In our recently published six-year-long study on drug-related problems in older adults, we reported 1031 DRPs in 666 patients out of 10400 patient registrations.[4] The majority of DRPs belonged to adverse drug reactions (ADRs, n = 933). DDIs were implicated in 189 cases of ADRs, and in 10 cases both DDIs and drug-disease interactions were involved. Thus, DDIs were noted in a total of 199 cases of ADRs (21.3%). These 199 cases developed in 165 patients suggesting that more than one ADR developed due to DDIs in the same patient. Metabolic disturbances (n = 194), nervous system disorders (169), gastrointestinal disturbances (164), and vascular disorders (n = 83) were the common manifestations of ADRs and were because of DDIs in 58 (29.89%), 36 (21.3%), 28(17.1%) and 22 (26.5%), cases respectively. While pedal edema, sedation, and gastritis were the common manifestations of ADRs, DDIs commonly manifested as movement disorders (n = 18), hypotension (n = 16), hypoglycemia (n = 15), and hyperuricemia (n = 11).
3.1. Details of the common three DDIs are provided below.
3.1.1. DDIs causing movement disorders: (n = 18)
Extrapyramidal manifestations were the major drug-related movement disorders. As mentioned in Table 1, antipsychotics were involved in 12 (66.6%) cases. Seven cases (38.8%) were because of antipsychotics used therapeutically for neuropsychiatric conditions such as risperidone, trifluoperazine, aripiprazole, olanzapine, chlorpromazine, and quetiapine. Antipsychotics used at antiemetic doses such as levosulpiride were implicated in six cases (33.3%) of movement disorders. Next to antipsychotics was the pharmacologic subgroup of selective calcium channel blockers with mainly vascular effects, followed by propulsives and antidementia drugs, which were involved in four, three, and three cases respectively. Ten (55.5%) cases of movement disorders happened because of the involvement of drugs having the property of blocking voltage-gated calcium channels. Such drugs included valproate, amlodipine, cilnidipine and cinnarizine, trimetazidine, and xanthines such as theophylline.
Table 1
Common clinically manifest drug-drug interactions (DDIs), individual drugs and the pharmacologic subgroups (WHO-ATC) of drugs causing DDIs, in older adults in outpatient settings. NK: Not known, WHO-ATC: World Health Organization-Anatomic Therapeutic Chemical Classification of drugs
Clinically manifest DDIs | Drugs involved | Pharmacological subgroups |
---|
Movement disorders (n = 18) | Cilnidipine + Levosulpiride (1), Amlodipine + Tramadol (1), Trimetazidine + Amlodipine (1), Chlorpromazine + Metoclopramide (1), Quetiapine + Risperidone (1), Trifluoperazine + Levosulpiride (1), Cinnarizine + Levosulpiride (1), Olanzapine + Donepezil (1), Trimetazidine + Domperidone (1), Aripiprazole + Pregabalin (1), Risperidone + Donepezil (1), Valproate + Caffeine (1), Cinnarizine + Domperidone + Levosulpiride (1), Risperidone + Citicholine + Donepezil (1), Paroxetine + Fluoxetine + Levosulpiride (1) Salmeterol + Terbutaline + Etophylline + Theophylline (1), Cilnidipine + Acebrophylline + Levosulpiride + Meropenem (1), Entacapone + Levodopa + Carbidopa (1) | Antipsychotics = 12, Selective Calcium Channel Blockers with mainly vascular Effects = 4, Propulsives = 3, antidementia drugs = 3, Other cardiac preparations = 2, Antivertigo preparations = 2, Psychostimulants, Agents Used for ADHD And Nootropics = 2, Other Systemic Drugs for Obstructive Airway Diseases = 2, Adrenergics, inhalants = 1, Opioids = 1, Other Analgesics and Antipyretics = 1, Antiepileptics = 1, Dopaminergic agents = 1, Other Beta-Lactam Antibacterials = 1, Antidepressants = 1 |
