Downloaded from
`
`http://ndt.oxfordjournals.org/
`
` at Pennsylvania State University on March 4, 2014
`
`Nephrol Dial Transplant (2005) 20: 1438–1442
`doi:10.1093/ndt/gfh811
`Advance Access publication 19 April 2005
`
`Original Article
`
`Benchmarking iron dextran sensitivity: reactions requiring
`resuscitative medication in incident and prevalent patients
`
`Brian A. J. Walters1 and David B. Van Wyck2
`
`1Division of Nephrology and Hypertension, University of Miami School of Medicine, Miami, FL and
`2Department of Medicine, University of Arizona College of Medicine, Tucson, AZ, USA
`
`Abstract
`Background. Reliable information on the incidence
`of
`severe
`reactions
`to iron dextran is
`limited.
`Administration of agents of resuscitation in acute
`anaphylaxis may serve as a marker to quantify life-
`threatening adverse drug reactions.
`Methods. To determine the incidence of the most
`serious reactions to intravenous (i.v.) iron dextran, we
`searched the Gambro Healthcare US medical database
`for evidence of same-day administration of both i.v.
`iron dextran and parenteral adrenaline, corticosteroids
`or antihistamines. We confirmed each case as an iron
`dextran sensitivity reaction by direct inquiry. We also
`determined the total reported number of suspected
`adverse iron dextran reactions.
`Results. During the 16 month study period, we
`determined that 1 066 099 doses of i.v. iron dextran
`were given to 48 509 patients, including 20 213 patients
`who had not previously received iron dextran (iron
`dextran naı¨ ve). We identified seven patients who
`experienced reactions requiring resuscitative agents,
`all in response to a test dose (five patients) or first
`therapeutic dose (two patients), and therefore all in
`the iron-naı¨ ve (incident) group. Thus, we found the
`incidence of iron dextran reactions requiring resusci-
`tative agents to be 0.035% (7 out of 20 213). No
`reaction was fatal. In a combined group of incident
`and prevalent patients, we found 337 total reports of
`suspected adverse reactions to iron dextran, without
`regard to severity of reaction, yielding an overall per
`patient adverse drug event (ADE) rate of 0.69% (337
`out of 48 509) and per exposure rate of 0.03% (337 out
`of 1 066 099).
`to iron
`reactions
`incidence of
`Conclusions. The
`dextran requiring resuscitative medications, per expo-
`is 0.035%. Reactions of this
`sure or per patient,
`
`Correspondence and offprint requests to: David B. Van Wyck, MD,
`Kidney Health Institute, LLC, 6720 North Nanini Drive, Tucson,
`AZ 85704-6128, USA. Email: dvanwyck@sprynet.com
`
`severity occur after either the test dose or first dose
`of iron dextran.
`
`Keywords: adverse reactions; anaemia; anaphylaxis;
`chronic renal failure; iron dextran; iron deficiency
`
`Introduction
`
`iron
`for
`Agents administered intravenously (i.v.)
`deficiency consist of
`colloidal
`iron–carbohydrate
`compounds distinguished structurally by differences
`in core size and carbohydrate chemistry, and clinically
`by differences in pharmacokinetics, maximum dose
`size and maximum rate of infusion [1]. That members
`of the iron–carbohydrate family are also distinguished
`by the rate of adverse reactions is frequently pro-
`posed but incompletely supported by the literature.
`No direct comparative studies have been performed
`among agents available in the USA and Europe,
`including iron dextran,
`ferric gluconate and iron
`sucrose. Recent prospective clinical trials have reported
`adverse drug events (ADEs) after ferric gluconate [2]
`and after iron sucrose [3]. Iron dextran reaction rates
`are frequently cited as a benchmark against which non-
`dextran iron agents are compared [2,4]. However,
`existing iron dextran benchmarks remain unreliable
`because available reports of iron dextran reaction
`rates lack crucial
`information required to calculate
`the true incidence and prevalence. Specifically, key
`studies fail to report the total number of iron dextran
`doses administered [5], the number of patients treated
`[6,7] or whether patients had been exposed to iron
`dextran previously [6–8].
