Drug Interactions Doctors Encounter

A patient is admitted to hospital, and an antibiotic is added to the regular medication list. The antibiotic is appropriate. The patient's chronic medicines are appropriate. Yet, a few days later, the patient develops an unexpected adverse effect.

What went wrong?

Sometimes, the problem is not an individual drug but the interaction between two medicines.

Drug–drug interactions (DDIs) are an important medication-safety concern, particularly in hospitalized patients receiving multiple medicines. However, a crucial distinction must be made: a potential drug interaction is not the same as a clinically manifested adverse event.

A 2020 systematic review and meta-analysis estimated the pooled prevalence of clinically manifested DDIs in hospitalized patients at 9.2% (95% CI 4.0–19.7%). A later meta-analysis reported a pooled prevalence of potential DDIs of 64.9% and a frequency of clinically evident DDIs of 17.17%. These estimates should not be directly compared as if they measured the same outcome; study populations, definitions and methods differed. [1,2]

An interaction database can identify a possible problem. Clinical pharmacology helps determine whether that interaction matters for an individual patient, what consequences are plausible, and what action is appropriate.

Let's explore a few interaction patterns through realistic clinical scenarios.

Clarithromycin and statins

A patient taking atorvastatin is prescribed clarithromycin for a bacterial infection. A few days later, the patient develops muscle pain and weakness.

Could the antibiotic be contributing?

What is happening?

Clarithromycin inhibits CYP3A4 and can also inhibit hepatic statin uptake transporters. It can substantially increase exposure to certain statins, particularly simvastatin and lovastatin. Atorvastatin exposure may also increase. These changes can raise the risk of myopathy and, rarely, rhabdomyolysis. [3]

The magnitude of the interaction differs among statins.

What should the clinician do?

• Check the exact statin and dose before prescribing clarithromycin.

• Consider an appropriate antibiotic alternative when clinically suitable.

• Follow product-specific recommendations regarding contraindications, temporary interruption or dose restrictions.

• Ask about new muscle pain, weakness or dark urine, and investigate suspected muscle injury with creatine kinase and renal function when indicated.

Clarithromycin is used for selected bacterial infections, including certain respiratory infections and Helicobacter pylori eradication regimens when appropriate. It is not the preferred antibiotic for every respiratory infection; indication, resistance patterns and patient factors matter.

Before adding a new drug, consider whether modifying an existing medicine would be safer than accepting the interaction.

Methotrexate and interacting medicines

A patient receiving low-dose weekly methotrexate for rheumatoid arthritis is prescribed trimethoprim–sulfamethoxazole for an infection. Subsequently, the patient develops oral ulcers and a falling blood-cell count.

Could the combination be responsible? What is happening?

Methotrexate toxicity can be increased by medicines that interfere with its elimination or add antifolate or bone-marrow-suppressive effects. Trimethoprim–sulfamethoxazole is an important combination to review because severe bone-marrow toxicity has been reported with concurrent use. [4]

Renal impairment, dehydration, dosing errors and additional interacting medicines can further increase risk. NSAIDs may also be relevant, particularly when renal function is impaired or methotrexate exposure is high.

Low-dose methotrexate used in rheumatology and high-dose methotrexate used in oncology require different clinical contexts when assessing risk.

What should the clinician do?

• Confirm that methotrexate is prescribed weekly, not daily, when used for common inflammatory diseases.

• Review antibiotics and other medicines before prescribing them.

• Consider an appropriate alternative to trimethoprim–sulfamethoxazole where indicated.

• Assess renal function and monitor blood counts and other relevant laboratory parameters.

• Investigate mucositis, unexplained fever, bruising or cytopenias promptly.

Clinical pharmacology pearl: Methotrexate interactions cannot be judged from the drug pair alone. Dose, indication, renal function and additive toxicity all influence risk.

Direct oral anticoagulants and CYP3A4/P-glycoprotein modifiers

A patient receiving apixaban for atrial fibrillation is prescribed a new medicine. An interaction alert appears, but the patient has no bleeding and the anticoagulant dose has not changed.

Is the combination safe? What is happening?

Direct oral anticoagulants (DOACs) differ in their pharmacokinetics. All are substrates of P-glycoprotein (P-gp), while CYP3A4-mediated metabolism is particularly relevant to apixaban and rivaroxaban.

