient belonging to these extremely harmful interaction categories was obtained by dividing the total number of all extremely harmful, clinically nonbeneficial DDIs observed by the total number of all patients belonging to these extremely harmful interaction categories.

The mean for all extremely harmful, clinically nonbeneficial DDIs observed per prescription was obtained by dividing the total number of all extremely harmful, clinically nonbeneficial DDIs observed by the total number of all prescriptions issued.

Ethical clearance for the study was obtained from the Health Research Ethical Committee of the Tertiary Kidney Care Centre. The Ethical Clearance/Protocol Research Number issued for the study was UNIMED/KCCO/NOV 2014/0007. In addition, a verbal informed consent was obtained from each of the adult CKD patients whose medical case records were used, while the medical case records for those who did not grant their informed consent were excluded from the study. Consent was sought from patient’s relative where the patient had impaired level of consciousness. Participants’ confidentiality was respected and maintained by ensuring that no unauthorized person had access to the information on the questionnaires, no information could be traced to the respondents (as coding system was used for the questionnaires instead of writing the patients’ names on them), and no unauthorized use of information was made. Those who were at risk of extremely harmful DDIs were monitored closely, and feasible alternative prescriptions were made where the concomitant risk and adverse effects outweigh the benefits.

Data collected were encoded and analyzed using the Statistical Package for the Social Sciences (SPSS) Version 17 (released 2008; SPSS Incorporations, Chicago, Illinois, USA). Results were expressed as mean ± standard deviation or using frequency and percentage values where necessary. The t-test and chi-square test were used to compare means and proportions, respectively. The level of statistical significance was set at P<0.05.

Results

This study involved 123 consented adult CKD patients comprising of 82 (66.67%) males and 41 (33.33%) females. The mean age of the study participants was 53.81±16.03 years. Forty-eight (39.0%) patients were between 20 and 49 years, 52 (42.3%) patients were between 50 and 69 years while the remaining 23 (18.7%) patients were 70 years and older (see Table 2).

Table 2.

Characteristics of the study population.

Characteristics	F (%)/mean ± SD
Gender
Male	82 (66.67%)
Female	41 (33.33%)
Mean age (years)	53.81 ± 16.03
Age-group
20–49 years	48 (39.0%)
50–69 years	52 (42.3%)
≥70 years	23 (18.7%)
Level of education
No formal education	16 (13.0%)
Primary	18 (14.6%)
Secondary	36 (29.3%)
Tertiary	53 (43.1%)
CKD stage
1	1 (0.8%)
2	2 (1.6%)
3	19 (15.5%)
4	15 (12.2%)
5	86 (69.9%)
Mean comorbidities (diseases)	2.33 ± 1.09
Number of comorbidities (diseases)
0	4 (3.52%)
1	23 (18.70%)
2	45 (36.59%)
3	36 (29.27%)
4	11 (8.94%)
5	3 (2.44%)
6	1 (0.8%)
Specific comorbidities
Hypertension	103 (83.70%)
Diabetes mellitus	39 (31.70%)
Obesity	24 (19.51%)
Heart failure	11 (8.90%)
Obstructive uropathy	8 (6.50%)
HIV	7 (5.70%)
Stroke	5 (4.10%)
Adult polycystic kidney disease	5 (4.10%)
HBV infection	5 (4.10%)
Mean prescribed medication per patient	10.06 ± 3.97
Number of prescribed medications (drugs)
≤5	18 (14.6%)
6–10	43 (35.0%)
11–15	51 (41.5%)
≥16	11 (8.9%)
Form of nephrological interventions
Maintenance dialysis	66 (53.66%)
Conservative care	53 (43.09%)
Renal transplantation	4 (3.25%)
Mean BMI (Kg/m²)	25.71 ± 5.09
BMI (Kg/m²)
Underweight	5 (4.10%)
Normal	55 (44.72%)
Overweight	39 (31.71%)
Mild (grade 1) obesity	18 (14.63%)
Moderate (grade 2) obesity	5 (4.10%)
Morbid (grade-3) obesity	1 (0.8%)

BMI: body mass index; CKD: chronic kidney disease; HBV: hepatitis B virus; HIV: human immunodeficiency virus; SD: standard deviation.

Fifty-three (43.1%) subjects had tertiary education, 36 (29.3%) subjects had secondary education, 18 (14.6%) subjects had primary education, and 16 (13.0%) subjects had no formal education. A total of 86 (69.9%) study subjects were in CKD stage 5, 15 (12.2%) subjects were in CKD stage 4, 19 (15.5%) subjects were in CKD stage 3, 2 (1.6%) subjects were in CKD stage 2, and the remaining 1 (0.8%) subject was in CKD stage 1 (see Table 2).

Regarding the form of nephrological interventions offered, the majority of the respondents, 66 (53.66%) subjects, were on maintenance dialysis, followed by 53 (43.09%) subjects on conservative care, while 4 (3.25%) subjects were on renal transplantation (see Table 2). Seventy patients were on renal replacement therapy. The prevalence of renal replacement therapy in this study was 57%.

In this study, the range for number of associated comorbidities was 0–6 diseases with a mean of 2.33 ± 1.09 diseases per patient. Most proportion of the respondents, 45 (36.59%), had two comorbidities, followed by 36 (29.27%) with three comorbidities, 23 (18.70%) had only one comorbidity, while 11 (8.94%) had four comorbidities (see Table 2). The most common specific comorbidities were hypertension in 103 (83.70%), diabetes mellitus in 39 (31.70%), obesity in 24 (19.51%), heart failure in 11 (8.9%), obstructive uropathy in 8 (6.5%), HIV in 7 (5.7%), and stroke in 5 (4.1%; see Table 2).

A total number of 1237 medications were prescribed for the study subjects, and the mean number of prescribed medications per patient was 10.06 ± 3.97. A total of 18 (14.6%) CKD subjects were on ≤5 medications, 43 (35.0%) subjects were on 6–10 medications, 51 (41.5%) subjects were on 11–15 medications, and 11 (8.9%) subjects were on ≥16 medications (see Table 2).

