Premedication Use Prior to Infliximab Administration: A Cross-Sectional Analysis

Abstract

Background

Premedications are commonly given to inflammatory bowel disease (IBD) patients prior to intravenous infliximab administration. We aimed to (1) describe practice variability; and (2) determine clinician rationale for premedicating IBD patients prior to infliximab administration.

Methods

We developed a cross-sectional electronic survey after comprehensive literature review to assess practice variability and clinician rationale for premedication use prior to infliximab. An optional post-survey quiz assessed clinicians’ understanding of available literature. The survey was distributed through members-only NASPGHAN and Crohn’s and Colitis Foundation of America (CCFA) listservs and American Gastroenterological Association (AGA) and American College of Gastroenterology (ACG) web-based discussion boards.

Results

379 unique respondents with a 93.3% survey completion rate comprised 331 (87%) and 45 (12%) pediatric and adult gastroenterologists. Among numerous options for premedications, acetaminophen (66%) and diphenhydramine (64%) were most often given prior to each infliximab infusion. Only 20% did not routinely use premedications. There was heterogeneity of premedication use between gastroenterologists within the same clinical practice. Of 328 (87%) respondents who completed the knowledge assessment quiz, only 18% identified the association of diphenhydramine use with increased reaction.

Conclusion

There is high inter- and intra-practice variability for premedication use prior to infliximab administration. Clinician rationale for premedicating patients appears to be driven by individual preference or group practice habit. Improved knowledge of the evidence may assist in decreasing over-use of premedications, particularly diphenhydramine.

INTRODUCTION

Infliximab, a monoclonal antibody binding tumor necrosis factor (anti-TNF) alpha, is the most commonly used first-line biologic agent for treatment of inflammatory bowel disease (IBD), in both Crohn’s disease (CD) and ulcerative colitis (UC).^1,2,3 Administered as an intravenous therapy, induction and maintenance regimens include dosing intervals based on clinical trials and clinician expertise, with each infusion duration ranging from 1 to >2 hours.^4,5,6,7 Although there is general clinical practice consesus in dosing, administration frequency, and duration, standardizing the use of premedications prior to infliximab has not been adequately explored.

Infusion reactions associated with infliximab range from mild reactions, including fever and chills, dyspnea, pruritus or urticaria (occurring in approximately 10%), to severe reactions including anaphylaxis, convulsions and hypotension (less than 1%).⁸ Monoclonal antibody therapy can be complicated by the development of antibodies to the medication. Acute infusion reactions – but not delayed hypersensitivity reactions – are more likely to occur in the presence of antibodies to infliximab (ATIs).^9,10 Of note, infusion reaction risks increase with the development of ATIs, associated with lower drug trough levels.^11,12 Early studies suggested systemic steroid premedication could reduce ATI levels but not eliminate ATI formation and therefore did not prevent infusion reactions.^13,14 Evidence also supports a non-immunologic basis for infusion reactions,^5,7,15 and experiential data show that the likelihood of acute infusion reactions does increase following a drug hiatus.¹⁶

Given this background on reaction risk associated with infliximab, premedications are frequently and routinely used in clinical practice with a heterogeneous list of medication options, including anti-pyretics (acetaminophen), antihistamines (diphenhydramine, cetirizine) and corticosteroids (prednisone, hydrocortisone, and methylprednisone). A comprehensive review of the literature shows weak strength of evidence for all-case premedication use in patients receiving infliximab. In the largest prospective multicenter study of infliximab use in 1632 adult patients with rheumatologic conditions (25.5% with IBD) representing 24,852 infusions, reactions occurred in 12% of patients and 1.3% of all infusions with only 2 instances of anaphylaxis. Interestingly, the use of antihistamines such as diphenhydramine was associated with a significant increase in the incidence of infusion reactions (OR 1.58, p = 0.0007).¹⁵ Studies in adult and pediatric IBD patients^17,18 showed similar rates of infusion reactions (adults: 19.7% in 447 patients, 3.5% in 6,468 infusions; children: 16.5% in 243 patients and 3.6% in 1652 infusions). Acetaminophen has been associated with a decreased incidence when used alone, but the overall effect when administered with other premedications remains unclear.^14,15 Corticosteroids have shown no definitive difference in infusion reactions.^{4,15,16,17,18,19} The incidence of reactions in the setting of immunosuppressive therapy is unclear; some studies show lower incidence;^4,9,11,17 others show no difference.^15,16

We hypothesize that there exists a non-standardized approach to premedicating IBD patients receiving infliximab. The objectives of this study were to (1) describe the practice variability of premedication use; and (2) determine clinical rationale for premedication use among clinicians treating IBD patients.