Strong inhibitors of relevant pathways may increase drug exposure and bleeding risk. Strong inducers, including rifampicin, may reduce exposure and compromise anticoagulant efficacy. The magnitude depends on the specific DOAC, interacting drug, renal function and other patient factors. [5]

What should the clinician do?

• Identify the exact DOAC rather than treating the class as pharmacokinetically uniform.

• Review the interacting drug's effects on P-gp and CYP3A4.

• Check renal function, indication, dose and other medicines that increase bleeding risk.

• Follow product-specific recommendations on avoidance, dose adjustment or alternative therapy.

• Avoid unvalidated dose changes based solely on a theoretical interaction.

NSAIDs and antiplatelet medicines can also increase bleeding risk through pharmacodynamic effects, even without substantially increasing DOAC concentrations.

An interaction can cause harm by increasing drug exposure—or by reducing it enough to compromise treatment efficacy.

Rifampicin: The interaction that can make other medicines less effective

A patient taking several long-term medicines begins rifampicin for tuberculosis. Over the following weeks, the response to one of the patient's other treatments deteriorates.

Could enzyme induction explain the change? What is happening?

Rifampicin activates the pregnane X receptor, increasing the expression of drug-metabolizing enzymes and transporters, including CYP3A4 and P-glycoprotein. Concentrations of many susceptible medicines can consequently fall, potentially causing treatment failure. [6]

Examples include selected anticoagulants, immunosuppressants, hormonal contraceptives and cardiovascular drugs. The magnitude varies by medicine.

Importantly, enzyme induction develops over time and does not disappear immediately when rifampicin is stopped.

What should the clinician do?

• Review the complete medication list before rifampicin is initiated.

• Identify medicines for which reduced exposure could cause serious harm.

• Select alternatives or monitoring strategies where recommended.

• Reassess the regimen when rifampicin is discontinued, because previously increased doses or alternative treatments may then become unsafe.

When an interacting medicine is started or stopped, ask two questions: What happens to the other drug now, and what happens when the interaction resolves?

Linezolid and serotonergic medicines

A patient taking an antidepressant requires linezolid for a serious infection. An interaction alert raises concern about serotonin syndrome.

Should linezolid automatically be avoided? What is happening?

Linezolid has monoamine oxidase-inhibiting activity. When combined with serotonergic medicines, it can increase the risk of serotonin toxicity, which may include agitation, sweating, fever, tremor, clonus and hyperreflexia.

However, the risk is not inevitable. A 2023 systematic review found that reported serotonin toxicity was uncommon, including among patients receiving linezolid with serotonergic agents. This supports individualized assessment rather than assuming every combination will cause harm. [7]

What should the clinician do?

• Review the serotonergic medicines, doses and availability of alternative antibiotics.

• Assess the urgency and expected benefit of linezolid.

• Follow current prescribing information and local protocols for managing the combination.

• If co-administration is necessary, establish an appropriate monitoring and management plan.

• Evaluate new neuromuscular, autonomic or mental-status changes promptly.

A serious potential interaction deserves attention, but management should reflect both the probability and severity of harm and the consequences of withholding effective treatment.

Multiple QT-prolonging medicines

A hospitalized patient receives an antibiotic and an antiemetic, both associated with QT prolongation. The patient is older, has impaired renal function and has developed hypokalaemia.

The medication list contains the warning, but the patient's clinical context determines its urgency.

What is happening?

Some medicines delay cardiac repolarization. Combining them may produce additive effects, while other medicines can increase concentrations of QT-prolonging drugs through metabolic inhibition.

The risk of torsades de pointes is influenced by baseline QTc, bradycardia, electrolyte abnormalities, cardiac disease, dose and the number of contributing medicines. [8]

What should the clinician do?

• Review the indication and availability of alternatives.

• Correct potassium and magnesium abnormalities when present.

• Assess the baseline ECG and repeat it when clinically indicated.

• Consider renal function, dose and metabolic interactions.

• Reassess treatment promptly if QTc becomes markedly prolonged or concerning symptoms develop.

A QTc above 500 ms generally raises concern for torsades risk, although interpretation must account for the clinical situation and the method used to correct QT for heart rate.

Two QT-prolonging medicines do not produce a predictable, simple doubling of risk. Assess the combination alongside the patient's susceptibility and modifiable risk factors.

