R E V I E W S & A N A LY S E S Analysis of Reported Drug Interactions: A Recipe for Harm to Patients Matthew Grissinger, RPh, FISMP, FASCP INTRODUCTION Manager, Medication Safety Analysis Pennsylvania Patient Safety Authority Drug interactions may occur inside (drug-drug interaction [DDI]) or outside (drug incompatibility) the body. When an interaction occurs, the pharmacological effect and/or physical characteristics of one or both drugs is altered. As a result, the phar- ABSTRACT macological effect of one or both drugs may be increased or decreased, or a new and Patients admitted to a hospital often unanticipated adverse effect may occur. receive many medications. Concomitant DDIs may result from pharmacokinetic interactions (absorption, distribution, metabo- use of multiple medications can lead lism, and excretion) or from interactions at drug receptors. Often these interactions are to drug interactions that occur prior to not benign. The risk of patient harm and the potential financial burden from DDIs administration (i.e., drug incompatibility) is significant. For example, DDIs have been estimated to account for up to 30% of all or after ingestion or injection (i.e., drug- adverse drug events (ADEs).1,2 Certain patient factors (e.g., age, impaired renal func- drug interaction). When an interaction tion, current medications) can increase the risk and potential harm from DDIs. occurs, the effects and characteristics A drug incompatibility occurs when two or more injectable drugs are mixed and the of the drugs may be altered, leading to stability or structure of the drugs is altered by physical or chemical reactions. The increased or decreased drug activity or resulting solution is often no longer optimal or safe for the patient. For example, physi- new and unanticipated adverse effects. cal changes to the solution may lead to precipitate formation that can cause catheter Analysts searched the Pennsylvania occlusion and embolism and can contribute to a range of ADEs, from thrombophle- Patient Safety Reporting System database bitis to multi-organ failure. Additionally, the reduction or elimination of the active for reports submitted as “Medication drug can lead to a therapeutic failure. The consequences of drug incompatibilities can error/Monitoring error/Drug-drug inter- be particularly severe in neonatal and pediatric patients. Unfortunately, inappropriate action” that occurred from April 2009 Y-site combinations (used to infuse multiple medications through one venous access through March 2016. A total of 815 point) of continuously infused drugs may be common. In an observational study of 13 event reports were included in the final intensive care units (ICUs) in Canada, the prevalence of inappropriate drug combina- analysis. The most commonly reported tions was 8.5% among all patients but rose to 18.7% in patients receiving at least two type of drug interaction involved drug continuously infused drugs.3 incompatibilities (41.8%, n = 341) and drug-drug interactions (27.9%, n = 227). As the number of approved drugs increases, the risk for DDIs and drug incompat- Healthcare facilities can help reduce the ibilities increases. Pennsylvania Patient Safety Authority analysts have not previously opportunity for drug interactions reaching explored drug interactions reported through the Authority’s Pennsylvania Patient patients by addressing all areas of the Safety Reporting System (PA-PSRS). With this analysis of drug interactions reported to medication-use process and not relying the Authority, analysts sought to characterize contributing factors and identify appro- solely on the effectiveness of alerts when priate system-based risk reduction strategies to help facilities identify potential risk and orders are entered into electronic health minimize potential patient harm. records. (Pa Patient Saf Advis 2016 Dec;13[4]:137-148.) METHODS Analysts queried the PA-PSRS database for reports submitted as “Medication Error/ Monitoring error/Drug-drug interaction” that occurred from April 2009 through March 2016. This query yielded 870 event reports. Fifty-five reports (6.3%) were excluded from final analysis because upon review of the event’s description, the error did not involve a drug interaction. A total of 815 event reports remained for final analysis. The medication name, patient care area, event type, event description, phase(s) of the medication use process, and harm score, adapted from the National Coordinating Council for Medication Error Reporting and Prevention (NCC-MERP) harm index,4 were provided by the reporting facility. When a medication name data field was left blank or incomplete, but the name was provided in the event description, an analyst adjusted the medication name field appropriately. Reports were categorized into four categories: DDI, therapeutic duplication, contraindication, and drug incompatibility. The drug classes involved in the events also were identified. Intravenous solutions with or without electrolytes were considered to be drugs for this analysis. In the context of Vol. 13, No. 4—December 2016 Pennsylvania Patient Safety Advisory Page 137 ©2016 Pennsylvania Patient Safety Authority R E V I E W S & A N A LY S E S this analysis, therapeutic duplications are Figure 1. Types of Drug Interactions Identified in Events Reported to the Pennsylvania Patient Safety Authority, April 2009 through March 2016 (N = 815) errors when two or more medications from a similar pharmacotherapeutic class and for similar indications are prescribed 66 and/or administered to a patient. Error (8.1%) reports were further evaluated to identify contributing factors and potential system- based risk reduction strategies. 181 341 Drug incompatibility (22.2%) (41.8%) RESULTS Drug-drug interaction Results were categorized by the type of Therapeutic duplication drug interaction. The largest percent- age of drug interaction events were drug Drug contraindication 227 incompatibilities (Figure 1). (27.9%) MS16832 Drug Incompatibilities Most of the drug incompatibility events (88.3%, n = 301 of 341) reached the patient (harm score C through I). Patient harm was noted in only 0.6% (n = 2) of Figure 2. Harm Scores of Drug Interaction Events as Reported to the Pennsylvania the drug incompatibility reports and were Patient Safety Authority, April 2009 through March 2016 (N = 815)* reported as errors that may have contrib- uted to or resulted in temporary harm NUMBER to the patient and required intervention OF REPORTS (harm score E; Figure 2). 