Hypotension (n = 16) | Torsemide + Spironolactone (4), Amlodipine + Metoprolol (2) Levodopa + Carbidopa (1), Amlodipine + Chlortalidone (1) Cilnidipine + Chlortalidone (1), Hydrochlorothiazide + Amlodipine (1) Isosorbide Dinitrate + Amlodipine (1) Chlortalidone + Nitrate (1), Telmisartan + Hydrochlorothiazide + Amlodipine (1) Olmesartan + Ramipril + Hydrochlorothiazide (1) Telmisartan + NK (1) NK (1) | Selective Calcium Channel Blockers with mainly Vascular Effects = 7, High Ceiling Diuretics = 4, Aldosterone Antagonists and Other Potassium-Sparing Agents = 4, Beta Blocking Agents = 2, Low Ceiling Diuretics, Excl. Thiazides = 3, Low-Ceiling Diuretics, Thiazides = 3, Vasodilators Used in Cardiac Diseases = 2, Angiotensin II Receptor Blockers (ARBS), Plain = 3 ACE Inhibitors, Plain = 1, Dopaminergic Agents = 1 |
Hypoglycaemia(n = 15) | Gliclazide + Metformin (2), Glargine + Regular Insulin (1), Glimepiride + Insulin glargine (1), Metformin + Metoprolol (1), Teneligliptin + Metformin (1), Glimepiride + Metformin + Nebivolol (1) Glipizide + Glimepiride + Metformin (1), Metformin + Pioglitazone + Glimepiride (1) Glimepiride + Metformin + Isoniazid (1), Glimepiride + Metformin + Escitalopram (1) Glimepiride + NK (3), NK (1) | Blood Glucose Lowering Drugs, Excl. Insulins = 13, Insulins And Analogues = 2, Beta Blocking Agents = 2, Drugs For Treatment of Tuberculosis = 1, Antidepressants = 1 |
Hyperuricemia (n = 11) | Pyrazinamide + Torsemide (2), isoniazid + rifampicin (2), Antitubercular drugs NK (2), Propranolol + Telmisartan (1), Cilnidipine + Telmisartan (1), Torsemide + Spironolactone (1), Telmisartan + Torsemide (1), Torsemide + Spironolactone + Chlortalidone (1) | Drugs For Treatment of Tuberculosis = 6, High Ceiling Diuretics = 5, Angiotensin Receptor Blockers (ARBS), Plain = 3, Aldosterone Antagonists and Other Potassium-Sparing Agents = 2, Low-Ceiling Diuretics, Excl. Thiazides = 1, Selective Calcium Channel Blockers with mainly Vascular Effects = 1, Beta Blocking Agents = 1 |
Gastritis (n = 10) | Indomethacin + Prednisolone (1), Mebeverine + Chlordiazepoxide (1) Linezolid + Piperacillin Tazobactam (1), Aspirin + Escitalopram (1) Aspirin + Prednisolone (1) Etoricoxib + Thiocolchicoside (1), Diltiazem + Amlodipine + Iron (1), Aceclofenac + Cefixime + Ofloxacin (1) Norfloxacin + Tinidazole + Potassium supplements (1) Antitubercular drugs (NK) + Methotrexate + Leflunomide (1) | Anti-inflammatory And antirheumatic Products, Non-Steroids = 3, Anti-inflammatory And Antirheumatic Products, Non-Steroids (Antithrombotic Agents) = 2, Corticosteroids for Systemic Use, Plain = 2, Other Antibacterials = 2, Quinolone Antibacterials = 2, Drugs For Functional Gastrointestinal Disorders = 1, Anxiolytics = 1, Antidepressants = 1, Beta-Lactam Antibacterials, Penicillins = 1, Other Beta-Lactam Antibacterials = 1, Potassium = 1, Selective Calcium Channel Blockers with Direct Cardiac Effects = 1, Selective Calcium Channel Blockers with mainly Vascular Effects = 1, Muscle relaxants, centrally acting agents = 1, Iron Preparations = 1, Drugs For Treatment of Tuberculosis = 1, Immunosuppressants = 2 |