`Adverse reactions to parenteral iron agents range
`from minor to life-threatening. Although the clini-
`cal features of reactions to iron agents have been
`reviewed extensively and listed in the literature [5,9,10],
`information on the incidence of the most severe reac-
`tions is limited, in part because reactions described as
`
`ß The Author [2005]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
`For Permissions, please email: journals.permissions@oupjournals.org
`
`

`

`Downloaded from
`
`http://ndt.oxfordjournals.org/
`
` at Pennsylvania State University on March 4, 2014
`
`Incidence of severe iron dextran reactions
`
`anaphylactic include manifestations that are relatively
`benign [5], the severity of reactions described as serious
`is difficult to assess objectively, and the study popula-
`tions upon which estimates are based have been
`relatively small. Moreover, because the pathogenesis
`of parenteral iron reactions is unknown [11], labora-
`tory markers to distinguish degrees of severity are
`absent. Thus, reliable incidence information on the
`most severe reactions is critically needed to determine
`the proper role of iron dextran in anaemia management
`and to evaluate and compare the safety of parenteral
`iron agents.
`Large medical databases provide powerful tools to
`elucidate iron dextran reactions in dialysis patients.
`Analysis of clinical variance reports for haemodialysis
`patients in the Fresenius Medical Care North America
`(FMCNA) database in the 6 month period October
`1998–March 1999 yielded 165 suspected ADEs among
`841 252 i.v. iron dextran administrations [6]. Among
`the 165 cases identified, 43 required emergency depart-
`ment evaluation, 11 were hospitalized and one died;
`113 reactions occurred after the first dose in a series
`(maintenance or a planned course of injections); and
`50% of patients with ADEs had received iron dextran
`safely in the past. However, because this study did
`not identify the total number of patients administered
`i.v. iron dextran, information on the incidence and
`prevalence of iron dextran sensitivity per patient at
`risk was not available, and direct comparison with
`results of previous reports was not possible.
`life-
`To determine the incidence of
`the most
`threatening reactions
`to iron dextran in dialysis
`patients, we examined a large medical database
`containing the pharmacological treatment history of
`>48 000 dialysis patients who received iron dextran
`during a 16 month period from January 1999 to April
`2000. Using medication entry fields in the database,
`we identified patients who received iron dextran and
`either adrenaline, corticosteroids or antihistamines
`parenterally during the same dialysis day. We reasoned
`that same-day administration of iron dextran and
`one or more of these i.v. resuscitative agents should
`serve as an objective marker for the most severe iron
`dextran reactions. We confirmed each identified epi-
`sode by direct clinical
`inquiry and compared the
`number of identified episodes with the total number
`of i.v. iron doses given during the study period.
`
`Patients and methods
`
`Study design
`
`searched the
`study. We
`retrospective
`a
`This was
`Õ
`Gambro Healthcare database (RIMS
`—Renal Information
`Management System, Gambro Healthcare, Denver, CO) for
`Õ
`each record of
`i.v.
`iron dextran [INFeD
`; Watson
`Pharmaceuticals, Corona, CA (available in Europe as
`Õ
`Cosmofer
`; Nebo A/S, Denmark
`and
`Vitaline
`Pharmaceuticals, Long Crendon, Buckinghamshire, UK) or
`Õ
`Dexferrum
`; American Regent, Inc., Shirley, NY (available
`
`1439
`
`Õ
`
`in Canada as DexIron
`; Genpharm, Inc, Etobicoke, Ontario,
`Canada] administration in the period between January 1,
`1999 and April 30, 2000. Among resulting records, we
`distinguished patients who had received iron dextran at any
`time prior to the study period from those who received iron
`dextran during the study period. Among patients in both
`groups, we then identified individuals who received i.v.,
`subcutaneous (s.c.) or intramuscular (i.m.) administration of
`adrenaline, corticosteroid or antihistamine agents during a
`dialysis day in which iron dextran was also administered. In
`each resulting patient, we contacted the patient-care team
`directly to confirm whether the identified resuscitative agents
`were administered for a suspected iron dextran reaction. We
`also determined whether the reaction occurred after the test
`dose, first therapeutic dose or subsequent therapeutic doses.
`To exclude the possibility that pre-treatment with either
`form of iron dextran would influence reactions to the other,
`we identified all records which included administration of
`both agents during the study period.