250 216 More drug incompatibility events involved adult patients (41.6%, n = 142 of 341) than 200 elderly or pediatric patients (Figure 3). Overall, 48 unique patient care areas 150 were associated with a drug incompat- 100 93 ibility event, with medical/surgical units 83 involved in 13.2% (n = 45 of 341) of the 69 69 56 events. Taken together, intensive care 45 45 50 26 units (ICUs; e.g., cardiac ICU, neonatal 24 17 12 11 3 6 4 6 4 2 13 MS16833 ICU), where patients are often on mul- 2 2 3 1 0 1 1 1 tiple intravenous (IV) medications, were 0 A B1 B2 C D E F cited in 29.6% (n = 101) of reports. HARM SCORE Analysts reviewed event description fields to determine whether an actual incompat- Drug incompatibility (n = 341) Drug-drug interaction (n = 227) ibility took place or whether the reported Therapeutic duplication (n = 181) Drug contraindication (n = 66) event was a “close call” (e.g., nursing iden- tified the potential for an incompatibility * There were no reported events with harm score G, H, or I. before administration to the patient). Almost one out of five drug incompat- ibility reports (18.8%, n = 64) mentioned the formation of a precipitate and 2.1 % (n = 7) stated an infiltration took place. The largest percentage of reported events Page 138 Pennsylvania Patient Safety Advisory Vol. 13, No. 4—December 2016 ©2016 Pennsylvania Patient Safety Authority (49.3%, n = 168) described situations in Figure 3. Age of Patients Involved in Drug Interactions, April 2009 through which the potential for incompatibility was March 2016 (N = 815) identified before administration (i.e., close NUMBER call) and almost 30% (28.2%, n = 96) OF REPORTS described events where two incompatible 160 drugs were infused, but no visible precipi- 142 tate formed. 140 117 115 Drug incompatibility reports cited 117 120 109 unique medications. The most com- 98 100 mon medications mentioned in reports 82 included IV fluids (e.g., dextrose 5%, 80 69 sodium chloride 0.9%; 16.7%, n = 57), 60 heparin (14.4%, n = 49), pantoprazole 37 40 28 (8.5%, n = 29), and parenteral nutri- 20 14 MS16834 tion solutions (8.5%, n = 29). The most 3 1 common pairs of medications included 0 cefTRIAXone with lactated Ringer’s Drug Drug-drug Therapeutic Drug solution (2.6%, n = 9) and heparin with incompatibility interaction duplication contraindication diltiazem (2.6%, n = 9). Figure 4 shows (n = 341) (n = 227) (n = 181) (n = 66) the most common pairs of medica- tions involved in these reported events. TYPE OF DRUG INTERACTION The most common reported pair of Pediatric (0 to 17 years) Adult (18 to 64 years) medications that led to actual precipitate formation was the combination of cip- Elderly (65 years or older) rofloxacin and hydration solutions with electrolytes (9.4%, n = 6 of 64). Figure 4. Most Common Pairs in Drug-Drug Interaction Reports Identified as Drug Incompatibilities, as Reported to the Pennsylvania Patient Safety Authority, April Following are examples of reported errors 2009 through March 2016 (n = 341) involving patients receiving IV therapy and a precipitate occurred:* DRUG PAIR Pentamidine mixed with NSS [normal cefTRIAXone and 9 (2.6%) saline solution]. Precipitate found in lactated Ringer tubing. Medication stopped and tub- Heparin and diltiazem 9 (2.6%) ing taken to pharmacy for discussion. Ciprofloxacin and Pentamidine found to have been mixed 7 (2.1%) IV fluids with electrolytes* with NSS even though the label said it was mixed with D5 [dextrose 5%]. Heparin and vancomycin 7 (2.1%) No indication that this impacted Furosemide and 5 (1.5%) the patient or the IV site. Patient magnesium sulfate completed his medications and chemo- Furosemide and milrinone 5 (1.5%) therapy and went home as planned. Patient had a piggyback line infus- Heparin and amiodarone 5 (1.5%) ing Merrem® IV [meropenem] and Ketamine and a dose of ciprofloxacin was due. 5 (1.5%) HYDROmorphone I flushed out the line after the MS16835 Merrem was completed and hung the Pantoprazole and IV fluids 5 (1.5%) 0 1 2 3 4 5 6 7 8 9 10 * The details of the PA-PSRS event narratives NUMBER OF REPORTS in this article have been modified to preserve * Accounted for more reports of precipitate formation (n = 6 of 64, 9.4%) than other drug pairs. confidentiality. IV, intravenous. Vol. 13, No. 4—December 2016 Pennsylvania Patient Safety Advisory Page 139 ©2016 Pennsylvania Patient Safety Authority R E V I E W S & A N A LY S E S ciprofloxacin. Within five minutes include IV contrast and metFORMIN The patient was prescribed flucon- after leaving the room, patient rang (23.8%, n = 54), omeprazole and clopido- azole by an orthopedic surgeon after the call bell for assistance. Inside the grel (8.8%, n = 20), and simvastatin with consulting with infections disease. tubing I noticed large amounts of amiodarone (4.0%, n = 9). The patient is s/p [status post] renal whitish flakes. I disconnected and Analysts queried event descriptions to transplant and taking Prograf® [tacro- discarded the IV set and drew back determine whether alerts were mentioned limus]. The patient was admitted [with a syringe] to withdraw medica- or involved in the event and found that and Prograf was held till creatinine tions from his port. I noticed very few only 5.3% (n = 12 of 227) of the reports and tacrolimus levels returned within white flakes in what I drew back. I mentioned “alert,” “flag,” or “warning.” normal limits. then flushed the line with fluid and placed a new bag of ciprofloxacin on Following are examples of reported drug Therapeutic Duplications a new set of iv tubing and am com- interactions that reached patients: Nearly a quarter (22.2%, n = 181 of 815) pleting the infusion. Physicians are to A patient taking metFORMIN was of reports were identified as involving be notified, as well as Pharmacy. admitted for a cardiac catheterization therapeutic duplications. Therapeutic 80-year-old female patient had with contrast. MetFORMIN was not duplication events reached patients in DOBUTamine infusion and Zosyn® addressed on admission or discharge 65.2% (n = 118 of 181) of the reports. [piperacillin sodium and tazobactam and was not ordered as inpatient. Only 2.2% (n = 4) of the events were sodium] piggybacked via y site. The Discharged the next day. Instructions reported as Serious Events (harm score combined medications crystalized given to patient had no information E and F; see Figure 2). and occluded the PICC [peripherally for the patient on whether or when Elderly patients were involved in most inserted central catheter] line which to resume [the metFORMIN]. The (60.2%, n = 109) of these events (see had to be removed and replaced with patient presented [approximately two Figure 3). left internal jugular central line. weeks later] with myopathy, renal failure, and serum creatinine of 2.1. Therapeutic duplications occurred in 36 (Normal results are 0.7 to 1.3 mg/ unique patient care areas