The prescribed medications frequently involved in extremely harmful, clinically nonbeneficial DDIs in this study were heparin in 67 (54.47%) subjects, lisinopril in 65 (52.9%) subjects, oral calcium carbonate (CaCO₃) in 63 (51.2%) subjects, amlodipine in 56 (45.5%) subjects, oral ferrous sulfate (OFS) in 50 (40.7%) subjects, α-methyldopa in 32 (26%) subjects, and atorvastatin in 30 (24.4%) subjects (see Table 3). Prophylactic low molecular weight heparin (enoxaparin/clexane) was prescribed for these CKD patients with symptomatic ESRD who are undergoing maintenance hemodialysis (MHD) sections so as to prevent hemocoagulopathy disorder in them. In addition, these CKD patients with diabetic ketoacidosis/hyperosmolar hyperglycemic state (DKA/HHS), ischemic stroke, varicose veins, valvular heart diseases, ischemic heart disease, atrial fibrillation, or/and prolonged immobility may also be prescribed prophylactic enoxaparin/clexane in order to prevent the occurrence of deep vein thrombosis or cardiogenic thrombosis and subsequent thromboembolism phenomenon in them. The route of administration is subcutaneous usually at a dose of 40 mg (4000 IU) once daily prior to the commencement of hemodialysis section. Duration and frequency of administration among CKD patients are determined by the clotting profile (International Normalized Ratio [INR] value between the range of 0.8–1.3 is usually the desirable target) and also at the discretion of the nephrologist physician depending on the primary indication/purpose for usage such as for maintenance hemodialysis sections in symptomatic ESRD patients or for ESRD/pre-ESRD CKD patients with DKA/HHS, ischemic stroke, varicose veins, valvular heart diseases, ischemic heart disease, atrial fibrillation, or/and prolonged immobility.

Table 3.

List of the prescribed medications frequently involved in extremely harmful, clinically nonbeneficial DDIs among the study population.

Drugs involved in extremely harmful DDIs	Frequency (%)
Heparin	67 (54.5)
Lisinopril	65 (53)
Calcium carbonate (oral)	63 (51.2)
Amlodipine	56 (45.5)
Oral ferrous sulfate	50 (40.7)
α-Methyldopa	32 (26)
Atorvastatin	30 (24.4)
Nifedipine	24 (19.5)
Metoclopramide	22 (18)
Ciprofloxacin	18 (14.6)
Calcium gluconate (intravenous)	12 (9.8)
Digoxin	10 (8.1)
Amoxicillin^a–clavulanate	10 (8.1)
Losartan	9 (7.3)
Omeprazole	9 (7.3)
Clopidogrel	7 (5.7)
Artemether–lumefantrine^a	6 (4.9)
Prednisolone	5 (4.1)
Mycophenolate mofetil	5 (4.1)
Simvastatin	4 (3.3)
Ceftriaxone	4 (3.3)
Sulfamethoxazole^a–trimethoprim (cotrim)	4 (3.3)
Tacrolimus	3 (2.4)
Abacavir	3 (2.4)
Dipyridamole	3 (2.4)
Rabeprazole	3 (2.4)
Ritonavir^a–lopinavir	2 (1.6)
Warfarin	2 (1.6)
Allopurinol	2 (1.6)
Hydroxychloroquine	2 (1.6)
Haloperidol	2 (1.6)
Cyclosporine	1 (0.8)
Phenobarbital	1 (0.8)
Cefpodoxime	1 (0.8)
Promethazine	1 (0.8)
Verapamil	1 (0.8)
Valganciclovir	1 (0.8)

DDIs: drug–drug interactions.

^a Component drug(s) of the fixed combination involved in extremely harmful, clinically nonbeneficial DDIs.

A total number of 1851 interactions were identified among 118 (95.9%) subjects, while 5 (4.1%) subjects had no known interaction. Twenty-nine (23.6%) subjects had between 6 and 10 interactions, 26 (21.1%) subjects had between 11 and 15 interactions, while 16 (13.0%) subjects had between 21 and 25 interactions; the rest are shown in Table 4.

Table 4.

Number of DDIs observed among the study population.^a

Number of DDIs	Frequency (%)
None	5 (4.1%)
1–5	15 (12.2%)
6–10	29 (23.6%)
11–15	26 (21.1%)
16–20	15 (12.2%)
21–25	16 (13.0%)
≥26	17 (13.8%)

DDIs: drug–drug interactions.

^a Mean DDIs = 15.69 ± 11.00 interactions per patient.

The mean for all DDIs observed per prescription in this study was 1.5, which means that for each additional medication prescription issued out to CKD patients in this study, there is additional chance of at least 1–2 DDIs occurring.

The mean for all extremely harmful, clinically nonbeneficial DDIs observed per patient belonging to these extremely harmful interaction categories in this study was 1.9; this revealed that there was likelihood of at least two extremely harmful DDIs in the prescription for each CKD patient belonging to these extremely harmful interaction categories in this study.

Also, the mean for all extremely harmful, clinically nonbeneficial DDIs observed per prescription in this study was 0.046 (equivalent to 4.6% or 46 extremely harmful interactions per a thousand prescriptions issued), which means that for every 1000 prescriptions issued out to CKD patients in this study, there was chance for the occurrence of at least 46 extremely harmful DDIs.

Furthermore, in this study, the prevalence of extremely harmful, clinically nonbeneficial DDIs (that is, type D or type X interaction categories only) was 24.4%, while the overall prevalence for all the observed DDIs was 95.9%.