`To determine the prevalence of all reactions to iron,
`without regard to severity, we searched for the term ‘iron’
`or ‘iron dextran’ including trade names in the allergy field
`of the patient record. Entries in the allergy field, a component
`Õ
`of the patient medical history file in RIMS
`, are generated
`by allergic reactions experienced by the patient within a
`Gambro dialysis facility, suffered outside the facility or
`reported by the patient from previous medical history.
`To shed light on the incidence of all reactions to iron
`dextran without regard to degree of severity, we determined
`the total number of patients in whom adverse iron dextran
`reactions were reported for the first time during the study
`period. The search strategy specifically sought the word
`‘iron’ or ‘iron dextran’ including trade names in the adverse
`events field of each record. We divided the result by the
`total number of patients who received iron dextran for the
`first time during the study period. We made no attempt to
`validate each adverse reaction report.
`To assess the overall prevalence of iron dextran sensitivity
`in the study population, we searched the allergy field in every
`patient record, whether or not the patient received iron
`dextran during the study period, determined the number
`of records containing the word ‘iron’ and medications
`containing iron dextran in the allergy field, and divided the
`result by the total number of patients in the database.
`
`Medical database
`
`The Gambro Healthcare database contains the electronic
`medical record for every dialysis patient being treated at each
`out-patient Gambro facility in the USA and has several
`different levels of security to maintain patient confidentiality.
`The electronic medical record contains additional
`inter-
`active functions in a relational database format propagated
`through Informix 7.20 (Merlo Park, CA). The resulting
`Õ
`architecture forms the basis of
`the proprietary RIMS
`software, which links data elements from patient demo-
`graphics, medical history, clinical
`laboratory test results,
`medications, medical
`interventions and hospitalizations.
`Medication history includes the dose, time and date of
`medication administration, as well as a listing of prescribed
`ongoing medications. Allergies and adverse reactions are
`reported in distinct fields in separate components of the
`database: allergies as a component of the medical history,
`and adverse reactions to i.v. medications administered as
`
`

`

`Downloaded from
`
`http://ndt.oxfordjournals.org/
`
` at Pennsylvania State University on March 4, 2014
`
`1440
`
`a component of the medication history or event reporting
`management system. Thus, a record of iron dextran allergy
`may reflect
`information from distant medical history
`obtained by interviewing the patient, and may include infor-
`mation on iron dextran reactions experienced either inside
`the Gambro facility, or at previous facilities or hospitals.
`Adverse reaction reports, on the other hand, include infor-
`mation only on those reactions that were experienced
`within Gambro facilities. Since adverse reaction reporting
`is encouraged, no attempt is made to screen serious from
`minor reactions.
`
`Statistical analysis
`
`We examined the statistical significance of the form of iron
`dextran used and the presence or absence of angiotensin-
`converting enzyme (ACE) inhibitor therapy using Poisson
`regression modelling (SAS Institute, Cary, NC).
`
`Results
`
`The Gambro Healthcare database from January 1,
`1999 to April 30, 2000 contains information on a total
`of 61 950 dialysis patients. Our retrospective analysis
`yielded 1 066 099 episodes of iron dextran administra-
`tion in 48 509 patients, including 28 296 patients who
`had received iron dextran prior to the study period
`(prevalent patients) and 20 213 who had not previously
`received iron dextran (incident patients). We identified
`seven episodes, in seven patients, in which i.v. iron
`dextran administration coincided with i.v administra-
`tion of resuscitative agents. There were no deaths.
`All seven episodes were confirmed as iron dextran
`reactions upon direct clinical inquiry. Thus, the overall
`per exposure rate of adverse reaction requiring resus-
`citative i.v. medication was 7 out of 1 066 099
`(0.0007%). However, clinical inquiry determined that
`all seven reactions occurred after either the test dose
`(five patients) or first dose (two patients). Thus, among
`prevalent patients, there were no reactions requiring
`resuscitative medications. Among incident patients,
`where the meaningful exposures were effectively the
`test or first dose, the per exposure rate was equivalent
`to the per patient rate, or 7 out of 20 213 (0.035%).
`Õ
`(five episodes in 317 097
`Patients receiving INFeD
`exposures to 13 765 patients) were no more likely to
`experience life-threatening reactions than those receiv-
`Õ
`(two episodes in 123 309 exposures
`ing Dexferrum
`to 6448 patients; P ¼ 0.9733). Patients receiving ACE
`inhibitor therapy were more likely to suffer a life-
`Õ
`threatening reaction to iron dextran (either INFeD
`Õ
`or Dexferrum
`)
`than patients not receiving ACE
`inhibitors (P ¼ 0.0082).