NSAIDs and anticoagulants

A patient taking an oral anticoagulant develops musculoskeletal pain and begins taking an NSAID. The patient assumes the combination is safe because the anticoagulant dose has not changed.

What is happening?

NSAIDs can damage gastrointestinal mucosa and impair platelet function. Anticoagulants interfere with coagulation. These effects can combine to increase bleeding risk without requiring a pharmacokinetic interaction.

A systematic review and meta-analysis of 27 studies found that concomitant NSAID use was associated with increased risks of any bleeding and gastrointestinal bleeding among patients taking oral anticoagulants. In the DOAC subgroup, the association with major bleeding was not statistically significant, although the estimates were imprecise. [9]

What should the clinician do?

• Ask specifically about prescribed and over-the-counter NSAIDs.

• Consider non-NSAID alternatives when suitable.

• If an NSAID is necessary, assess the indication, dose, duration, gastrointestinal risk and other bleeding-promoting medicines.

• Counsel patients to seek urgent assessment for significant bleeding, black stools, vomiting blood or other concerning symptoms.

A proton pump inhibitor may reduce upper gastrointestinal risk in selected patients, but it does not eliminate all bleeding risk.

Not every important interaction changes drug concentration. Some combinations increase harm because two medicines affect different components of the same physiological process.

Proton pump inhibitors and clopidogrel

A patient taking clopidogrel after coronary intervention is also prescribed omeprazole for gastrointestinal protection. An interaction alert suggests that omeprazole could reduce clopidogrel activation.

Should the PPI be stopped? What is happening?

Clopidogrel is a prodrug that requires metabolic activation, involving CYP2C19. Omeprazole can inhibit CYP2C19 and reduce formation of clopidogrel's active metabolite.

However, a pharmacokinetic or platelet-function effect does not automatically establish a clinically meaningful increase in cardiovascular events.

In the COGENT randomized trial, patients receiving aspirin and clopidogrel who were assigned omeprazole had fewer gastrointestinal events. The trial did not detect a significant difference in cardiovascular events, although it ended prematurely and could not exclude a clinically meaningful difference. [10]

What should the clinician do?

• Assess the patient's gastrointestinal bleeding risk and indication for acid suppression.

• Consider the specific PPI and its interaction profile.

• Follow current cardiology guidance and product information when selecting gastroprotection.

• Avoid stopping clinically indicated gastroprotection solely because an interaction alert appears.

Mechanistic plausibility is important, but patient outcomes help determine whether an interaction should change practice.

A practical bedside approach: Five questions before acting on a DDI alert

When an interaction is identified, ask:

1. What is the mechanism?
Does the combination affect metabolism, transport, absorption, renal elimination or a shared physiological pathway?

2. How large is the effect likely to be?
Is the interaction established, dose-dependent, variable or supported mainly by theoretical evidence?

3. What makes this patient vulnerable?
Consider age, renal and hepatic function, comorbidities, dose, duration and other medicines.

4. What outcome are we trying to prevent?
Examples include bleeding, nephrotoxicity, arrhythmia, treatment failure and excessive drug exposure.

5. What is the least harmful effective intervention?
Options include avoiding the combination, choosing an alternative, adjusting a dose, monitoring laboratory parameters, obtaining a test or using therapeutic drug monitoring.

The correct response is not always discontinuation. Sometimes the combination is necessary, and the safest strategy is targeted monitoring.

Hospitals can create interactions as well as inherit them

A study of 851 internal medicine patients evaluated potential DDIs at hospital admission, during hospitalization and at discharge. It found that 47% of major or moderate potential DDIs identified at discharge were attributable to medication changes made during the hospital stay. [11]

This reminds us that medication safety cannot end with admission medication reconciliation. New antibiotics, analgesics, anticoagulants and other treatments may introduce interactions that were not present when the patient arrived.

Medication review is especially important at three points:

• Admission: identify interactions in the patient's existing regimen.

• During hospitalization: reassess interactions whenever treatment changes.

• Discharge: ensure that medication changes and monitoring plans are communicated clearly.

Conclusion

Drug interactions are not simply a database problem. They are a clinical reasoning problem.