with pharmacy Drug-Drug Interactions dL for men and 0.6 to 1.1 mg/dL for (27.1%, n = 49 of 181) cited most often. DDIs accounted for 27.9% (n = 227 of Medical/surgical and telemetry units were women.) The emergency department 815) of reported events. Although more involved in 13.3% (n = 24) and 11.6% physician did not address metFOR- than half (51.5%, n= 117 of 227) of DDI (n = 21) of the events, respectively. MIN. The patient was hydrated and events reached the patient, only 1.3% discharged. Discharge instructions for Overall, 78 unique medications were men- (n = 3) were reported as Serious Events patient instructed patient to continue tioned. The most common medications (harm score E and F; see Figure 2). metFORMIN. were anticoagulants, a class of high-alert Most of these events involved an elderly medications,5 including heparin (47%, n A physician identified a Tapazole® patient (50.7%, n = 115) while 43.2% (n = = 85 of 181), enoxaparin (34.8%, n = 63), [methimazole] and Synthroid® 98) involved adult patients (see Figure 3). rivaroxaban (13.3%, n = 24) and dabiga- [levothyroxine] drug interaction. The Forty-eight unique patient care areas were admission orders were processed the tran (11%, n = 20). The most common associated with a DDI event, with phar- previous night which included both pairs of medications were combinations of macy cited most often (31.7%, n = 72) Synthroid and Tapazole. The interac- anticoagulants (Figure 5). Overall, antico- followed by medical/surgical units (11%, tion did not flag [in the order entry agulants were mentioned 215 times (more n = 25) and emergency departments system] and was missed. Pharmacy than one anticoagulant was mentioned in (4.8%, n = 11). investigated and found that the some reports) in 181 reports. There were 116 unique medications and computer master inventory entry for Following are examples of reported thera- more than 112 unique pairs of medica- Synthroid was missing [appropriate] peutic duplications that reached patients: tions cited in DDI reports. The most codes. Thus, Synthroid was not flag- ging for therapeutic duplications, The patient received 3 doses of common medications mentioned in Lovenox® [enoxaparin] and 2 doses reports included IV contrast (e.g., iohexol, drug interactions, etc. The physician discontinued Tapazole. Pharmacy of heparin in a 24-hour period. A iopamidol; 23.8%, n = 54), metFORMIN physician ordered Lovenox and did (23.8%, n = 54), clopidogrel (9.3%, updated the entry for Synthroid to reflect the proper coding. not discontinue heparin patient was n = 21), and simvastatin (9.3%, n = 21). already receiving. Pharmacy did not The most common pairs of medications Page 140 Pennsylvania Patient Safety Advisory Vol. 13, No. 4—December 2016 ©2016 Pennsylvania Patient Safety Authority Figure 5. Most Common Pairs* Involved in Drug-Drug Interaction Reports Identified Most (56.1%, n = 37 of 66) of these events as Therapeutic Duplications, as Reported to the Pennsylvania Patient Safety involved an adult patient, but 42.4% (n = Authority, April 2009 through March 2016 (n = 181) 28) of events involved elderly patients (see Figure 3). DRUG PAIR Thirty-three unique patient care areas Heparin and enoxaparin 40 (22.1%) were associated with drug contraindica- tion events with medical/surgical units Heparin and dabigatran 12 (6.6%) (13.6%, n = 9), pharmacy (10.6%, n = 7), and telemetry (10.6%, n = 7) the top cited Heparin and rivaroxaban 12 (6.6%) care areas. Enoxaparin and rivaroxaban 10 (5.5%) Fifty-six unique medications and 21 unique pairs of medications were cited in MS16836 Heparin and apixaban 8 (4.4%) reports. The most common medications mentioned in reports included nitroglyc- 0 5 10 15 20 25 30 35 40 45 50 erin (12.1%, n = 8 of 66) and enoxaparin (9.1%, n = 6). The most common pairs of NUMBER OF REPORTS medications included nitroglycerin and * These drug pairs consist of anticoagulants, a class of high-alert medications. sildenafil (10.6%, n = 7), methadone with nalbuphine (7.6%, n = 5), and dofetilide with sulfamethoxazole/trimethoprim note warning when Lovenox was catch the therapeutic duplication. (4.5%, n = 3). entered into the pharmacy computer. Both medications were sent and both There was a variety of contraindications Nurses did not clarify order on were administered. The patients to drug therapy, with the most common when to give Lovenox and gave both experienced a rise in potassium to involving drug-drug contraindications Lovenox and heparin at the same greater than 5.0, which sent an alert (51.5%, n = 34). Contraindications due to time. Patient sent to acute care for to pharmacy concerning the ACE allergies were involved in 15.2% (n = 10 possible GI [gastrointestinal] bleed. [angiotensin-converting enzyme] of 66) and contraindications to a thera- Patient presented in hypertensive inhibitors. The pharmacist called the peutic intervention (e.g., administration emergency and was started on IV physician and received an order to dis- of an anticoagulant with an epidural line furosemide, IV nitroglycerin, and IV continue the captopril and continue in place) were involved in 12.1% (n = 8) nicardipine. The next day, the patient the lisinopril. of the reported events (see Table). was being transitioned back to his home oral medication regimen. The Drug Contraindications Following are examples of reported drug Drugs may be contraindicated when the contraindications that reached patients: patient was given some of his home medications while the nicardipine benefit of the combination of a drug Patient who was pregnant and on was still running and he became and another drug, comorbid condition, methadone for opioid addiction severely hypotensive requiring cardio- or procedure does not outweigh the risk was admitted to rule out sepsis. vascular support. He was discharged (e.g., aspirin is relatively contraindicated Nalbuphine ordered by resident, order home a week later with no further for children with viral infections because processed, filled and administered complications. it increases the risk of Reye’s syndrome). to patient. Directly after the dose Drug contraindications accounted for was administered, the patient began Upon admission to the hospital the 8.1% (n = 66 of 815) of events. Reports to exhibit signs of withdrawal (i.e. home medication list was obtained involving drug contraindications were tachycardia, pain, n/v [nausea and and verified with the patient. The categorized by harm score with 68.2% vomiting], cramps). The fetus was patient stated that he was taking (n = 45 of 66) of the events reaching the also tachycardic. Patient was trans- captopril 50 mg TID [three times patient (harm score C through I). Only ferred. Morphine was administered a day] and lisinopril 5 mg daily. 