METHODS

Designing the Provider Survey

The North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN) Clinical Care and Quality (CCQ) and IBD committees designed a cross-sectional provider survey of premedication use prior to infliximab among practicing clinicians treating IBD patients. Subcommittee members from the CCQ Committee (JP, GW, KP) and the IBD Committee (WE, JM, AG) performed a comprehensive review of the literature to assess the strength of evidence for premedication use prior to infliximab administration and developed survey and self-quiz content. An adult gastroenterologist (CS) with expertise in IBD and quality improvement contributed to the survey construction and reviewed content relevance for clinicians treating adult IBD patients.

The final survey included 13 content questions as guided by 5 prioritization items (Table 1). Survey content priorities were: provider and practice variables (e.g., practice setting and size, IBD patient volume by provider and practice, and years of practice); practice-specific infusion standards (e.g., presence of a standardized infliximab order set, duration of infusion, dosing protocols); practice agreement in infliximab ordering based on respondent observation (all-case premedication use by provider and practice); and justification for premedication use (i.e., all-case and specific indications). The survey was designed to take 3 minutes to complete.

Table 1.

Survey & Self-Quiz Content Emphasis

Prioritization list for the 13-question clinician survey

Provider and Practice Variables Practice-Specific Infusion Standards Practice Agreement in Infliximab Ordering All-Case Premedication Use (by provider and by practice) Justification for Premedication Use

Prioritization list for the 4-question self-quiz

Rate of Acute Reactions to Infliximab Predominance of Non-Immunologic Basis of Reactions Association of Diphenhydramine with Increased Reactions Association of Anti-TNF Antibodies with Increased Reactions

Premedication Options

Premedication choices included Hydrocortisone (Solu Cortef) IV, Methyl-prednisolone (Solumedrol) IV, Prednisone or prednisolone PO, Dexamethasone (Decadron) PO, Diphenhydramine (Benadryl), Cetirizine (Zyrtec), Loratadine (Claritin), Acetaminophen (Tylenol), Ondansetron (Zofran) and Other. The survey was designed to determine variation in all-case premedication use among providers within a particular practice and between providers across all practice settings. Respondents were specifically asked to indicate rationale for premedication use, both in all-case and for specific indications only. All-case justification choices included “Prevent infusion reaction”, “Prevent antibodies to infliximab”, “Practice standard of my group”, “Other” and “I don’t know”. Specific indication justification choices included “I would not use under any circumstance”, “History of prior infusion reaction”, “History of predisposition to drug allergy”, “If on infliximab monotherapy”, “Presence of antibodies to infliximab”, “If prolonged lapse between doses of infliximab”, “Home infusion”, and “Other”.

Developing the Self-Quiz

Minimizing the time to complete all components of the survey and quiz was a high priority for the subcommittee. In order to assess knowledge base among respondents, 4 prioritization items were deemed important to include in the optional self-quiz as they were relevant to inform evidence-based practice (Table 1). These 4 questions assessed knowledge of the rate of acute infusion reactions to infliximab (correct answer: 1–4%), the predominance of non-immunologic basis of infusion reactions, the association of diphenhydramine use with increased reaction, and the association of anti-TNF antibodies with increased reaction.

Distribution of the Qualtrics Electronic Survey

The survey and the study protocol received approval by the Stanford University Human Subjects Research Institutional Review Board. The survey and optional self-quiz questions were populated in Qualtrics web-based platform (Qualtrics, Provo, UT). To ensure that only one survey was captured per respondent, the Qualtrics software matched computer IP addresses with provided e-mails at the time of initiating the survey. Electronic distribution occurred to all emails on the members-only NASPGHAN and CCFA listservs. A separate discussion with web link to the Qualtrics survey was available on the American Gastroenterological Association (AGA Community Forum) and American College of Gastroenterology (ACG & CCFA IBD Circle) web-based discussion boards. The survey was secondarily disseminated by these members to the international community of pediatric gastroenterologists that was not necessarily encompassed by the listservs or discussion boards.