`There were 337 reports of adverse reactions to
`iron dextran during the study period. Thus the inci-
`dence of all reported adverse iron reactions during
`the study period without regard to degree of severity
`of reaction or history of previous exposure was (337
`reports out of 48 509 patients) 0.6947%. The overall
`per exposure adverse reaction rate was therefore
`(337 events out of 1 066 099 exposures) 0.0316%.
`
`B. A. J. Walters and D. B. Van Wyck
`
`We separately analysed records in patients who
`Õ
`Õ
`and Dexferrum
`during the
`received both INFeD
`study period. Among these 2075 patients, we found
`no episodes of i.v. resuscitative medication adminis-
`tration in 76 474 iron dextran exposures.
`
`Discussion
`
`This report represents the second use of a large-scale
`medical database to elucidate iron dextran reactions,
`the first to determine the per patient incidence of
`adverse reactions to iron dextran, and the first to assess
`risk in incident compared with prevalent patients.
`The potential advantage of the current report resides
`in its scale, the ability to distinguish previously exposed
`from previously unexposed patients, and the use
`of an objective definition of serious iron dextran
`ADEs:
`same-day administration of
`iron dextran
`and either parenteral adrenaline, corticosteroids or
`antihistamines.
`Our findings suggest that the most severe reactions
`to iron dextran are seen in naı¨ ve patients, that success-
`ful administration of a test dose does not preclude
`a life-threatening reaction to a first therapeutic dose,
`and that successful administration of a first dose
`seems to render the risk of developing a life-threatening
`reaction to subsequent doses of either form of iron
`dextran substantially less likely.
`We used another feature of the database, the record
`of adverse reactions to medications, to determine the
`total number of iron dextran reactions serious enough
`to prompt withdrawal from further drug exposure.
`We found that adverse reactions serious enough to
`be reported to the database occurred in 0.7% of
`48 509 dialysis patients receiving iron dextran. Previous
`studies conducted on a smaller scale found serious
`reactions in 0.7% of 573 dialysis patients [5] and
`0.6% of 481 non-uraemic patients [12] examined
`retrospectively after iron dextran injection. The con-
`cordance between our findings and previously pub-
`lished results of per patient reaction rates suggests
`that reporting of serious adverse events to the Gambro
`database is relatively complete. Since the purpose
`of reporting to the Gambro database is to prevent
`further administration of iron dextran, we can con-
`sider the adverse event report rate to be equivalent to
`an iron dextran intolerance rate. Thus, in our mixed
`patient population (58% previously exposed, 42%
`previously unexposed), 0.7% of patients developed
`iron dextran intolerance over 16 months, yielding
`an intolerance incidence of 0.5% per year.
`We chose same-day administration of resuscitative
`agents as an objective marker for life-threatening
`reactions because broader definitions lack precision,
`consistency and reproducibility. Many adverse reac-
`tions to iron dextran,
`including rashes that satisfy
`published definitions of anaphylaxis, are not serious,
`and not all serious reactions meet criteria for ana-
`phylaxis [5]. Some but not all patients were with-
`drawn from further iron dextran administration after
`
`

`

`Downloaded from
`
`http://ndt.oxfordjournals.org/
`
` at Pennsylvania State University on March 4, 2014
`
`Incidence of severe iron dextran reactions
`
`reactions that were not serious. Some patients who
`have experienced reactions described as serious, ana-
`phylactic, anaphylactoid or allergic have not been
`withdrawn from further iron dextran therapy but
`have subsequently received multiple doses of i.v. iron
`dextran without adverse events [13]. Finally, some
`patients have experienced serious reactions to iron
`dextran despite successfully receiving a test dose and
`multiple previous therapeutic doses [5]. These observa-
`tions and the paucity of information to support a
`single pathogenesis for all iron dextran reactions [11]
`argue persuasively for a single, objective standard to
`quantify the most serious iron dextran reactions. Our
`finding that resuscitative agents are required after the
`test dose or first dose in 0.035% of patients provides
`the first information on such a standard, helps to assess
`the role of i.v. iron dextran in the management of
`iron deficiency and provides a benchmark against
`which other parenteral
`iron preparations can be
`compared.
