Critical Care Medicine

Severe drug reactions

Assessment of Adverse Drug Events

Also known as: Adverse Drug Reactions, Side Effects; Drug-induced injury

Related terms: Medication Safety

1. Description of the problem


An adverse drug event is defined as injury related to a drug. There are many variations of definitions for an adverse drug event but overall, injury is an accepted endpoint. Injury itself requires clarification. Injury should be defined as end-organ damage. So, drug-induced events such as abnormal electrolyte values that have not yet resulted in injury (ie, arrhythmia) would not be considered an adverse drug event but are considered drug-related hazardous conditions, since they are an antecedent to injury.

What every clinician needs to know

Adverse drug events are more frequent and severe in the critically ill population. Patients in the intensive care unit are at risk for adverse drug events because they are in a fast-paced environment, they receive about two times more medications and more high-risk medications than ward patients, they receive drugs administered intravenously and they have changing organ function that alerts the pharmacokinetics of medications.

Clinical features

Adverse drug events can manifest as numerous signs and symptoms depending on the drug. Common types of adverse drug events in critically ill patients are acute kidney injury (AKI), cognitive impairment, dysrhythmias and bleeding.

Key management points

The key to the management of adverse drug events is to prevent the transition of an event such as a drug related hazardous condition from becoming patient injury or at minimum prevent temporary to permanent injury. A first step is to assure that the reason for the injury is identified, which includes differentiating drug-induced and disease-induced events. After conclusive rationale is identified for suspecting a drug-related cause, then isolating the specific drug is necessary. There are adverse drug event causality instruments that can assist with this determination (refer to diagnosis section). Safe medication use guidelines are being developed by the Society of Critical Care Medicine.

2. Emergency Management

1. Stabilize the drug-induced end-organ function.

2. Provide antidote medications as necessary. The following represents common antidotes and corresponding dosages (obtained from Lexicomp®):

Naloxone: An initial dose of 0.4 to 2 mg intravenously (IV) is recommended. If the desired degree of reversal and improvement in respiratory function is not obtained, the dose may be repeated at 2- to 3-minute intervals. If no response is observed after 10 mg have been administered, the diagnosis of opioid toxicity should be questioned.

Flumazenil: For known or suspected benzodiazepine overdose, initially administer 0.2 mg IV over 30 seconds; if desired level of consciousness not obtained after an additional 30 seconds, administer 0.3 mg IV over 30 seconds; further doses of 0.5 mg IV over 30 seconds may be given at 1-minute intervals if needed to MAX total dose of 3 mg; patients with only partial response to 3 mg may require additional slow titration to a total dose of 5 mg; if no response 5 minutes after receiving total dose of 5 mg, overdose is unlikely to be benzodiazepine and further treatment with flumazenil will not help.

Vitamin K: For reversal of anticoagulant, 2.5 to 25 mg ORALLY (rarely up to 50 mg); if prothrombin time is not satisfactory within 12 to 48 h, repeat dose, alternatively 2.5 to 25 mg IV or subcutaneously (rarely up to 50 mg); if prothrombin time is not satisfactory within 6 to 8 h, repeat dose.

N-Acetylcysteine: For acetaminophen overdose, if body weight 40 kg or greater, IV loading dose, 150 mg/kg in 200 mL of compatible solution (D5W, 0.45NS, or WFI) IV over 60 minutes, then 50 mg/kg in 500 mL of solution IV over 4 h, followed by 100 mg/kg in 1000 mL of solution IV over 16 h.

Physostigmine: For anticholinergic toxicity, administer at a slow controlled rate not exceeding 1 mg/min; may repeat dosage at intervals of 10 to 30 minutes.

Digoxin Immune Fab (Ovine) For digoxin toxicity: Acute ingestion of unknown amounts, 10 vials (380 mg) IV, observe response; repeat with 10 vials as needed; life-threatening, 20 VIALS (760 mg) IV.