The range for the number of DDIs among the 118 respondents was 1–85 interactions with a mean value of 15.69 ± 11.00 interactions per patient. A total of 639 (34.5%) DDIs were of mild severity (type B), 1155 (62.4%) DDIs were of moderate severity (type C), 45 (2.4%) DDIs were of major severity (type D), while 12 (1%) DDIs were of contraindicated/avoid drug combination (type X; see Table 5). Furthermore, a total number of 639 (34.5%) type B interactions with a mean value of 5.86 ± 4.35 number of interactions per patient were identified in 109 (88.6%) subjects. A total number of 1155 (62.4%) type C interactions with a mean value of 9.80 ± 7.37 number of interactions per patient were identified in 118 (95.9%) subjects. A total number of 45 (2.4%) type D interactions with a mean value of 1.50 ± 1.02 number of interactions per patient were identified in 30 (24.4%) subjects. Also, a total number of 12 (1%) type X interactions with a mean value of 1.20 ± 0.86 number of interactions per patient were identified in 10 (8.1%) subjects. While five (4.1%) subjects belonged to type A (no known drug interaction) category.

Table 5.

Level of clinical significance (severity) of DDIs among the population.

Level of clinical significance (severity) of DDIs	Number of interactions (%)
Type B (mild or minor DDIs)	639 (34.5%)
Type C (moderate or significant DDIs)	1155 (62.4%)
Type D (major or serious DDIs)	45 (2.4%)
Type X (contraindication or avoid drug combination)	12 (1%)

DDIs: drug–drug interactions.

A total of 960 (51.8%) DDIs were pharmacokinetic interactions (P_k), 890 (48.1%) DDIs were pharmacodynamics interactions (P_d), while the remaining 1 (0.1%) DDI was a pharmaceutical interaction (P_C; see Table 6). Furthermore, a total number of 960 (51.8%) pharmacokinetic interactions with a mean value of 8.90±7.19 number of interactions per patient were identified in 108 (87.8%) subjects. A total number of 890 (48.1%) pharmacodynamics interactions with a mean value of 7.54±5.14 number of interactions per patient were identified in 118 (95.9%) subjects. While a total number of one (0.1%) pharmaceutical interaction with a mean value of 1.00±0.00 number of interaction per patient was identified in only one (0.81%) subject.

Table 6.

Mechanism of DDIs observed among the study population.

Mechanism of DDIs	Number of interactions (%)
Pharmacodynamics interactions (P_d)	890 (48.1%)
Pharmacokinetic interactions (P_k)	960 (51.8%)
Pharmaceutical interactions (P_C)	1 (0.1%)

DDIs: drug–drug interactions.

The most frequent major severity interactions (type D) in this study occured between α-methyldopa and metoclopramide: eight (0.45%) interactions in eight (6.5%) patients; mycophenolate mofetil (MMF) and tacrolimus: three (0.2%) interactions in three (2.4%) patients; MMF and OFS: three (0.2%) interactions in three (2.4%) patients; while dipyridamole and heparin: three (0.2%) interactions in three (2.4%) patients; and the rest were shown in Table 7.

Table 7.

Pattern of extremely harmful, clinically nonbeneficial DDIs observed among the study population.