Hyponatremia (n = 10) | Furosemide + Aspirin + Pantoprazole (1), Chlortalidone + Imipramine (1) Chlortalidone + Losartan (1) Furosemide + Torsemide (1), Carbamazepine + Hydrochlorothiazide + Olmesartan (1) Chlortalidone + Telmisartan + Rabeprazole (1), Telmisartan + Silodosin + Rabeprazole (1) Dosulepin + Escitalopram + Hydrochlorothiazide + Nebivolol (1), Chlortalidone + Pregabalin + Nortriptyline + Tramadol + Rabeprazole (1), NK (1) | Angiotensin II Receptor Blockers (ARBS), Plain = 4, Drugs for Peptic Ulcer and Gastro-Oesophageal Reflux Disease (GORD) = 4, Low-Ceiling Diuretics, Excl. Thiazides = 4, Antidepressants = 3, High Ceiling Diuretics = 2, Anti-inflammatory And Antirheumatic Products, Non-Steroids (Antithrombotic Agents) = 1, Low-Ceiling Diuretics, Thiazides = 2, Drugs Used in Benign Prostatic Hypertrophy = 1, Beta Blocking Agents = 1, Antiepileptics = 1, Other Analgesics and Antipyretics = 1, Opioids = 1 |
3.1.2. DDIs causing hypotension (n = 16)
As shown in Table 1, selective calcium channel blockers with mainly vascular effects were the major pharmacologic subgroup involved in DDIs, n = 6(37.5%). This was followed by aldosterone antagonists, n = 4 (25%) and high ceiling diuretics, n = 4(25%)
3.1.3. DDIs causing hypoglycemia (n = 15)
As shown in Table 1, the pharmacologic subgroup of blood glucose-lowering drugs, excluding insulins was involved in 13 (86.7%) cases of DDIs causing hypoglycemia. In five cases (33.3%), hypoglycemia happened because of the sole involvement of two or more drugs belonging to this subgroup. In two cases, hypoglycemia happened because of the use of beta-blocking agents in conjunction with oral blood glucose-lowering drugs.
Other clinically manifest DDIs included acute kidney injury, hypertension, hypokalemia, hyperkalemia, and hepatitis. Details of such DDIs are described in Supplementary Table 1.
3.2. Nature and severity of DDIs:
The majority of DDIs (n = 165,83%) happened because of pharmacodynamic interaction between the drugs. Examples included hyperkalemia happening because of simultaneous use of spironolactone and ramipril and hyponatremia occurring due to concomitant use of carbamazepine, hydrochlorothiazide, and olmesartan. A total of 26(13.1%) DDIs were immune-mediated. Examples included hepatitis happening because of antitubercular drugs and immune-mediated cutaneous disturbances because of concomitant use of multiple antimicrobials. Both pharmacokinetic and pharmacodynamic interactions were involved in seven and only pharmacokinetic interaction was suspected in one case of DDI.
Severity-wise, 100, 82, and 17 DDIs were ‘moderate’, ‘mild’, and ‘severe’. The latter included cases of neuroleptic malignant syndrome, pemphigus associated with drug interactions, and hyperkalemia among others.
3.3. DDIs causing hospitalization
Out of the total 1031 DRPs, hospitalization was needed in 267 cases (25.9%). Among 267, 245 happened because of ADRs and in 149 cases, ADRs were the direct contributors to hospitalization.[4] In the present study, hospitalization was required in 70 cases of DDIs (35.2%) and in the majority, DDIs were the direct cause of hospitalization (43/70, 61.4%). Thus, DDIs accounted for 28.6% of ADR-related hospital admissions. Movement disorders, hyponatremia, and hypokalemia were the common DDIs that were followed by hospitalizations As shown in Fig. 1, all cases (100%) of hypokalemia, 90% cases of hyponatremia, and 50% cases of movement disorders occurring because of DDIs were followed by hospitalization. Hyponatremia and movement disorders were the direct reasons for hospitalization in the majority. On the other hand, hypokalemia-related hospitalizations were largely coincidental. Angiotensin receptor blockers (ARBs) were the commonest pharmacologic subgroup involved in hyponatremia followed by low-ceiling diuretics, excluding thiazides, and can be interpreted from Table 2.