Clarithromycin can increase exposure to susceptible statins. Methotrexate toxicity can be amplified by interacting medicines. Rifampicin can reduce the exposure of other drugs. NSAIDs can add bleeding risk to anticoagulant treatment. Other combinations may alter drug concentrations or laboratory measures without producing clinically meaningful harm; some can cause serious toxicity or treatment failure.

The challenge is to recognize the interactions that matter for the patient in front of us.

The EvidenceRx takeaway: Do not stop at “These drugs interact.” Ask: “What is the likely clinical consequence in this patient, and what action will best preserve safety and treatment effectiveness?”

This educational article supports clinical reasoning and does not replace product-specific prescribing information, institutional protocols or specialist advice. Management should be individualized to the indication, patient factors and urgency of treatment.

References

1. de Oliveira Filho AD, Santos TNGDA, Macieira GMDC, et al. Prevalence of clinically manifested drug interactions in hospitalized patients: A systematic review and meta-analysis. PLoS One. 2020;15(7). doi:10.1371/journal.pone.0235353. https://pubmed.ncbi.nlm.nih.gov/32609783/

2. Aksoy N, Ozturk N. A meta-analysis assessing the prevalence of drug-drug interactions among hospitalized patients. Pharmacoepidemiol Drug Saf. 2023. https://pubmed.ncbi.nlm.nih.gov/37705139/

3. Christensen MMH, Haastrup MB, Øhlenschlaeger T, et al. Interaction potential between clarithromycin and individual statins—a systematic review. Basic Clin Pharmacol Toxicol. 2020;126(4):307–317. doi:10.1111/bcpt.13343. https://pubmed.ncbi.nlm.nih.gov/31628882/

4. Kivity S, et al. Practical clinical pharmacology and drug interactions of low-dose methotrexate therapy in rheumatoid arthritis. Clin Pharmacokinet. 1996. https://pubmed.ncbi.nlm.nih.gov/8535644/

5. Steffel J, Collins R, Antz M, et al. 2021 European Heart Rhythm Association Practical Guide on the use of non-vitamin K antagonist oral anticoagulants in patients with atrial fibrillation. Europace. 2021;23(10):1612–1676. doi:10.1093/europace/euab065. https://academic.oup.com/europace/article/23/10/1612/6247378

6. Niemi M, Backman JT, Fromm MF, Neuvonen PJ, Kivistö KT. Pharmacokinetic interactions with rifampicin: clinical relevance. Clin Pharmacokinet. 2003;42(9):819–850. doi:10.2165/00003088-200342090-00003. https://pubmed.ncbi.nlm.nih.gov/12882588/

7. Elbarbry F, Moshirian N. Linezolid-associated serotonin toxicity: a systematic review. Eur J Clin Pharmacol. 2023;79(7):875–883. doi:10.1007/s00228-023-03500-9. https://pubmed.ncbi.nlm.nih.gov/37129603/

8. Khatib R, Sabir FRN, Omari C, et al. Managing drug-induced QT prolongation in clinical practice. Postgrad Med J. 2021;97(1149):452–458. doi:10.1136/postgradmedj-2020-138661. https://pubmed.ncbi.nlm.nih.gov/33122341/

9. Zheng Y, et al. Co-administered oral anticoagulants with nonsteroidal anti-inflammatory drugs and the risk of bleeding: A systematic review and meta-analysis. Thromb Res. 2025. https://pubmed.ncbi.nlm.nih.gov/39492109/

10. Bhatt DL, Cryer BL, Contant CF, et al. Clopidogrel with or without omeprazole in coronary artery disease. N Engl J Med. 2010;363:1909–1917. doi:10.1056/NEJMoa1007964. https://pubmed.ncbi.nlm.nih.gov/20925534/

11. Vonbach P, Dubied A, Krähenbühl S, Beer JH. Prevalence of drug-drug interactions at hospital entry and during hospital stay of patients in internal medicine. Eur J Intern Med. 2008;19(6):413–420. doi:10.1016/j.ejim.2007.12.002. https://pubmed.ncbi.nlm.nih.gov/18848174/

12. Bethi Y, et al. Prevalence and predictors of potential drug-drug interactions in patients of internal medicine wards of a tertiary care hospital in India. Eur J Hosp Pharm. 2019;26(6):317–321. https://pubmed.ncbi.nlm.nih.gov/31157049/