Digoxin toxicity, acute ingestion of known amount of digoxin, each vial (38 mg) IV will bind approximately 0.5 mg digoxin; bioavailability of digoxin is 0.8 for 0.25 mg tablets or 1.0 for 0.2 mg Lanoxicaps; use the following formula, dose (in vials) = digoxin ingested (mg) x bioavailability / 0.5 mg of digoxin bound per vial.

Digoxin toxicity, digoxin: chronic digoxin toxicity, 6 vials (228 mg) IV or use the following formula: dose (in vials) = (serum digoxin concentration in ng/mL) x (wt in kg)/100

Thiosulfate/sodium nitrate: For sodium nitroprusside toxicity, treat with sodium nitrate 4 to 6 mg/kg (0.2mL/kg of a 3% solution) IV over 2 to 4 minutes, followed by sodium thiosulfate 150 to 200 mg/kg IV (cyanide antidote kit). Hydroxocobalamin 5g IV over 15 minutes (adult dose) is an alternative treatment for cyanide toxicity.

3. Elucidate drug from disease. The response to the antidote may provide insight into the diagnosis of the drug-induced event. For example, if the patient responded to the flumazenil then a benzodiazepine is the culprit. If there is no response to the flumazenil then the event may be disease induced. The temporal sequence between drug and event is also helpful in making this determination. Also, additional tests may assist with diagnosis such as a urine sodium for drug-induced syndrome of antidiuretic hormone.

4. Identify and discontinue suspect medication.

5. Provide alternate therapy for discontinued medication if needed. Evaluate whether the patient may potentially cross-react to this drug.

Management Points

  1. Evaluate the common medication causes of the event.

  2. Assess temporal sequence for potential drug-related causes in relation to the event.

  3. If possible, discontinue the drug with the greatest likelihood of causality then streamline the medication therapy removing potentially contributing medications. An atypical reaction may require a primary literature review instead of referring to a tertiary drug information resource.

  4. Consider alterations in renal and hepatic function related to critical illness as a possible reason for altered clearance of the medications. Maybe a dosing adjustment will suffice instead of drug discontinuation.

  5. Consider drug-drug interactions as a possible contributing factor that may be altering the pharmacokinetics of the medication and contributing to the occurrence of the event.

  6. Be sure that an assessment of patient allergies was conducted, as a medication error may have lead to the event.

  7. If this adverse drug event is related to a new allergic reaction for the patient, then be sure the allergy is documented in the patient's permanent records.

3. Diagnosis

Diagnostic criteria

There are several instruments available to aid in the assessment of adverse drug events, which is further described in the annotated bibliography. The majority of these instruments contain common themes that assist with evaluation:

1. Has this reaction been reported before in association with this drug or its drug class?

2. Has the patient had a reaction to this drug or similar drugs previously?

3. Are there laboratory tests that suggest that this drug is the culprit?

4. Is the temporal sequence between the event and drug administration suggestive of a relationship?

5. When you discontinue the drug or de-escalate the dose, does the event subside?

6. If you have opted to readminister the suspect medication to the patient, does the reaction reappear?

Confirming assessment

Confirming the diagnosis of an adverse drug event, ideally, requires discontinuation of the drug when possible to determine if the signs and symptoms are relieved.

Other possible diagnosis

An evaluation of alternate, non–drug-related causes is necessary. The discontinuation of the drug (if possible) or administration of a diagnostic drug (antidote) is the ideal approach to diagnosis.

Confirmatory tests

A confirmatory test that could be performed is rechallenging the patient to the drug to determine the occurrence of similar signs and symptoms. This criteria of evaluation is included in some of the causality instruments. However, based on the tenuous nature of the patient and the severity of the event, this may not be advised. Sometimes the rechallenge occurs naturally, for example, each time the patient gets a dose of insulin at night and consequently becomes hypoglycemic by morning, then after a couple of these occurrences the diagnosis would be confirmatory.

4. Specific Treatment

First-line therapy

The first-line therapy of an adverse drug event is discontinue the offending drug. Some antidotes exist for specific drugs. Examples of treatments for common adverse drug reactions are provided below.