Table 2
Clinically manifest drug-drug interactions (DDIs) and hospitalizations in older outpatients. ‘Direct’ hospitalizations refer to the hospital admissions happening directly due to DDIs. Co-incidental hospitalizations refer to hospital admissions where DDIs were present but were not the primary reason of hospitalization. NK: Not known
Hospitalization | MedDRA Low level term | Drug interactions |
---|
Direct | Hyponatremia (n = 8) | ● Chlortalidone/imipramine (1), Chlortalidone/losartan (1), Telmisartan/silodosin/rabeprazole (1), Chlortalidone/telmisartan/rabeprazole (1), Carbamazepine/hydrochlorothiazide/olmesartan (1), Dosulepin/escitalopram/hydrochlorothiazide with nebivolol (1), Furosemide/torsemide (1), NK (1) |
Movement disorders (n = 7) | ● Risperidone/citicholine/donepezil (1), Cilnidipine/levosulpiride (1), Trimetazidine/domperidone (1), Olanzapine/donepezil (1), Chlorpromazine/metoclopramide (1), Entacapone/levodopa/carbidopa (1), Cinnerazine/domperidone/levosulpiride (1) |
Acute kidney injury (n = 5) | ● Ramipril/spironolactone/aspirin/torsemide (1), Telmisartan/diclofenac (1), Ramipril/spironolactone (1), Spironolactone/torsemide (1), Antitubercular drugs (1) |
Hypoglycemia (n = 5) | ● Gliclazide/metformin (2), Glimepiride/herbal (1), Glipizide/glimepiride/metformin (1), Glimepiride/others (1) |
Co-incidental | Hypokalemia (n = 8) | ● Chlorthalidone/rabeprazole (1), Formoterol/theophylline/etophylline (1), Insulin/formoterol inhaler (1), Hydrochlorothiazide/pantoprazole/deflazacort (1) ● Salbutamol/theophylline/etophylline (1), Furosemide/insulin (1), Chlorthalidone/piperacillin-tazobactam (1), Furosemide/torsemide (1) |
| Hyperuricemia (n = 5) | ● Cilnidipine/telmisartan (1), Pyrazinamide/torsemide (1), Torsemide/spironolactone/chlorthalidone (1), Torsemide/spironolactone (1), Antitubercular drugs (1) |
Hepatitis (n = 3) | ● Donepezil/quetiapine (1), Atorvastatin/chlorpromazine (1), NK (1) |
Among seven deaths reported due to DRPs, six happened because of ADRs, and DDIs were implicated in one case of acute onset liver and kidney injury with antitubercular drugs. Details of individual drugs and the pharmacologic subgroups causing hospitalization are mentioned in Table 2.
3.4. Risk factors of clinically manifest DDIs
Among older adults developing ADRs (n = 933), risk factors of ADRs happening due to DDIs were determined. The results of the unadjusted bivariate analysis are mentioned in Supplementary Table 2. After adjusting for confounders, patients with Parkinsonism, acute infectious etiology, coronary artery disease, neuropsychiatric illnesses, and diabetes were at 3.14, 2.78, 1.97, 1.82- and 1.77 times higher odds of developing ADRs due to DDIs (Table 3a)
Table 3
Adjusted analysis of risk factors of clinically manifest drug-drug interactions out of 933 cases (Table 3a) and 523 cases (Table 3b)
Table 3a (N = 933) | Table 3b (N = 523) * |
---|
VARIABLES | aOR (CI) | P- VALUE | aOR (CI) | P- VALUE |
AGE (in years) 50–64 (reference) 65–74 ≥ 75 | 0.80(0.54–1.18) 1.10(0.69–1.74) | 0.30 0.27 0.68 | 0.89(0.53–1.50) 1.14(0.64–2.03) | 0.67 0.66 0.66 |
Sex Male (reference) Female | 1.03 (0.72–1.47) | 0.88 | 0.79(0.48–1.28) | 0.34 |
Diabetes mellitus NO (reference) YES | 1.77(1.22–2.56) | 0.003 | 1.08(0.68–1.72) | 0.72 |