Pattern of DDIs observed	Number of patients involved (%)	Number of Interactions (%)	Level of clinical significance	Mechanism of DDIs	Adverse effect(s) of the extremely harmful DDIs
α-Methyldopa + Metoclopramide	8 (6.5)	16 (0.9) Note: D = 8 (0.45); X = 8 (0.45)	(a) D(b) X	(a) P_k (b) P_d	(a) Metoclopramide decreases the level of α-methyldopa by inhibition of GIT absorption; applies to only oral formulations of both agents (type D; pharmacokinetic) (b) α-Methyldopa decreases the antiemetic effects of metoclopramide by pharmacodynamics antagonism at the CTZ site D₂ receptors (type X; pharmacodynamics).
α-Methyldopa + haloperidol	2 (1.6)	2 (0.1)	X	P_d	α-Methyldopa decreases the antiemetic and antipsychotic effects of haloperidol by pharmacodynamics antagonism at the CTZ site and mesocorticolimbic system D₂ receptors, respectively.
α-Methyldopa + promethazine	1 (0.81)	1 (0.1)	X	P_d	α-Methyldopa decreases the antiemetic effects of promethazine by pharmacodynamics antagonism at the CTZ site D₂ receptors.
Calcium gluconate + ceftriaxone	1 (0.81)	1 (0.1)	X	P_C	Do not administer any intravenous calcium-containing solution (including IV calcium gluconate, Ringer’s lactate, or Hartmann’s solution) in combination with IV ceftriaxone because of the high potential risk of fatal particulate ceftriaxone–calcium complex precipitates being deposited in the lungs, heart, and kidneys. Administration interval should be separated by at least 48 hours apart.
MMF + tacrolimus	3 (2.4)	3 (0.2)	D	P_d	Both MMF and tacrolimus increase immunosuppressive effects of each other. This predisposes to high risk of infections.
MMF + OFS	3 (2.4)	3 (0.2)	D	P_k	OFS decreases the level of MMF by inhibition of GIT absorption; applies to oral form of both agents.
Dipyridamole + heparin	3 (2.4)	3 (0.2)	D	P_d	Both agents enhanced the risk of hemorrhage through pharmacodynamics synergism.
Atorvastatin + ritonavir	2 (1.6)	2 (0.1)	D	P_k	Ritonavir increases the rhabdomyopathy toxicity of atorvastatin by inhibiting the membrane influx transporters OATP1B1, which belong to the solute-linked carriers (SLCO) superfamily particularly SLCO21A family, and they are ATP-independent polypeptides.
Cyclosporine + MMF	1 (0.81)	2 (0.1)	D	(a) P_k, (b) P_d	(a) Cyclosporine will decrease the level/effect of MMF by interfering with the enterohepatic recirculation of mycophenolic acid (type D; pharmacodynamics). (b) Both cyclosporine and MMF increase the immunosuppressive effects of each other, thereby increasing the risk of infections (type D; pharmacodynamics).
Nifedipine + simvastatin	2 (1.6)	2 (0.1)	D	P_k	Nifedipine will increase the level/concentration of simvastatin by inhibiting the hepatic/intestinal CYP3A4 metabolism.
CaCO3 + digoxin	2 (1.6)	2 (0.1)	D	P_k	CaCO₃ will increase the absorption concentration and effects of digoxin by increasing GIT pH; this applies to only oral form of both agents.
Ceftriaxone + heparin	2 (1.6)	2 (0.1)	D	P_d	Ceftriaxone increases the anticoagulant activity of heparin. Cephalosporins generally decrease prothrombin (factor II) activity.
Abacavir + valganciclovir	1 (0.81)	1 (0.1)	D	P_d	Both agents increase the hematologic toxicity of each other.
Cyclosporine + sulfamethoxazole	1 (0.81)	1 (0.1)	D	P_d	Both combinations have significantly increased nephrotoxicity effects.
Cyclosporine + phenobarbital	1 (0.81)	1 (0.1)	D	P_k	Phenobarbital decreases the level/concentration of cyclosporine by increasing metabolism via hepatic and intestinal cytochrome P450 isoenzymes—CYP3A4; CYP1A2; and P-glycoprotein (MDR1) efflux transporter pumps.
Lisinopril + losartan	1 (0.81)	1 (0.1)	D	P_d	Both drugs increase the toxicity of each other by pharmacodynamics synergism; dual blockade of renin–angiotensin systems increases the risk of renal impairment, hyperkalemia, and hypotension.
Prednisolone + simvastatin	1 (0.81)	1 (0.1)	D	P_k	Prednisolone will decrease the level or effect of simvastatin by affecting/interfering with hepatic and intestinal CYP3A4 metabolism.
Amlodipine + simvastatin	1 (0.81)	1 (0.1)	D	P_k	Amlodipine increases the level/concentration of simvastatin with potential for increased risk of myopathy/rhabdomyolysis; limits simvastatin dose to no more than 20 mg/day when both drugs are used concurrently.
Digoxin + omeprazole	1 (0.81)	1 (0.1)	D	P_k	Omeprazole will increase the absorption concentration and effects of digoxin by increasing the GIT pH; prolonged use of PPIs may also cause hypomagnesemia with an increased risk of digoxin toxicity. This interaction applies to only oral form of both agents.
Clopidogrel + omeprazole	1 (0.81)	1 (0.1)	D	P_k	Omeprazole decreases the pharmacological antiplatelet effects of clopidogrel (prodrug) by inhibiting hepatic CYP2C19 metabolism. Clopidogrel is metabolized/converted to its active metabolite in part by CYP2C19 microsomal enzymes.
Cefpodoxime + warfarin	1 (0.81)	1 (0.1)	D	P_d	Cefpodoxime will increase the anticoagulant activity of warfarin. Cephalosporins decrease Prothrombin activity in part by decreasing vitamin K producing intestinal flora.
Cefpodoxime + heparin	1 (0.81)	1 (0.1)	D	P_d	Cefpodoxime will increase the anticoagulant activity of heparin. Cephalosporins usually decrease prothrombin activity.
Allopurinol + warfarin	1 (0.81)	1 (0.1)	D	P_k	Allopurinol increases the anticoagulant effects of warfarin by decreasing its metabolism.
Heparin + warfarin	1 (0.81)	1 (0.1)	D	P_d	Both agents exhibit pharmacodynamics synergism to increase anticoagulant effects.
Lumefantrine + promethazine	1 (0.81)	1 (0.1)	D	P_d	Both promethazine and Lumefantrine increase QTC interval and risk of cardiotoxicity.
Amoxicillin + MMF	1 (0.81)	1 (0.1)	D	P_k	Amoxicillin will decrease the serum level/concentration of mycophenolate by impairing its enterohepatic circulation.
Hydroxychloroquine + MMF	1 (0.81)	1 (0.1)	D	P_d	Both agents increase the immunosuppressive effects of each other, thereby predisposing to a high risk of infections.
Ciprofloxacin + OFS	1 (0.81)	1 (0.1)	D	P_k	OFS decreases the blood level of ciprofloxacin by inhibition of GIT absorption via chelating effect. This applies to only oral form of both drugs. Coadministration of oral ciprofloxacin with multivalent cation-containing products will reduce its bioavailability by about 90%, administer ciprofloxacin orally at least 2 hours before or 6 hours after using these products.
Clopidogrel + rabeprazole	1 (0.81)	1 (0.1)	D	P_k	Rabeprazole decreases the antiplatelet effects of clopidogrel by inhibiting hepatic CYP2C19 metabolism. Clopidogrel is a prodrug that required metabolism to its active metabolite, and this is being carried out in part by CYP2C19.
Simvastatin + verapamil	1 (0.81)	1 (0.1)	D	P_k	Verapamil will increase the serum level and effect of simvastatin by inhibiting hepatic/intestinal enzyme CYP3A4 metabolism. This potentially increases the risk of myopathy/rhabdomyolysis, do not exceed simvastatin dose of 10 mg daily when given verapamil concurrently.

CTZ: chemoreceptor trigger zone; DDI: drug–drug interaction; MDR1: multidrug resistance transporter 1; MMF: mycophenolate mofetil; OFS: oral ferrous sulfate; P_C: pharmaceutical DDIs; P_d: pharmacodynamics DDIs; P_k: pharmacokinetic DDIs.

^a Total number of type D interactions = 45 interactions (2.4%) in 30 patients (24.4%). Total number of type X interactions = 12 interactions (1%) in 10 patients (8.1%). Total number of DDIs observed = 1851 interactions (100%) in 123 patients (100%).