Examples of treatment reversals

Heparin reversal

Protamine sulfate, 1 mg IV for every 100 units of heparin remaining in patient; if 30 minutes have elapsed since the injection of heparin, one-half the dose may be sufficient; maximum 50 mg given over 10 minutes.

Low molecular weight heparin (LMWH) reversal

Mild to moderate toxicity: Treatment is symptomatic and supportive. Monitor for clinical evidence of bleeding.

Severe toxicity: Treatment is symptomatic and supportive. Treat patients with severe bleeding with protamine sulfate. Blood transfusions or fresh frozen plasma may be indicated in addition to protamine sulfate.

In patients with strongly suspected or confirmed heparin-induced thrombocytopenia (HIT), with or without thrombosis, discontinue LMWH agent and start an alternative, nonheparin anticoagulant (lepirudin, argatroban). An anaphylactoid or anaphylactic reaction may require aggressive airway management and support.

Opiate reversal

See naloxone above

Benzodiazepine reversal

See flumazenil above

Hypotension reversal

Administer IV fluids to achieve the desired blood pressure. If needed, administer a vasopressor or inotrope depending on the cause of the hypotension.

AKI reversal

While there is no therapy to directly treat AKI, renal replacement therapy is used to remove fluids and toxic substances. The key is to prevent AKI by maintaining hemodynamic support, treating sepsis, and minimizing exposure to nephrotoxic agents, especially nephrotoxins used in combination.

Dysrhythmia reversal

For acute symptomatic bradycardia, administer atropine 0.5 mg IV every 3 to 5 minutes to a maximum total dose of 3 mg.

For tachycardia, administer the appropriate antidysrhythmic agent depending on the type of tachycardia.

Hyperkalemia reversal

Calcium chloride: 5mL of 10% solution IV over 2 minutes (stop infusion if bradycardia develops)

Calcium gluconate: 10mL of 10% solution IV over 2 minutes (stop infusion if bradycardia develops)

Dextrose: 1 to 2 amps D50W and 5 to 10 units regular insulin IV

Sodium polystyrene sulfonate (kayexalate) 25 to 50g mixed with 100mL of 20% sorbitol orally or rectally

Hyperglycemia reversal

5-10 Units regular insulin and 1 to 2 amps D50W IV bolus

Sodium bicarbonate: 1mEq/kg slow IV push or continuous IV infusion; not to exceed 50 to 100 mEq

Albuterol 5 mg mixed with 3mL isotonic saline via high-flow nebulizer q20min as tolerated

Furosemide 20 to 40 mg IV push in patients not already on this drug. Double daily PO dose as IV slow push in patients already taking this drug.

Refractory cases

Re-evaluate the drug- and disease-related causes of the event in a refractory case. Potentially the event was not drug induced or another drug was contributing to the event.

5. Disease monitoring, follow-up and disposition

Expected response to treatment

Determine the elimination half-life of the suspected drug and calculate 4 to 5 half-lives for an expected time in which the drug should be cleared from the bloodstream. This is the latest time in which you would expect improvement of signs and symptoms of an adverse drug event for most drugs.

Unexpected response

If the patient does not improve within 4 to 5 half-lives then you may consider that the half-life of the drug may be longer than anticipated. A patient with a decline in renal or hepatic function may have a half-life that is longer than typical. For example, a patient with an altered mental status in which prolonged intravenous morphine administration is suspected does not become communicative after 8 to 10 hours (approximately a 2-hour half-life multiplied by 4 to 5).

An assessment of the patient's renal function indicates that his creatinine clearance is 20 mL/min. The half-life of morphine in this patient is longer than 2 hours since the drug and its active metabolite, 6-morphine glucoronide, can accumulate in patients with renal dysfunction.

Also, in patients with drug-induced end-organ damage such as AKI, it may take longer than 4 to 5 half-lives of the drug for this injury to resolve. For these types of events, it may simply take additional time. It is necessary to monitor the patient's renal function to assure it is improving. If the patient's end organ damage worsens after the suspected offending drug is discontinued, then it is time to consider other possible causes including drugs and diseases.


After identifying the offending drug it may take some time for the end-organ damage to resolve. If resolution does not occur in an appropriate amount of time, a further evaluation of drug and disease states is necessary.