Hypertension NO (reference) YES | 1.36(0.96–1.94) | 0.08 | 1.20(0.77–1.88) | 0.41 |
Benign prostatic hyperplasia NO (reference) YES | 1.01(0.53–1.93) | 0.96 | 0.79(0.38–1.63) | 0.52 |
Lung disease NO (reference) YES | 0.81(0.45–1.42) | 0.46 | 0.63(0.34–1.16) | 0.14 |
Parkinson’s disease NO (reference) YES | 3.14(1.58–6.24) | 0.001 | 1.82(0.74–4.52) | 0.19 |
Coronary artery disease NO (reference) YES | 1.98(1.26–3.08) | 0.003 | 1.45(0.81–2.60) | 0.21 |
Congestive heart failure NO (reference) YES | 1.56(0.72–3.38) | 0.26 | 0.89(0.28–2.82) | 0.84 |
Chronic liver disease NO (reference) YES | 2.06(0.85–4.95) | 0.11 | 1.72(0.64–4.61) | 0.28 |
Kidney disease NO (reference) YES | 1.14(0.64–2.02) | 0.65 | 0.90(0.46–1.77) | 0.76 |
Inflammatory arthritis NO (reference) YES | 1.33(0.66–2.71) | 0.42 | 1.10(0.49–2.48) | 0.81 |
Thyroid disorder NO (reference) YES | 0.77(0.40–1.45) | 0.42 | 0.61(0.31–1.18) | 0.14 |
Neuropsychiatric disorder NO (reference) YES | 1.82(1.09–3.02) | 0.022 | 1.82(0.97–3.41) | 0.059 |
Infection NO (reference) YES | 2.78(1.86–4.15) | 0.000 | 1.96(1.15–3.35) | 0.014 |
Stroke NO (reference) YES | 0.79(0.27–2.30) | 0.66 | 0.60(0.19–1.92) | 0.39 |
Functional gastrointestinal disorder NO (reference) YES | 1.50(0.68–3.29) | 0.31 | 0.78(0.30–1.98) | 0.59 |
Dementia NO (reference) YES | 0.39(0.08–1.92) | 0.25 | 0.84(0.15–4.54) | 0.83 |
Number of drugs 1–4 (reference) 5–9 ≥ 10 | -- | -- | 2.36(1.33–4.19) 5.33(2.49–11.40) | 0.000 0.003 0.000 |
Number of drugs (5–9) (reference) ≥ 10 | -- | -- | 2.27(1.26-4.00) | 0.006 |
of adverse drug reactions occurring in older outpatients. *Information on exact number of drugs was available for n = 523 cases. |
In a separate regression done involving older adults in whom information about the exact number of drugs was available (n = 523), a 1.96 times higher odds of ADRs due to DDIs was observed in patients with acute infectious etiology. With statistical significance, individuals receiving 10 drugs, or more were at 5.33- and 2.27-times higher odds of clinically manifest DDIs compared respectively to the group receiving 1–4 drugs and 5–9 drugs. A 2.36 times higher odds of clinically manifest DDIs were observed in older adults receiving 5–9 drugs compared to the group receiving 1–4 drugs. With marginal statistical significance (P value- 0.059), individuals with neuropsychiatric diseases were at 1.82 times higher odds of clinically manifest DDIs (Table 3b) No independent risk was observed however with Parkinsonism and coronary artery disease.
3.5. Preventability of clinically manifest DDIs:
Out of 199 ADRs caused due to DDIs, 133 (66.8%) can be avoided in the future by the adoption of suitable preventive measures. Figure 2 shows the details of such preventive measures. Avoiding the causative drug, optimal monitoring through physical examinations and blood investigations, and adopting of ‘start low and go slow’ policy could prevent 39 (29.3%), 38 (28.6%), and 36 (27.1%) cases of avoidable DDIs respectively. Other useful approaches include the prescription of better alternate therapy (4.5%), providing preventive management (3.7%), and avoiding, or decreasing the dose in altered pathophysiological states (2.2% each).