The most frequent contraindicated/avoid drug combination interactions (type X) in this study occurred between α-methyldopa and metoclopramide: eight (0.45%) interactions in eight (6.5%) patients; α-methyldopa and haloperidol: two (0.1%) interactions in two (1.6%) patients; α-methyldopa and promethazine: one (0.1%) interaction in one (0.81%) patient; while IV calcium gluconate and IV ceftriaxone: one (0.1%) interaction in one (0.81%) patient (see Table 7).

The only one pharmaceutical DDI in this study occurred between IV calcium gluconate and IV ceftriaxone (specifically, type X; pharmaceutical interaction), which carries the potential risk of fatal particulate ceftriaxone–calcium complex precipitates being deposited in the lungs, heart, and kidneys. The administration interval between IV ceftriaxone and any IV calcium-containing solutions (such as Ringer’s solution, Hartman’s solution, and IV calcium gluconate) must be separated by at least 48 hours apart (see Table 7).

Regarding the patterns of extremely harmful, clinically nonbeneficial DDIs among the respondents, a total number of 30 different patterns of interactions were observed with the most frequent extremely harmful DDI patterns occurring between α-methyldopa and metoclopramide: 16 (0.9%) interactions in eight (6.5%) patients; MMF and tacrolimus: three (0.2%) interactions in three (2.4%) patients; MMF and OFS: three (0.2%) interactions in three (2.4%) patients; while dipyridamole and heparin: three (0.2%) interactions in three (2.4%) patients. The rest are shown in Table 7.

With respect to the Venn’s diagram in Figure 1 for the level of clinically significant DDIs observed among the patients, it was noticed that 10 (8.1%) subjects had all the categories of clinically significant interactions type B, type C, type D, and type X altogether. Twenty (16.3%) subjects had only type D interactions, 79 (64.2%) subjects had only type B interactions, while 9 (7.3%) subjects had only type C interactions. Five (4.1%) subjects belonged to type A (no known drug interaction) category, that is; neither had type B nor type C nor type D nor type X interactions. Furthermore, 30 (24.4%) subjects had either type D or type X interactions, 109 (88.6%) subjects had either type B or type D or type X interactions, 88 (71.5%) subjects had either type B or type C interactions, 99 (80.5%) subjects had either type B or type D interactions, 29 (23.6%) subjects had either type C or type D interactions, 108 (87.8%) subjects had either type B or type C or type D interactions while 118 (95.9%) subjects had either type B or type C or type D or type X interactions. It was also observed that the respondents with type X interactions were found to be a subset of the respondents with type D interactions, while those respondents with type D interactions were also a subset of the respondents with type B interactions. In addition, those respondents with type B interactions were also found to be a subset of the respondents with type C interactions. In summary, all the respondents with either type B or type D or type X interactions were all subsets of the respondents with type C interactions. What this implies is that an individual patient with the propensity to develop drug interactions can only be prone to type B, type D and type X interactions when he/she has first developed type C interaction. Likewise an individual patient can only develop type X interaction when he/she has first developed type C, type B and type D interactions.

Figure 1.

Venn’s diagram for the level of clinically significant interactions among the study population.

Also, with respect to the Venn’s diagram in Figure 2 for the mechanism of DDIs observed among the patients, it was noticed that only one (0.81%) subject had all the known mechanisms of pharmacodynamics, pharmacokinetic and pharmaceutical interactions altogether. One hundred and seven (87%) subjects had only pharmacokinetic interactions, while 10 (8.1%) subjects had only pharmacodynamics interactions. Five (4.1%) subjects had “no known mechanism of drug interaction”, that is; neither had pharmacodynamics nor pharmacokinetic nor pharmaceutical interactions. Furthermore, 117 (95.1%) subjects had either pharmacokinetic or pharmacodynamics interactions, 108 (87.8%) subjects had either pharmacokinetic or pharmaceutical interactions, while 118 (95.9%) subjects had either pharmacodynamics or pharmacokinetic or pharmaceutical interactions. It was also observed that the patient with pharmaceutical interaction was a subset of those with pharmacokinetic interactions, while the patients with pharmacokinetic interactions were also found to be a subset of those with pharmacodynamics interactions in this study. In summary, all the patients with either pharmacokinetic or pharmaceutical interactions were all subsets of the patients with pharmacodynamics interactions. What this implies is that an individual patient with the propensity to develop drug interactions can only be prone to pharmacokinetic and pharmaceutical interactions when he/she has first developed pharmacodynamics interaction. Likewise an individual patient can only develop pharmaceutical interaction when he/she has first developed both pharmacodynamics and pharmacokinetic interactions.

Figure 2.

Venn’s diagram for the mechanism of interactions among the study population. P_C: pharmaceutical interaction; P_d: pharmacodynamics interaction; P_k: pharmacokinetic interaction.

In this study, there was a statistically significant association between the number of prescribed medications and the eGFR (pre-ESRD and ESRD stages) for the respondents with a p value less than 0.0001. This implies that the number of prescribed medications increases as their eGFR decline in advance CKD stage patients (Table 8).

Table 8.

Test for association between the number of prescribed medications and the estimated glomerular filtration rate for study population.^a

	Pre-ESRD stage (eGFR ≥ 15 mL/min)	ESRD stage (eGFR = 0–14 mL/min)
1–5 medications	13	5
6–10 medications	18	25
11–15 medications	5	46
≥16 medications	1	10

eGFR: estimated glomerular filtration rate; ESRD: end-stage renal disease.

^a χ ² = 30.31, df = 3, p < 0.0001 (significant), critical value = 7.815, α = 0.05.

Furthermore, there was a statistically significant association between the number of prescribed medications and the occurrence of DDIs among the respondents with a p value less than 0.0001. This implies that as the number of prescribed medications increases, the chance for occurrence of DDIs also increases among these CKD patients. About 77.8% of those CKD patients taking around 1–5 medications are at risk of developing DDIs, while 99.05% of those taking around 6 or more medications are at risk of developing DDIs (Table 9).

Table 9.