Analysis of Results

The Qualtrics link was open for survey completion from April 15 to June 30, 2016. Upon completion of the surveys, individual and aggregate data from Qualtrics were exported to Microsoft Excel (Microsoft Corporation, Redmond, WA) for data cleaning and descriptive statistics.

RESULTS

In total there were 379 unique respondents with a 93.3% survey completion rate collected from 331 (87.3%) and 45 (11.9%) board-eligible or certified pediatric and adult gastroenterologists, respectively (Table 2). The majority (67%) practiced in an academic setting and large portions were in private practice (15.3%) or a health-system practice (13.5%). The number of providers per practice and the volume of IBD patients by provider and practice varied. The majority of practitioners ordered infliximab from a standardized order set (64.9%) and infused over 2 or 3 hours (74.6%).

Table 2.

Provider and Practice Characteristics

	Total (n)	Pediatric (n)	Adult (n)
Training path
Pediatric	331
Adult	45
Other	3
Years in practice
<10	49.1% (186)	45.4% (159)	57.8% (26)
10–19	21.9% (83)	21.7% (76)	13.3% (6)
20–30	21.4% (81)	20.3% (71)	20% (9)
>30	7.1% (27)	6.6% (23)	8.9% (4)
Practice setting
Academic	67% (254)	63.4% (222)	68.9% (31)
Private Practice	15.3% (58)	14.3% (50)	15.6% (7)
Health-System Practice	13.5% (51)	12.3% (43)	15.6% (7)
Other	4.2% (16)	4.6% (16)	0% (0)
Practice size (# providers)
<5	35.4% (134)	36.0% (119)	28.9% (13)
6–10	38.3 % (145)	37.5% (124)	44.4% (20)
11–30	19.5% (74)	19.9% (66)	17.8% (8)
>30	6.9% (26)	6.6% (22)	8.9% (4)
Provider patient volume (# IBD patients)
<10	25.9% (98)	28.4% (94)	8.9% (4)
10–30	43.3% (164)	44.4% (147)	35.6% (16)
31–50	14% (53)	13.6% (45)	17.8% (8)
>50	16.9% (64)	13.6% (45)	37.8% (17)
Practice patient volume (# IBD patients)
<50	9.8% (37)	10.9% (36)	2.2% (1)
51–200	30.3% (115)	32.6% (108)	13.3% (6)
201–500	35.1% (133)	35.6% (118)	28.9% (13)
501–1000	16.6% (63)	16.6% (55)	17.8% (8)
>1000	8.2% (31)	4.2% (14)	37.8% (17)

Premedications were commonly used prior to every infliximab infusion (Figures 1 and 2). Regardless of practice type, the majority of respondents reported using an antihistamine (69.7%) or acetaminophen (66.2%) routinely and diphenhydramine was the second most commonly used premedication (63.6%). Many respondents used a steroid (48%) prior to every infliximab infusion with similar rates of methylprednisolone and hydrocortisone use. Despite common use of premedications, a sizeable proportion (19.8%) of the respondents did not use premedications routinely.

Figure 1.

Proportion of all-case premed use by type

Figure 2.

Proportion of all-case premed use by medication

In addition to variability of premedication use throughout all respondents, there was reported heterogeneity of premedication use between gastroenterologists within the same clinical practice. About a third (35.4%) of respondents indicated ordering was based on individual preference only. Of the respondents who reported a written protocol for premedication (44.5% of respondents), a majority (58.6%) had provider agreement always, although the rest agreed “often” or “sometimes”. Of those with no written protocol (19.8% of respondents), only 25.3% agreed “always”.

The majority of clinicians (61.2%) (Figure 3) reporting all-case premedication use justified this practice as prevention of an infusion reaction. A quarter justified all-case premedication use because it was the practice standard of their group (26.4%). Prevention of antibodies to infliximab was also a common justification (22.4%), and of these all but 2 respondents included a steroid.

Figure 3.