When an antidote is administered, follow-up should include evaluation of the possible side effects of the antidote. For example, vitamin K can result in excessive reversal of the INR or sodium polystyrene can result in excessive reversal of the serum potassium.

An evaluation of, if and when to restart the drug or begin an alternate therapy should be completed. You would not want to leave a condition untreated if medically necessary.

Depending on the event, certain drugs should never be restarted (eg, anaphylaxis), so be sure that there is appropriate documentation in the patient's medical record to prevent reoccurrence of the event. Also, be sure to provide disclosure to the patient and/or their family because their awareness could avoid a repeated reaction.

Reactions should be reported to the appropriate local and state regulatory bodies, as to increase awareness of the adverse drug event and allow for possible prevention for other patients in the future.


The cause of many ADEs is unexplained. Contributing factors for ADEs can be drug-drug interactions or altered pharmacokinetics of the drug. In some cases ADEs are the result of a medication error. In approximately 1% to 29% of cases the manifestation of an adverse drug event is a medication error, although 10% appears to be the more common rate cited.

The inappropriate use of drugs at any stage in the medication use process (prescribing, distributing, administering, monitoring) is considered a medication error. Medication errors can result in an adverse drug event but for the majority of medication errors, drug-related injury does not occur. Also, the majority of adverse drug events are not the result of a medication error.


Frequency compared to non-ICU patients

The rate of preventable ADEs and potential ADEs is nearly twice as frequent in ICUs as in general care units, with 19 and 10 per 1000 patient-days, respectively. After adjusting for the number of drugs used prior to the ADE, there were no differences in event rates between ICUs and general care units.

Frequency in the ICU

ADEs due to medication errors that have the potential to cause injury but do not result in injury are known as potential ADEs or near misses. An example would be a patient that received a drug despite a documented allergy to that drug but this error did not cause injury to the patient. The rate of potential ADEs in the ICU is 14 to 117 per 100 patient days. The rate of preventable ADEs or those ADEs that were associated with a medication error is 5.2 to 12.8 per 100 patient days.


ADEs and medication errors occurring in the ICU are more harmful than those occurring in non-ICUs. Also, patients with an ADE have an ICU length of stay that is twice that of a non-ICU patient having an ADE. The rate of potentially life-threatening errors in the ICU is 28 per 1000 new prescriptions.

Special considerations for nursing and allied health professionals.

Tracking drug-induced reactions before they result in harm to the patient would be an optimal preventive approach. Trigger systems (automated and manual) have been recommended by the Institute for Safe Medication Practices to track and prevent events. A structured institutional surveillance system that includes methods such as voluntary reporting and direct observation would also be beneficial.

Special considerations for nursing- Track events before they result in harm- surveillance systems- (suggested references for this section to those who will be completing it)

Stockwell, DC, Kane-Gill, SL. "Developing a patient safety surveillance system to identify adverse events in the intensive care unit". Crit Care Med. vol. 38. 2010. pp. S117-25.

Keers, RN, Williams, SD, Cooke Ashcroft, DM. "Prevalence and nature of medication administration errors in health care settings; a systematic review of direct observational evidence". Ann Pharmacother. vol. 47. 2013. pp. 237-56.

(Direction observation is a method of medication error detection that allows for a focus on administration errors. This review discusses studies that used direction observation and the medication errors identified with this approach.)

You must be a registered member of ONA to post a comment.

Sign Up for Free e-newsletters

Regimen and Drug Listings


Bone Cancer Regimens Drugs
Brain Cancer Regimens Drugs
Breast Cancer Regimens Drugs
Endocrine Cancer Regimens Drugs
Gastrointestinal Cancer Regimens Drugs
Genitourinary Cancer Regimens Drugs
Gynecologic Cancer Regimens Drugs
Head and Neck Cancer Regimens Drugs
Hematologic Cancer Regimens Drugs
Lung Cancer Regimens Drugs
Other Cancers Regimens
Rare Cancers Regimens
Skin Cancer Regimens Drugs