Test for the association between the number of prescribed medications and the occurrence of DDIs among the study population.^a

	DDIs present	DDIs absent
1–5 medications	14	4
≥6 medications	104	1

DDIs: drug–drug interactions.

^a χ ² = 17·83, df =1, p < 0.0001 (significant), critical value = 3.841, α = 0.05.

In addition, there was a statistically significant association between the eGFR and the occurrence of DDIs among the study population with a p value less than 0.01. This implies that the chance for the occurrence of DDIs increases among these CKD patients as their eGFR decline in advance CKD stage due to the substantial increase in the number of prescribed medications being ingested (Table 10).

Table 10.

Test for the association between the eGFR and the occurrence of DDIs among the study population.^a

	DDIs present	DDIs absent
Pre-CKD stages (eGFR ≥60 mL/min/1.73 m²)	2	1
CKD stages (eGFR < 60 mL/min/1.73 m²)	116	4

CKD: chronic kidney disease; DDIs: drug–drug interactions; eGFR: estimated glomerular filtration rate.

^a χ ² = 6.755, df =1, p < 0.01 (significant), critical value = 3.841, α = 0.05.

Last, among these CKD patients recruited for this study, there was also a statistically significant association between those with diabetes mellitus and obesity with p value less than 0.0001. This implies that those CKD patients with obesity are highly predisposed and at risk of developing diabetes mellitus (Table 11).

Table 11.

Test for the association between diabetes mellitus status and obesity status for the study population.^a

	Obesity present	Obesity absent
Diabetes mellitus present	16	23
Diabetes mellitus absent	8	76

^a χ ² = 16.83, df =1, p < 0.0001 (significant), critical value = 3.841, α = 0.05.

Discussion

The most frequent extremely harmful, clinically nonbeneficial DDI in this study was between α-methyldopa and metoclopramide: 16 (0.9%) interactions in 8 (6.5%) patients. Furthermore, α-methyldopa decreases the antiemetic effects of metoclopramide by pharmacodynamics antagonism at the chemoreceptor trigger zone site D₂-receptors (type X; pharmacodynamics). In addition, metoclopramide decreases the level of α-methyldopa by inhibition of GIT absorption, as this applies to only oral formulations of both agents (type D; pharmacokinetic). This was in contrast to the study done by Saleem et al.²¹ where the most frequent extremely harmful, clinically nonbeneficial DDI reported was between IV calcium gluconate and IV ceftriaxone 21 (3.6%) interactions (specifically, contraindicated type X; pharmaceutical interactions) in 21 (10.01%) CKD patients, which carries the potential risk of fatal particulate ceftriaxone–calcium complex precipitates being deposited in the lungs, heart, and kidneys. The administration interval between IV ceftriaxone and any IV calcium-containing solutions (such as Ringer’s solution, Hartman’s solution, and IV calcium gluconate) must be separated by at least 48 hours apart.

The prescribed medications frequently involved in extremely harmful, clinically nonbeneficial DDIs in this study were heparin, lisinopril, oral CaCO₃, amlodipine, and oral OFS. This agreed with previous studies where oral CaCO₃ (Sgnaolin et al.¹⁸) and lisinopril (Chinwendu et al.³⁰) were also reported as among the most frequently prescribed medications for CKD patients.^18,30

In this study, the mean prescribed medications per patient was 10.06 ± 3.97. This is slightly higher than 7.87 ± 2.44 reported by Al-Ramahi et al.²⁰ However, Rama et al.¹⁹ reported a higher mean prescribed drugs per patient of 12.08 ± 6.30 compared to the finding in this present study. Our study population involved both nondialyzed and dialyzed patients unlike the study by Al-Ramahi et al. that involved only CKD patients on MHD. Patients who are on regular MHD may require less number of drugs for blood pressure control; hence, this may possibly explain the lower mean number of prescribed medications. The practice of polypharmacy in the management of CKD is however not surprising because they have a high number of cardiovascular risk factors, comorbidities, and complications such as hypertension, diabetes mellitus, heart failure, arrhythmias, anemia, hyperuricemia, and calcium–phosphate abnormalities, which are managed by a combination of drugs. The most common comorbidities in this study were hypertension and diabetes, which agreed with previous studies done.^7,22,23 This can be attributed to the fact that both conditions are the leading etiologies of CKD in Nigeria, Sub-Saharan West Africa region, and worldwide. The overall prevalence of DDIs among the studied CKD patients was 95.9% with about 62.4% showing moderate (type C) clinical significance.

The mean for all DDIs observed per prescription in this study was 1.5, which was similar to the report by Sgnaolin et al.¹⁸ (1.0) but lower than the number reported by Rama et al.¹⁹ (2.7) and Marquito et al.²² (2.5). This showed that there was likelihood of at least one to two DDIs occurring for each additional medication prescription issued out to CKD patients in this study.

Furthermore, the mean for all extremely harmful, clinically nonbeneficial DDIs observed per patient belonging to these extremely harmful interaction categories in this study was 1.9; this revealed that there was likelihood of at least two extremely harmful DDIs in the prescription for each CKD patient belonging to these extremely harmful interaction categories in this study. Also, the mean for all extremely harmful, clinically nonbeneficial DDIs observed per prescription in this study was 0.046 (equivalent to 4.6% or 46 extremely harmful interactions per a 1000 prescriptions issued), which means that for every 1000 prescriptions issued out to CKD patients in this study, there was chance for the occurrence of at least 46 extremely harmful DDIs compared to the standard reference value of less than 0.00001 (equivalent to less than 0.001% or less than one extremely harmful interaction per a 100,000 prescriptions issued). This finding however leaves room for more important and nonnegotiable improvement in our rational prescriptions for CKD patients. This therefore emphasize the need for clinical pharmacologists and clinical pharmacists to regularly evaluate the prescriptions of CKD patients for extremely harmful, clinically nonbeneficial DDIs. The prevalence of extremely harmful, clinically nonbeneficial DDIs (i.e. type D or type X interaction categories only) in this study was 24.4% compared to the standard reference value of near or approximately 0% (i.e. value less than 0.001%) being required for efficient, accurate, and rational medical prescriptions in clinical practice. While the overall prevalence for all the DDIs observed in this study was 95.9% which was higher than the range of 56.9-89.1% reported in some previous studies.^19