Rationale for all-case premedication use

The rationale for conditional premedication use prior to infliximab administration resembled the rationale for empiric use. History of an infusion reaction – including pruritus, dyspnea or chest discomfort, urticaria or rash, and anaphylaxis – was the most frequent indication for conditional premedication use (Figure 4). A prolonged interruption in infliximab administration (19.5%) and presence of antibodies to infliximab (11.3%) were also common justifications.

Figure 4.

Rationale for conditional premedication use

Most respondents (86.5%) opted to take the four-question quiz to test knowledge about evidence-based premedication use. Most (68%) recognized that the presence of antibodies to infliximab is associated with an increased risk of infusion reaction. The non-immunologic basis of most infusion reactions was also accurately indicated by the majority of respondents (64.9%). Notably, only 18.3% of respondents correctly identified the association of diphenhydramine use with increased infusion reaction. About half (51.2%) of respondents correctly estimated the rate of infusion reaction with infliximab to be 1–4% while many overestimated how often these reactions occur.

DISCUSSION

Through a collaborative effort between the CCQ and IBD Committees of NASPGHAN, we performed a cross-sectional analysis aimed to understand practice variability and clinician rationale for premedication use prior to infliximab administration in IBD patients. Our investigation represents the largest and most representative sample of clinician participants managing IBD across various practice types, including both adult and pediatric gastroenterologists. The findings presented corroborate a previous study,²⁰ which focused on the variability of infliximab administration practices among pediatric gastroenterologists.

Two clinical implications are relevant for further discussion based on the primary findings in our study of premedication use. First, regardless of practice setting and training type, the majority of clinicians treating IBD patients with infliximab empirically use premedications. Although there is paucity of evidence that routine use of premedications (regardless of the type of premedication) reduces the incidence of infusion reactions, a substantial proportion of the respondents use acetaminophen and diphenhydramine prior to every infliximab infusion. Many also use corticosteroids. Furthermore, if there is any protective effect in patients with prior infusion reaction, that effect has not been clearly described in the literature.

The clinical impact of indiscriminate use of premedications in IBD patients is not inconsequential. Premedications are associated with side effects, and many of these may be misconstrued as reactions to infliximab. Well-investigated diphenhydramine, in particular, is a medication with commonly associated adverse or unwanted side effects.²¹ They include drowsiness, dizziness, constipation, upset stomach, blurred vision, rash and dry mouth. Blunting of mental alertness for IBD patients after receiving diphenhydramine with each infliximab therapy may represent an unnecessary and even unsafe reaction. Curtailing diphenhydramine misuse and overuse have been described in Choosing Wisely campaigns in non-gastroenterology practices.^22,23 Gastroenterologists are taking steps to reconsider how to judiciously administer “benign” medications,²⁴ including acetaminophen, which can exacerbate hepatotoxicity in patients with existing liver injury despite appropriate dosing.²⁵

Second, we found high variability between practices and providers and within the same practice. As expected, most clinicians justified the use of premedication with the rationale to prevent infusion reactions. However, the results of the post-survey self-test suggest limited awareness of the insufficient evidence to support premedication use. Interestingly, none of the respondents reported never using premedications. Although the majority of respondents knew reactions were predominantly non-immunologically driven, most respondents conditionally used premedications prior to infliximab if there was a prior history of a reaction and after a prolonged time lapse between infusions.

Given the discordance in evidence-based knowledge and real-world practice habits, a major driver of premedication use seems to be dependent on familiarity of ordering routines. Most of the respondents indicated reliance on infliximab order sets, which may have premedications pre-populated or default-ordered for each infliximab administration within electronic health records. The effect of defaults in an electronic health record environment has been shown to significantly influence clinicians’ ordering habits.²⁶

We acknowledge the limited number of respondents from adult gastroenterology practices in this analysis. However, the aggregate results suggest that practice variation exists regardless of the patient cohorts’ age. Selection bias in survey studies is a limitation, but there were no added incentives (e.g., monetary compensation) to skew recruitment, introduce repeated responses, or amplify cognitive bias prior to beginning the survey.

In conclusion, the risk-benefit ratio of wide-spread premedication use is ultimately the clinical question raised in our investigation. Opportunities for more judicious use of premedications in IBD patients prior to receiving infliximab may represent an implementable quality improvement topic for provider- and health system-driven processes.