`Our results support and extend previous findings
`arising from the use of a large medical database [6].
`Although design differences between the two studies
`make direct comparison difficult, both the current and
`previous study show that iron dextran ADEs after
`iron dextran administration can be life-threatening.
`The previous study included one fatal ADE. Though
`both studies showed that most ADEs follow adminis-
`tration of a test dose or first dose, the previous study
`identified serious ADEs in patients who had success-
`fully received previous test or treatment doses. The
`majority of those non-naı¨ ve patients who experienced
`ADEs after iron dextran administration did so at the
`time of the first dose of a planned series, suggesting
`that the risk of first-dose ADEs may recur in prev-
`alent patients after an interval free from iron dextran
`exposure. The results of our analysis of adverse
`reaction reports without regard to degree of severity
`or previous iron dextran exposure (337 reports out
`of 1 066 099 exposures, or 0.0316%) closely approxi-
`mate the results of the previous report (165 events
`in 841 252 exposures, rate 0.0196%) [6], which used
`similar ADE reports and a mixed population of
`incident and prevalent patients. Again, concordance
`between our results and those previously reported for
`per exposure reaction rates provide further evidence
`that reporting to large medical databases is relatively
`complete.
`Assessing the quality of published evidence on
`the safety of i.v.
`iron agents requires considerable
`caution. Few retrospective studies provide informa-
`tion on a history of previous exposure [12,14], and
`many lack information on the number of exposures [5]
`or the number of patients [6]. Prospective trials may
`include only incident patients [3,4], only prevalent
`patients [15,16] or higher i.v.
`iron doses than are
`generally given [17]. We found not only that the rate
`of i.v. iron dextran ADEs requiring resuscitative medi-
`cation is 0.035% (seven events out of 20 213 patients
`or exposures) but also that this risk is confined to
`incident patients. We conclude the obvious,
`that
`
`1441
`
`in the absence of specific
`evidence of iron safety,
`information on numbers of doses, prevalent patients
`and incident patients,
`is unreliable and may be
`misleading.
`Investigators recently examined the safety of iron
`dextran using the large voluntary reporting database
`of the World Health Organization [7]. Their finding
`that serious ADE rates for iron dextran range from
`11.6 to 57.9 per million exposures is higher than
`that of our current study (7 per 1 066 099 actual
`exposures) and those of others [5,6]. Either an under-
`estimate of exposures or an increased proportion
`of naive patients could explain the higher results.
`In the WHO study, the number of doses administered
`was projected and the method for calculating the
`projection was not completely described. Since the
`study period included agents new to the market,
`the results could also reflect a high proportion of
`incident patients who would therefore be at increased
`risk for ADEs.
`for current
`Our studies provide strong support
`European Best Practice Guidelines [18] and NKF-K/
`DOQI [19] recommendations regarding administration
`of a test dose of iron dextran. These evidence-based
`clinical practice guidelines recommend administration
`of a single 25 mg test dose. Though no further test
`doses are required, our findings suggest that special
`precautions should also be taken with administration
`of the first therapeutic dose, and, as the previous
`report confirmed [6], with administration of the first
`dose of each newly planned series. The dispropor-
`tionate risk born by incident patients suggests that
`introduction of
`i.v.
`iron dextran to a previously
`unexposed population is likely to provoke more life-
`threatening reactions than would be expected on the
`basis of previous experience with mixed populations
`of incident and prevalent patients. In the current
`study, for example, 58% of the study population
`consisted of prevalent patients previously exposed to
`iron dextran. Thus,
`if the entire study population
`had been iron dextran naı¨ ve, the projected number
`of life-threatening reactions encountered would have
`been 2- to 3-fold higher. Taken together, our findings
`and those of others prompt
`the conclusion that
`caution and, in particular, ready access to resuscitative
`medication, should attend each administration of iron
`dextran.
`
`Acknowledgements. This
`supported by Luitpold
`study was
`Pharmaceuticals and American Regent, Inc.
`
`Conflict of interest statement. D.B.V.W. has served as investigator,
`consultant and speaker, and B.A.J.W. as an investigator,
`for
`American Regent, Inc.