–23 Furthermore, the estimated overall prevalence for DDIs in this study was higher than the value of 56.9% reported by Sgnaolin et al.¹⁸ Differences in methodology, average number of medications per prescription, number of specific associated comorbidities or complications per patient, and CKD stages of the studied population could account for the variations seen in the prevalence rates for overall DDIs in the different studies. For example, in the study by Sgnaolin et al.¹⁸ where a lower prevalence of overall DDIs was reported compared to this study, the average number of medications per prescription was 6.3 ± 3.1, but unlike this present study where the average number of medications per prescription was 10.06 ± 3.97. In fact, the number of medications per prescription and number of specific associated comorbidities or complications per patient were the major determinants of the prevalence rate for extremely harmful DDIs in this study.

Regarding the level of clinically significant interactions in this study, majority of the DDIs (62.4%) were of moderate severity (type C) and this was similar to previous reports.^18,19,22,23 In this study, the chance for the occurrence of type D (major/serious) interactions was 2.4%, which were found in 30 patients. While the chance for the occurrence of type X interactions (contraindication/avoid drug combination) was 1%, which were found in 10 patients; this is comparable to the value of 0.4% type X interactions reported by Marquito et al.²² This finding is however at variance to that of Saleem et al.²¹ who reported that 13.4% of DDIs observed in their study as contraindication/avoid drug combination.

A slightly higher proportion 960 (51.8%) interactions of the total DDIs was found to be from pharmacokinetic interactions. This is different from the reports in other studies where pharmacodynamics interactions were found to be predominant.^19,23 Also, most of the total DDIs were of delayed onset, which was similar to the report by Rama et al.¹⁹ The clinical significance of this is that the patients may not manifest the effects of DDIs early, hence the need for long-term follow-up of these patients.

There was a statistically significant association between the number of prescribed medications and the eGFR (pre-ESRD and ESRD stages) for the respondents in this study with a p value less than 0.0001. This implies that the number of prescribed medications increases as their eGFR decline in advance CKD stage patients. This finding also agreed with previous studies done by Sgnaolin et al.,¹⁸ Rama et al.,¹⁹ and Marquito et al.²²

Furthermore, there was a statistically significant association between the number of prescribed medications and the occurrence of DDIs among the respondents with a p value less than 0.0001. This implies that as the number of prescribed medications increases, the chance for the occurrence of DDIs also increases among these CKD patients. About 77.8% of those CKD patients taking around 1–5 medications are at risk of developing DDIs (compared to the value of 50% in previous reference literatures), while 99.05% of those taking around 6 or more medications are at risk of developing DDIs (compared to the value of 100% in previous reference literatures).^18,19,22,23

Last, among these CKD patients recruited for this study, there was also a statistically significant association between those with diabetes mellitus and obesity with a p value less than 0.0001. This implies that those CKD patients with obesity are highly predisposed and at risk of developing diabetes mellitus.

The limitation of this study was that the extremely harmful, clinically nonbeneficial DDIs detected in this study were theoretically discovered using Medscape drug interaction checker. Also, the Medscape drug interaction checker used in this study did not take into consideration the prescribed dose, frequency of administration, route of administration, and duration of medication use.

Conclusion

In this study, the occurrence and burden of extremely harmful, clinically nonbeneficial DDIs is significantly high among these CKD patients. Nephrologists, clinical pharmacologists and clinical pharmacists should utilize available interaction detecting softwares in order to identify and avoid extremely harmful DDIs among CKD patients. There is also a critical need to minimize the number of prescribed medications for these patients in order to prevent the occurrence and burden of extremely harmful DDIs. This approach will save lives and avoid the scourge of associated morbidity, mortality, enormous healthcare cost, increase frequency and length (duration) of hospitalization. However, this study has also brought to limelight the magnitude of extremely harmful, clinically nonbeneficial DDIs among CKD patients in renal practice. In addition, there is a proper need to take proactive steps to reduce these additional burdens and optimize the care of our renal patients.

Recommendations

Nephrologists, clinical pharmacologists and clinical pharmacists should make use of available interaction detecting softwares to check all prescribed medications for CKD patients so as to identify and avoid the presence of extremely harmful, clinically nonbeneficial DDIs.

A cordial integrated relationship between healthcare professionals (Nephrologists, clinical pharmacologists, clinical pharmacists and nephrology nurses) should be encouraged in order to optimize CKD patients’ care and to reduce the occurrence and burden of extremely harmful DDIs in them.

Whenever the clinical pharmacists that are involved in the rational dispensing of CKD patients medications detect the presence of any harmful DDIs in their prescription, he/she should alert the doctors (Nephrologists and clinical pharmacologists) involved and inform them of the impending problem so that feasible alternative prescriptions can be made and the error corrected as this will involve more demanding effort in the area of clinical pharmacist and doctor’s commitment time but the hospital should make this mandatory. As this approach will save lives, reduce the morbidity, mortality, frequency of hospitalisation, length of hospital stay and healthcare costs associated with the burden and occurrence of extremely harmful DDIs in clinical practice.

Regular organisation of health education awareness program on the prevention of CKD and its associated complications among the general public should be done by healthcare professionals (Nephrologists, clinical pharmacologists, clinical pharmacists and nephrology nurses) coupled with adequate support from the government.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Olumuyiwa John Fasipe

References

Fatoba

Oke

Hirst

. Global prevalence of chronic kidney disease: a systematic review and meta analysis. PLoS One 2016; 11(7): e0158765.