3% (n = 2) were reported as Serious Events and symptoms abated. Length of stay The medication reconciliation list with patient harm (harm score E and F; increased by one day for monitoring. was printed and both orders were see Figure 2). profiled. The pharmacist did not Vol. 13, No. 4—December 2016 Pennsylvania Patient Safety Advisory Page 141 ©2016 Pennsylvania Patient Safety Authority R E V I E W S & A N A LY S E S Table 1. Most Common Drug Contraindications, as Reported to the Pennsylvania Patient chemical stability data published for Safety Authority, April 2009 through March 2016 (n = 66) commonly used continuously infused EVENT TYPE NO. OF REPORTS % medications in ICU.10 The authors found 93 studies; 86 (92%) studies evaluated Drug-drug 34 51.5 physical compatibility and 35 (38%) stud- Allergy 10 15.2 ies evaluated chemical compatibility of at Therapeutic intervention 8 12.1 least one drug combination of interest. Physical and/or chemical compatibility Laboratory values 7 10.6 data existed for only 441 (54%) of the Diagnosis 5 7.6 possible 820 two-drug combinations, Test 1 1.5 whereas chemical compatibility data Procedure 1 1.5 existed for only 75 (9%) of the possible combinations. Of the 441 combinations for which compatibility data were avail- I verified an order to discontinue iso- pediatric intensive care ward showing that able, 67 (15%) represented incompatible sorbide mononitrate extended release 3.4 % of drug combinations were incom- combinations and 39 (9%) had conflict- 30 mg QAM [every morning]. Upon patible and thus potentially dangerous.6 ing information, with both compatible further profile review, I noticed the Tissot et al. found that 26% of incompat- and incompatible data identified. The patient was also on sildenafil 20 mg ibilities in an ICU were life threatening.7 authors concluded that physical compat- PO [by mouth] TID. Isosorbide Another study collected 78 different ibility studies are lacking for commonly mononitrate and sildenafil were veri- medication regimens and found 15% with used medications in ICU patients and fied three days earlier. It is unclear incompatibility reactions.8 may contribute to unsafe medication prac- whether the pharmacist verifying silde- tices. This is important because patients It is important to understand that there who are critically ill may require multiple nafil spoke with physician about the are many instances when two or more contraindication between [sildenafil IV medications administered by continu- medications have to be given concur- ous infusion. Obtaining separate venous and] isosorbide mononitrate. rently, but that does not mean that they access sites for each drug infusion would Patient was seen in urology clinic are compatible with each other. For be ideal, but in actuality, the number of and started on ciprofloxacin and example, the Institute for Safe Medication drug infusions often surpasses the avail- Bactrim™ [sulfamethoxazole and Practices (ISMP) reported that a phar- able access sites. trimethoprim]. Unfortunately, macist who was asked whether reteplase Bactrim is contraindicated with the injection could be infused with heparin While many new drugs are approved by patient’s Tikosyn® [dofetilide] therapy. consulted the product package insert, the U.S. Food and Drug Administration Patient was doing okay on therapy which stated that heparin frequently has (FDA) each year, research involving but was admitted for arrhythmia been given concomitantly with reteplase.9 incompatibilities has decreased. In 1991 evaluation which did not reveal any- Thus, the patient received both drugs and 1992, there were 245 newly pub- thing. Bactrim therapy stopped after through the same IV line. Unfortunately, lished clinical pharmaceutics research consultation with urology. Patient the pharmacist missed a sentence that articles incorporated into the seventh reeducated on drug interactions. appeared four lines above the informa- edition of Trissel’s Handbook on Injectable Unfortunately, patient used two phar- tion he had read that indicated heparin Drugs.11 Most of the studies came from macies and the insurance company and reteplase are incompatible when U.S. researchers in academia, pharmacy did not alert the outpatient pharmacy. combined in solution and should not be practice, and pharmacy students perform- administered together through the same ing laboratory-based research. Twenty DISCUSSION IV line. (Together, the drugs react to form years later, Trissel noted that the 17th a mass of solid or semi-solid material, edition of the Handbook on Injectable Drugs Information regarding the mechanisms incorporated fewer than 45 new research which can stop the infusion. Heparin and of drug incompatibilities is extensive, articles, most from foreign researchers. In reteplase can be given simultaneously but but quantitative information on the fre- other words, over a twenty-year span, new never mixed within the same container.) quency of occurrence and significance in research studies of drug compatibility and a clinical setting is limited. In one study, Kanji et al. performed a systematic review stability have declined more than 80%. incompatibilities were investigated in a to qualify and quantify the physical and Page 142 Pennsylvania Patient Safety Advisory Vol. 13, No. 4—December 2016 ©2016 Pennsylvania Patient Safety Authority The use of electronic order entry systems identified only 66% and 68% of the available. Information regarding the adop- with clinical decision support are among potential interactions.18,19 Weideman et al. tion and use of computerized prescriber the most promising strategies for detect- noted that none of the pharmacists were order entry systems and clinical decision ing and possibly preventing medication able to detect all potential interactions in support by the reporting facilities was also errors, including drug-drug interactions. a profile containing 8 or 16 drugs.19 This unavailable. However, they have not yet been tested reinforces the notion that a variety of risk with respect to preventing incompat- reduction strategies targeting system-based RISK REDUCTION STRATEGIES ibilities.12,13 To render a meaningful alert, causes of error, rather than relying solely The occurrence and perpetuation of drug electronic decision support systems would on human performance, is required to interactions involve many stages in the require, in addition to the drug name, intercept drug interaction and other medi- medication use process. This starts with dose, and so on, input on the number of cation errors. identifying possible interacting drugs available IV lines and the drugs currently