`
`References
`
`1. Danielson BG. Structure, chemistry, and pharmacokinetics
`J Am Soc Nephrol
`of
`intravenous
`iron agents.
`2004;
`15 [Suppl 2]: S93–S98
`2. Michael B, Coyne DW, Fishbane S et al. Sodium ferric gluconate
`complex in hemodialysis patients: adverse reactions compared
`to placebo and iron dextran. Kidney Int 2002; 61: 1830–1839
`
`

`

`Downloaded from
`
`http://ndt.oxfordjournals.org/
`
` at Pennsylvania State University on March 4, 2014
`
`1442
`
`B. A. J. Walters and D. B. Van Wyck
`
`3. Charytan C, Levin N, Al Saloum M, Hafeez T, Gagnon S,
`Van Wyck DB. Efficacy and safety of iron sucrose for iron
`deficiency in patients with dialysis-associated anemia: North
`American clinical trial. Am J Kidney Dis 2001; 37: 300–307
`4. Coyne DW, Adkinson NF, Nissenson AR et al. Sodium
`ferric gluconate complex in hemodialysis patients. II. Adverse
`reactions in iron dextran-sensitive and dextran-tolerant patients.
`Kidney Int 2003; 63: 217–224
`5. Fishbane S, Ungureanu VD, Maesaka JK, Kaupke CJ, Lim V,
`Wish J. The safety of intravenous iron dextran in hemodialysis
`patients. Am J Kidney Dis 1996; 28: 529–534
`6. Fletes R, Lazarus JM, Gage J, Chertow GM. Suspected iron
`dextran-related adverse drug events in hemodialysis patients.
`Am J Kidney Dis 2001; 37: 743–749
`7. Chertow GM, Mason PD, Vaage-Nilsen O, Ahlmen J. On the
`iron formulations. Nephrol Dial
`relative safety of parenteral
`Transplant 2004; 19: 1571–1575
`8. Fishbane S, Kowalski EA, Imbriano LJ, Maesaka JK. The
`evaluation of iron status in hemodialysis patients. J Am Soc
`Nephrol 1996; 7: 2654–2657
`9. Bailie GR, Johnson CA, Mason NA. Parenteral iron use in
`the management of anemia in end-stage renal disease patients.
`Am J Kidney Dis 2000; 35: 1–12
`10. Sunder-Plassmann G, Horl WH. Safety aspects of parenteral
`iron in patients with end-stage renal disease. Drug Safety 1997;
`17: 241–250
`
`Intravenous
`
`iron–dextran complex. Blood
`
`11. Van Wyck DB. Value of a test dose of iron dextran. Semin
`Dialysis 1999; 12: 53
`12. Hamstra RD, Block MH, Schocket AL. Intravenous iron dex-
`tran in clinical medicine. J Am Med Assoc 1980; 243: 1726–1731
`13. Hamstra RD, Block MH, Schocket AL. Intravenous iron
`dextran in clinical medicine. J Am Med Assoc 1980; 243:
`1726–1731
`14. Wallerstein RO.
`1968; 32: 690–695
`15. Folkert VW, Michael B, Agarwal R et al. Chronic use
`of sodium ferric gluconate complex in hemodialysis patients:
`safety of higher-dose (> or ¼250 mg) administration. Am J
`Kidney Dis 2003; 41: 651–657
`16. Michael B, Coyne DW, Folkert VW, Dahl NV, Warnock DG.
`Sodium ferric gluconate complex in haemodialysis patients:
`long-term safety. Nephrol Dial
`a prospective evaluation of
`Transplant 2004; 19: 1576–1580
`17. McCarthy JT, Regnier CE, Loebertmann CL, Bergstralh EJ.
`Adverse events
`in chronic hemodialysis patients
`receiving
`intravenous iron dextran—a comparison of two products. Am J
`Nephrol 2000; 20: 455–462
`18. SECTION III. Treatment of renal anaemia. Nephrol Dial
`Transplant 2004; 19: II16–II31
`19. IV. NKF-K/DOQI Clinical Practice Guidelines for Anemia of
`Chronic Kidney Disease: update 2000. Am J Kidney Dis 2001;
`37: S182–S238
`
`Received for publication: 1.7.04
`Accepted in revised form: 8.3.05
`
`