Cardone

Bacchus

Assimon

. Medication related problems in chronic kidney disease. Adv Chronic Kidney Dis 2010; 17(5): 404–412.

Kadiri

Arije

. Temporal variations and meteorological factors in hospital admissions of chronic renal failure in South West Nigeria. West Afr J Med 1999; 18(1): 49–51.

Akinsola

Odesanmi

Ogunniyi

. Diseases causing chronic renal failure in Nigerians—a prospective study of 100 cases. Afr J Med Med Sci 1989; 18(2): 131–137.

Ulasi

Ijoma

Onodugo

. Towards prevention of chronic kidney disease in Nigeria: a community based study in Southeast Nigeria. Kidney Int Suppl 2013; 3: 195–201.

Oluyombo

Ayodele

Akinwusi

. A community study of the prevalence, risk factors and pattern of chronic kidney disease in Osun State, South West Nigeria. West Afr J Med 2013; 32(2): 85–92.

Elam Org

Sitprija

. Co morbidities in patients with end stage renal disease in developing countries. Artif Organs 2002; 26(9): 753–756.

Liu

. Cardiovascular disease and its relationship with chronic kidney disease. Eur Rev Med Pharmacol Sci 2014; 18(19): 2918–2926.

Fernández Llimós

Tuneu

Baena

. Morbidity and mortality associated with pharmacotherapy. Evolution and current concept of drug related problems. Curr Pharm Des 2004; 10(31): 3947–3967.

10.

Mason

. Polypharmacy and medication related complications in the chronic kidney disease patient. Curr Opin Nephrol Hypertens 2011; 20(5): 492–497.

11.

Levin

Hemmelgarn

Culleton

. Guidelines for the management of chronic kidney disease. CMAJ 2008; 179(11): 1154–1162.

12.

Keith

Nichols

Gullion

. Longitudinal follow up and outcomes among a population with chronic kidney disease in a large managed care organization. Arch Intern Med 2004; 164(6): 659–663.

13.

Anavekar

Pfeffer

. Cardiovascular risk in chronic kidney disease. Kid Int Suppl 2004; (92): 111–115.

14.

Babua

Kalyesubula

Okello

. Cardiovascular risk factors among patients with chronic kidney disease attending a tertiary hospital in Uganda. Cardiovasc J Afr 2015; 26(4): 177–180.

15.

Pirmohamod

James

Meakin

. Drug–drug interaction as cause of admission to hospital: prospective analysis of 18820 patients. Br Med J 2004; 329: 5–19.

16.

Lazarou

Pomeranz

Corey

. Incidence of adverse drug reactions in hospitalized patients: a meta analysis of prospective studies. JAMA 1998; 279(15): 1200–1205.

17.

Al Hajje

Atoui

Awada

. Drug related problems identified by clinical pharmacist’s students and pharmacist’s interventions. Ann Pharm Fr 2012; 70(3): 169–176.

18.

Sgnaolin

Engroff

. Assessment of used medications and drug–drug interactions among chronic renal failure patients. Sci Med 2014; 24: 329–335.

19.

Rama

Viswanathan

Acharya

. Assessment of drug–drug interactions among renal failure patients of nephrology ward in a South Indian tertiary care hospital. Indian J Pharm Sci 2012; 74(1): 63–68.

20.

Al-Ramahi

Raddad

Rashed

. Evaluation of potential drug–drug interactions among Palestinian hemodialysis patients. BMC Nephrol 2016; 17: 96.

21.

Saleem

Mashood

Khan

. Clinical relevancy and determinants of potential drug–drug interactions in chronic kidney disease patients: result from a retrospective analysis. Integr Pharm Res Pract 2017; 6: 71–77.

22.

Marquito

Fernandes

Colugnati

FAB

. Identifying potential drug interactions in chronic kidney disease patients. J Bras Nefrol 2014; 36(1):26–34. DOI: 10.5935/0101-2800.2014006.

23.

Hedge

Udaykumar

Manjuprasad

. Potential drug interactions in chronic kidney disease patients. A cross sectional study. Int J Recent Trends Sci Technol 2015; 16: 56–60.

24.

Shad

Marsh

Preskorn

. The economic consequences of drug-drug interaction. J Clin Psychopharmacol 2001; 21(1): 119–120.

25.

Moura

Acurcio

Belo

. Drug-drug interactions associated with length of hospital stay and cost of hospitalization. J Pharm Sci 2009; 12(3): 266–272.

26.

Edwards

Aronson

. Adverse drug reactions: definitions, diagnosis, and management. Lancet 2000; 356(9237): 1255–1259.

27.

Stemer

Lemmens Gruber

. Clinical pharmacy activities in chronic kidney disease and end stage renal disease patients: a systematic literature review. BMC Nephrol 2011; 12: 35.

28.

International Society of Nephrology. Kidney Disease Improving Global Outcome (KDIGO) 2012 Clinical Practice Guideline for evaluation and management of CKD. Kidney Int Supplements 2013; 3: 1–150.

29.

Roblek

Trobec

Mrhar

. Potential drug–drug interactions in hospitalized patients with chronic heart failure and chronic obstructive pulmonary disease. Arch Med Sci 2014; 10(5): 920–932.

30.

Chinwendu

Joda

Oyetunde

. Evaluation of drug–drug interactions among chronic kidney disease patients of Nephrology Unit in the University of Nigeria teaching hospital, Ituku–Ozalla, Enugu state. J Basic Clin Pharma 2017; 8: S049–S053.

31.

Victoria

Matteo

Marco

. Polypharmacy in kidney disease patients. Curr Kidney Dis 2014; 11: 212–9.

32.

Flesch

Erdmann

. The problem of polypharmacy in chronic kidney disease. Curr Cardiol Rep 2006; 8: 217–25.

33.

Mason

Bakus

. Strategies for reducing polypharmacy and other medication–related problems in chronic kidney disease. Semin Dial 2010; 23: 55–61.