The most common pair of medications when obtaining a medication list during being delivered into a given lumen.14 This involved in DDIs reported to the Authority the medication reconciliation process type of information is not routinely avail- was IV contrast and metFORMIN. The upon patients’ admission to a facility. It able in ICU patient records, much less most significant adverse effect of met- includes reviewing the medication profile, in a structured format that would enable FORMIN therapy is the potential for the prescribing medications, and pharmacy electronic screening. development of metFORMIN-associated review of ordered medications. The DDIs that reach the patient can largely lactic acidosis, particularly in susceptible medication use process also involves com- be considered preventable. One study patients. Because metFORMIN is excreted municating to resolve clinically significant reports that 9% of medication-related by the kidneys, any patient with existing drug combinations and communicating errors are likely due to DDIs.15 In a study renal insufficiencies are more prone to relevant interactions to prescribers or that assessed the prevalence of 25 clini- these effects. Iodinated contrast agents are nursing (e.g., to possibly alter the timing cally important DDIs in the ambulatory also eliminated by the kidneys, thus the of administration of one of the offending care clinics of the Department of Veterans combination of both products could be a drugs). It extends to monitoring patients Affairs, the authors found an overall rate of concern. Originally, when metFORMIN was for the possible adverse effects. Finally, 2.15% for potential DDIs.16 Case exposure introduced to the market, the risk of acute it requires educating patients upon rates were greatest for patients receiving kidney injury and metFORMIN-associated discharge. selective serotonin reuptake inhibitors lactic acidosis led to recommendation for Efforts to prevent harm from these types (SSRIs) and monoamine oxidase inhibi- facilities to establish a process to “hold” of drug interactions can be focused on tors (MAOIs), ganciclovir and zidovudine, metFORMIN before or after IV contrast either reducing the occurrence of poten- anticoagulants and thyroid hormones, and was administered to all patients. However, tial interactions before they happen or warfarin and nonsteroidal anti-inflamma- both FDA and the American College of mitigating the risk of adverse outcomes tory drugs. With new drugs coming on the Radiology (ACR) have updated recommen- associated with interactions that reach the market each year, the potential for DDIs to dations that restrict the need to discontinue patient. The following strategies may be take place will only increase. In fact, more metFORMIN to only certain patient useful to healthcare facilities seeking to than a decade ago, Hansten noted that populations20,21 (see “Concomitant Use of reduce drug interaction events. more than 15,000 articles related to DDIs metFORMIN and IV Iodinated Contrast”). had been published.17 Drug Incompatibilities Much has been written about the use of Limitations —— Ensure drug information resources an alerting mechanism to make prescrib- In-depth analysis by the Authority of are available and up to date for ers and pharmacists aware of the potential Serious Events resulting from medica- prescribers, pharmacy, and nurs- for a DDI, therapeutic duplication, or tion prescribing errors is limited by the ing staff to assess for potential contraindication while also acknowledg- information reported through PA-PSRS, incompatibilities. ing concerns for alert fatigue. Although including the event descriptions. As with all reporting systems, the type and —— When determining the compatibility alerts may be overused, relying on practi- number of reports collected depend on between two drugs, it is important tioner diligence to catch drug interactions the degree to which facility reporting is to also evaluate the type of tubing can be unreliable, especially when there accurate and complete. Information about being used for IV administration are many drugs approved by the FDA each underlying patient conditions, which may or whether the drugs might be year. Studies have shown that pharmacists have impacted events, was not consistently combined into one syringe. For and soon-to-graduate pharmacy students Vol. 13, No. 4—December 2016 Pennsylvania Patient Safety Advisory Page 143 ©2016 Pennsylvania Patient Safety Authority R E V I E W S & A N A LY S E S example, when a drug is incom- before establishing a new drug potential for unfavorable outcomes patible with other intravenous administration plan at the ICU. In associated with their combined use. medications, tubing used for admin- that study, the visual compatibility of —— Explore the possibility of adjusting istration sets should not have a Y 115 clinically reasonable intragroup the dose of the object drug (i.e., the connector for flushing the line or drug mixtures was investigated. All drug that is altered by the interac- piggyback drug administration.22 drug combinations were compat- tion) to decrease the risks from the —— If incompatible drugs must be ible for six hours except mixtures drug interaction if concomitant use given sequentially through the same containing thiopental, which was of the drugs is necessary.25 line, flush the line adequately with reassigned to a single-line use. —— Space dosing times of drugs to avoid saline or other compatible solution —— The use of in-line filters can reduce an interaction.25 For example, some between the drugs.9 the risk of precipitates or particles, drug interactions involve drugs bind- —— Standardize the concentration and which result from incompatibilities, ing in the GI tract. These types of diluent of continuously administered entering the body. As a consequence, interactions can be avoided if the IV drugs. In an attempt to minimize the filter may become blocked if interacting drug is administered at the risk of incompatibilities in an precipitation occurs. A blocked filter least two hours before or four hours anesthesia ICU, the authors of one should signal the need to investi- after the other drug. This allows study built upon their drug stan- gate the situation and check the the interacting drug to be absorbed dardization process and grouped medications ordered to eliminate any before the second drug is introduced. the drugs according to pH, medical incompatibility.24 —— Refine and improve drug-interaction indication, and chemical structure.23 alerts. When practitioners become The ICU staff decided to use multi- Drug-Drug Interaction, accustomed to clinically unimport- lumen central venous catheters, and Therapeutic Duplication, and ant or irrelevant warnings, they each group of drugs was assigned Contraindication often ignore or bypass these “false to one lumen. Only drugs that —— Avoid the combination of interacting alarms.”26 There are strategies that belonged to the same group were medications when possible. How- can help optimize the effectiveness infused simultaneously through the ever, some combinations of drug may of alerts. same lumen; therefore, intragroup be clinically necessary even with the incompatibilities were excluded CONCOMITANT USE OF METFORMIN AND IV IODINATED CONTRAST In the past, guidelines and drug labelling called for doses of metFORMIN to be held before patients receive any type of iodinated contrast media. The reason cited for this was an increased risk of acute kidney injury (AKI) and lactic acidosis. However, no cases of lactic acidosis have been reported after intravenous (IV) iodinated contrast medium in patients without a contraindication to metFOR- MIN therapy (e.g., patients with normal renal function). Upon review of recent studies and current evidence, the U.S. Food and Drug Administration (FDA) and the American College of Radiology (ACR) have updated guidelines for concurrent use of metFORMIN and iodinated contrast media. In April 2016, FDA revised the labelling requirements for metFORMIN products to include recommenda- tions to discontinue use of metFORMIN prior to administering IV iodinated contrast to patients with estimated glomerular filtration rate (eGFR) between 30 and 60 mL/min/1.73m2 or with a history of liver disease, alcoholism, or heart failure.1 (Normal levels for eGFR are from 90 to 120 mL/min/1.73 m2.) Then, in May 2016, ACR updated its guidelines, stating there is no need to discontinue metFORMIN prior to IV iodinated contrast in patients with no signs of AKI with eGFR greater than 30 mL/min/1.73m2.2 It is still rec- ommended to discontinue metFORMIN if a patient is to receive intra-arterial iodinated contrast.1,2 Please see both the FDA and ACR updates for complete guidance and recommendations. Notes 1. U.S. Food and Drug Administration. FDA drug safety announcement: FDA revises warnings regarding use of the diabetes medicine metformin in certain patients with reduced kidney function [online]. 2016 Apr 8 [cited 2016 Aug 4]. http://www.fda.gov/drugs/drugsafety/ucm493244.htm 2. American College of Radiology. Metformin [online]. Chapter 10. In: ACR Manual on Contrast Media. Version 10.2. 2016 [cited 2016 Aug 4]. http://www.acr.org/quality-safety/resources/contrast-manual Page 144 Pennsylvania Patient Safety Advisory Vol. 13, No. 4—December 2016 ©2016 Pennsylvania Patient Safety Authority * Use a tiered system for interac- * Many systems can provide will be able make appropriate dos- tive warnings to allow staff to reports about all warnings that age changes or even discontinue the view and easily bypass less serious have been overridden.26 Assign drug(s) if necessary. issues if appropriate, but require a clinician or manager to review —— Use the Computerized Prescriber Order staff to make a text entry to the report daily to identify any Entry (CPOE) System Evaluation describe the rationale for overrid- problems. Consider focusing on Toolkit, available at http://patient- ing significant alerts.26 To further a small number of common but safetyauthority.org/educationaltools/ enhance the effectiveness of this critically important warnings to patientsafetytools/prescribing/ tiered system, work to reduce monitor the effectiveness of the Pages/home.aspx, to test the CPOE the frequency of warnings that computer’s alert system. system to see if potentially harmful are not clinically significant to * Work with the drug informa- drug-drug interactions or drug con- users.27 Engage frontline staff tion vendor to build or modify traindications are detected. who repeatedly encounter clinical the severity of alerts necessary —— Notify prescribers and pharmacists warnings in this effort, because to warn practitioners about of any changes (e.g., types of alerts they can help identify alerts that possible serious or fatal adverse not available or turned off) made to are not clinically significant. events, especially those for cer- the alerting system in the order entry * Once insignificant warnings tain conditions (e.g., drugs that systems.27 have been reduced, organiza- prolong the cardiac QT inter- —— Although technology can help tions may want to display the val).27 Keep in mind that the improve patient outcomes, educate highest-level alerts (e.g., contra- time to build custom alerts var- and then remind staff to avoid total indications, severe DDI) and ies depending on the technology reliance on any technology involved lower-level alerts (e.g., warnings, in use and expertise of staff. in the medication-use process and precautions) for pharmacists, but * Establish a system to gather and that it should be one part of a well- display only the highest level for document all comorbid condi- integrated set of safety strategies.27 prescribers.27 tions in a structured diagnosis —— Establish a process to educate staff * Create and regularly update or problem list field in the elec- on new and potentially serious drug a list of significant alerts that tronic health record (EHR).27 interactions identified in the litera- require direct prescriber notifica- Link this information to the ture or by FDA. tion. The use of such a list can prescriber and pharmacy order help guide appropriate com- entry systems to promote clinical CONCLUSION munication of and response to a screening when new drugs are significant alert.26 prescribed, to detect potential Review of events submitted to the contraindications to those drugs. Authority shows that the medication * Ask prescribers and pharma- use processes associated with the occur- cists who enter orders to note —— Monitor patients for early detection rence of drug interactions needs to be warnings that they feel are not of possible interactions and adverse assessed and improved. Interactions may clinically significant.26 Then, effects.25 When it is necessary to pose a significant risk to patients’ health. evaluate the severity level of these administer a pair of drugs that inter- Effective identification and preventive less significant warnings and act with one another, the interaction strategies need to include all stages of the adjust as necessary to minimize might be managed through close medication use process to prevent harm potential for overlooking more laboratory or clinical monitoring. to patients from the administration of clinically significant warnings. If evidence that an interaction is multiple medications. occurring, healthcare practitioners NOTES 1. Johnson JA, Bootman JL. Drug-related 3. Kanji S, Lam J, Goddard RD, et al. 4. National Coordinating Council for Medi- morbidity and mortality. A cost-of- Cross-sectional observational study of cation Error Reporting and Prevention. illness model. Arch Intern Med 1995 Oct continuously infused medication admin- NCC MERP index for categorizing medi- 9;155(18):1949-56. istration practices and venous access cation errors [online]. 2001 Feb [cited 2. Thomas A, Routledge PA. Drug devices in Canadian adult ICU patients. 2016 Aug 4]. http://www.nccmerp.org/ interactions in clinical practice. Focus Crit Care Med 2008;36(Suppl):A94. types-medication-errors Pharmacovigilance Bull 2003:1-7. Vol. 13, No. 4—December 2016 Pennsylvania Patient Safety Advisory Page 145 ©2016 Pennsylvania Patient Safety Authority R E V I E W S & A N A LY S E S 5. Institute for Safe Medication Practices. 13. Bates, DW, Teich JM, Lee J, et al. The 21. American College of Radiology. Met- ISMP list of high-alert medications in impact of computerized physician formin [online]. Chapter 10. In: ACR acute care settings [online]. 2014 [cited order entry on medication error pre- Manual on Contrast Media. Version 10.2. 2015 Aug 4]. http://www.ismp.org/ vention. J Am Med Inform Assoc 1999 2016 [cited 2016 Aug 4]. http://www. Tools/institutionalhighAlert.asp Jul-Aug;6(4):313-21. acr.org/quality-safety/resources/ 6. Gikic M, Di Paolo ER, Pannatier A, 14. Bertsche T, Mayer Y, Stahl R, et al. contrast-manual et al. Evaluation of physicochemical Prevention of intravenous drug incompat- 22. Institute for Safe Medication Practices. incompatibilities during parenteral drug ibilities in an intensive care unit. Am J Don’t use epidural tubing for an IV solu- administration in a paediatric intensive Health-Syst Pharm 2008 Oct 1;65(19): tion. ISMP Med Saf Alert Acute Care 2008 care unit. Pharm World Sci 2000 Jun;22(3): 1834-40. Jan 17;13(1):3. 88-91. 15. Leape LL, Bates DW, Cullen DJ et al. 23. Nemec K, Kopelent-Frank H, Greif R. 7. Tissot E, Cornette C, Demoly P, et al. Systems analysis of adverse drug events. Standardization of infusion solutions to Medication errors at the administration ADE prevention study group. JAMA 1995 reduce the risk of incompatibilities. Am J stage in an intensive care unit. Intensive Jul 5;274(1):35-43. Health Syst Pharm 2008 Sep 1;65(17): Care Med 1999 Apr;25(4):3539. 16. Mahmood M, Malone DC, Skrepnek 1648-54 8. Vogel Kahmann I, Bürki R, Denzler U, GH, et al. Potential drug-drug interac- 24. Ball PA. Intravenous in-line filters: filter- et al. Incompatibility reactions in the tions within Veterans Affairs medical ing the evidence. Curr Opin Clin Nutr intensive care unit. Five years after the centers. Am J Health Syst Pharm 2007 Jul Metab Care 2003 May;6(3):319-25. Also implementation of a simple “colour code 15;64(14):1500-5. available at https://www.researchgate.net/ system.” Anaesthesist 2003 May;52(5): 17. Hansten PD. Drug interaction publication/10809860_Intravenous_ 409-12. management. Pharm World Sci 2003 in-line_filters_Filtering_the_evidence 9. Institute for Safe Medication Practices. June;25(3):94-7. 25. Ansari J. Drug interaction and Safety Brief. ISMP Med Saf Alert Acute 18. Cavuto NJ, Woosley RL, Sale M. Pharma- pharmacist. J Young Pharm 2010 Jul- Care 2000 Feb 23;5(4):1. cies and prevention of potentially fatal Aug;2(3);326-31. 10. Kanji S, Lam J, Johanson C., et al. drug interactions. JAMA 1996 Apr 10; 26. Institute for Safe Medication Practices. Systematic review of physical and 275(14):1086-7. Optimizing the use of computer system chemical compatibility of commonly used 19. Weideman R, Bernstein I, McKinney W. clinical alerts. ISMP Med Saf Alert Acute medications administered by continuous Pharmacist recognition of potential drug Care 2000 Jan 26;5(2):1. infusion in intensive care units. Crit Care interactions. Am J Health-Syst Pharm 1999 27. Institute for Safe Medication Practices. Med 2010 Sep;38(9):1890-8. Aug 1;56(15):1524-9. The absence of a drug-disease interaction 11. Institute for Safe Medication Practices. 20. U.S. Food and Drug Administration. alert leads to a child’s death. ISMP Med Cheers honoree’s acceptance speech cre- FDA drug safety announcement: FDA Saf Alert Acute Care 2015 May 21;20(10):1- ates pause for reflection. ISMP Med Saf revises warnings regarding use of the 4. Also available at http://www.ismp. Alert Acute Care 2012 May 3;17(9):1-3. diabetes medicine metformin in certain org/Newsletters/acutecare/showarticle. 12. Bates DW, Leape LL, Cullen DJ, et al. patients with reduced kidney function aspx?id=109 Effect of computerized physician order [online]. 2016 Apr 8 [cited 2016 Aug 4]. entry and a team intervention on preven- http://www.fda.gov/drugs/drugsafety/ tion of serious medication errors. JAMA ucm493244.htm 1998 Oct 21;280(15):1311-6. Page 146 Pennsylvania Patient Safety Advisory Vol. 13, No. 4—December 2016 ©2016 Pennsylvania Patient Safety Authority LEARNING OBJECTIVES SELF-ASSESSMENT QUESTIONS —— Identify the most common types The following questions about this article may be useful for internal education and of drug interactions, as reported assessment. You may use the following examples or develop your own questions. to the Pennsylvania Patient Safety 1. Which of the following type of drug interaction was most frequently reported to Authority. the Authority as a Serious Event? —— Identify the most common drug a. Drug-drug interactions pairs involved in drug interactions, b. Drug incompatibilities as reported to the Authority. c. Therapeutic duplications —— Identify the mechanisms and poten- d. Drug contraindications tial outcomes for drug interactions. e. Drug-food interactions —— Assess risk reduction strategies that 2. Which of the following drug pairs involved in drug incompatibilities was the most can be implemented to help prevent commonly reported? drug interactions. a. Heparin and vancomycin b. Ciprofloxacin and IV fluids with electrolytes c. CefTRIAXone and lactated Ringer's solution d. Heparin and amiodarone e. Pantoprazole and IV fluids 3. Which of the following drug pairs involved in therapeutic duplications was the most commonly reported? a. Heparin and dabigatran b. Heparin and rivaroxaban c. Enoxaparin and rivaroxaban d. Heparin and enoxaparin e. Heparin and apixaban 4. Which of the following statements about drug interactions is FALSE? a. Drug interactions may occur inside (drug-drug interactions) or outside (drug incompatibilities) the body. b. Drug-drug interactions may result from pharmacokinetic interactions or from interactions at drug receptors. c. A drug incompatibility occurs when two or more injectable drugs are mixed and the stability or structure of the drugs is altered by physical or chemical reactions. d. The pharmacological effect of a drug interaction of one or both drugs may be increased or decreased, but rarely results in an adverse effect. e. Physical changes from drug incompatibilities may lead to precipitate formation that can cause catheter occlusion and embolism and can contribute to a range of adverse drug events (ADEs), from thrombophlebitis to multi-organ failure. 5. Which of the following statements about the use of metFORMIN and IV con- trast is FALSE? a. The most common pair of medications involved in drug-drug interactions reported to the Authority was IV contrast and metFORMIN. b. The most significant adverse effect of metFORMIN therapy is the potential for developing metFORMIN-associated lactic acidosis, particularly in suscep- tible patients. (continued on page 148) Vol. 13, No. 4—December 2016 Pennsylvania Patient Safety Advisory Page 147 ©2016 Pennsylvania Patient Safety Authority R E V I E W S & A N A LY S E S (continued from page 147) c. Because metFORMIN is excreted by the kidneys, patients with existing renal insufficiencies are more prone to adverse effects. d. According the U.S. Food and Drug Administration (FDA) and American College of Radiology (ACR), facilities should establish a process to “hold” met- FORMIN before or after IV contrast is administered to all patients. Questions 6 refers to the following case: A patient had an order for one liter of sodium chloride 0.9% to be given, as well as an order for oxaliplatin. The sodium chloride was run at a Y-site connection that was clos- est to the patient for the last hour of the infusion. After the infusion was completed, the patient was transferred to radiation therapy, where she developed rigors. The patient was treated with diphenhydrAMINE and famotidine for likely incompatibility reaction with the sodium chloride. 6. Which of the following risk reduction strategies would NOT help prevent this drug incompatibility? a. Ensure drug information resources are available and up to date for prescribers, pharmacy, and nursing staff to assess for potential incompatibilities. b. If incompatible drugs must be given sequentially through the same line, flush the line adequately with saline or other compatible solution after both drugs have been infused. c. When determining the compatibility between two drugs, evaluate the type of tubing being used for IV administration or whether the drugs might be com- bined into one syringe. d. Use in-line filters to reduce the risk of precipitates or particles, which result from incompatibilities, entering the body. Questions 7 refers to the following case: An order was received in the pharmacy for a patient for amiodarone 400 mg twice a day. This patient had previously received an order for fluconazole 100 mg daily and warfarin 5 mg daily by different physicians. The computer system generated two interaction warn- ings involving amiodarone, one for warfarin, and a second warning for fluconazole. The orders were allowed to become active, and the pharmacist did not enter any clinical inter- ventions indicating that the physician was questioned on either one. After discharge, the patient attempted to get prescriptions for amiodarone and fluconazole filled at a retail pharmacy and was notified of interaction. 7. Which of the following risk-reduction strategies would NOT help prevent this drug interaction? a. Avoid the combination of interacting medications when possible. b. Explore the possibility of adjusting the dose of the object drug (i.e., the drug that is altered by the interaction) to decrease the risks from the drug interac- tion if concomitant use of the drugs is necessary. c. Space dosing times of drugs to avoid an interaction to allow the interacting drug to be absorbed before the second drug is introduced. d. Use a system for interactive warnings to allow staff to view and easily bypass significant alerts. Page 148 Pennsylvania Patient Safety Advisory Vol. 13, No. 4—December 2016 ©2016 Pennsylvania Patient Safety Authority PENNSYLVANIA PATIENT SAFETY ADVISORY This article is reprinted from the Pennsylvania Patient Safety Advisory, Vol. 13, No. 4—December 2016. The Advisory is a publication of the Pennsylvania Patient Safety Authority, produced by ECRI Institute and ISMP under contract to the Authority. Copyright 2016 by the Pennsylvania Patient Safety Authority. This publication may be reprinted and distributed without restriction, provided it is printed or distributed in its entirety and without alteration. Individual articles may be reprinted in their entirety and without alteration provided the source is clearly attributed. This publication is disseminated via e-mail. To subscribe, go to http://visitor.constantcontact.com/ d.jsp?m=1103390819542&p=oi. To see other articles or issues of the Advisory, visit our website at http://www.patientsafetyauthority.org. Click on “Patient Safety Advisories” in the left-hand menu bar. THE PENNSYLVANIA PATIENT SAFETY AUTHORITY AND ITS CONTRACTORS The Pennsylvania Patient Safety Authority is an independent state agency created by Act 13 of 2002, the Medical Care Availability and Reduction of Error (Mcare) Act. Consistent with Act 13, ECRI Institute, as contractor for the Authority, is issuing this publication to advise medical facilities of immediate changes that can be instituted to reduce Serious Events and Incidents. For more information about the Pennsylvania Patient Safety Authority, see the Authority’s website at http://www.patientsafetyauthority.org. An Independent Agency of the Commonwealth of Pennsylvania ECRI Institute, a nonprofit organization, dedicates itself to bringing the discipline of applied scientific research in healthcare to uncover the best approaches to improving patient care. As pioneers in this science for nearly 50 years, ECRI Institute marries experience and indepen- dence with the objectivity of evidence-based research. More than 5,000 healthcare organizations worldwide rely on ECRI Institute’s expertise in patient safety improvement, risk and quality management, and healthcare processes, devices, procedures and drug technology. The Institute for Safe Medication Practices (ISMP) is an independent, nonprofit organization dedicated solely to medication error prevention and safe medication use. ISMP provides recommendations for the safe use of medications to the healthcare community including healthcare professionals, government agencies, accrediting organizations, and consumers. ISMP’s efforts are built on a nonpunitive approach and systems-based solutions. Scan this code with your mobile device’s QR reader to subscribe to receive the Advisory for free.