Risk of Nephrogenic Systemic Fibrosis in Patients With Stage 4 or 5 Chronic Kidney Disease Receiving a Group II Gadolinium-Based Contrast Agent: A Systematic Review and Meta-analysis | Chronic Kidney Disease | JAMA Internal Medicine | JAMA Network
[Skip to Navigation]
Figure 1.  Study Flow Diagram
Study Flow Diagram
Figure 2.  Incidence and Upper Bound of 95% CI of Nephrogenic Systemic Fibrosis (NSF) in Patients With Stage 4 or 5 Chronic Kidney Disease by Study
Incidence and Upper Bound of 95% CI of Nephrogenic Systemic Fibrosis (NSF) in Patients With Stage 4 or 5 Chronic Kidney Disease by Study

The 95% CI data are stratified by 16 studies included in the meta-analysis and represent NSF incidence across all studies (0 of 4931 [0%]). Pooled refers to pooled exposures of all studies.

Figure 3.  Incidence and Upper Bound of 95% CI of Nephrogenic Systemic Fibrosis (NSF) in Patients With Stage 4 or 5 Chronic Kidney Disease by Agent
Incidence and Upper Bound of 95% CI of Nephrogenic Systemic Fibrosis (NSF) in Patients With Stage 4 or 5 Chronic Kidney Disease by Agent

The 95% CI data are stratified by the 4 gadolinium-based contrast agents included in the study and represent NSF incidence across all studies (0 of 4931 [0%]). Pooled refers to pooled exposures of all agents.

Table 1.  Characteristics of Included Studies
Characteristics of Included Studies
Table 2.  Study Characteristics Related to Group II GBCA Administration and NSF
Study Characteristics Related to Group II GBCA Administration and NSF
1.
Agarwal  R, Brunelli  SM, Williams  K, Mitchell  MD, Feldman  HI, Umscheid  CA.  Gadolinium-based contrast agents and nephrogenic systemic fibrosis: a systematic review and meta-analysis.  Nephrol Dial Transplant. 2009;24(3):856-863. doi:10.1093/ndt/gfn593PubMedGoogle ScholarCrossref
2.
Gibson  SE, Farver  CF, Prayson  RA.  Multiorgan involvement in nephrogenic fibrosing dermopathy: an autopsy case and review of the literature.  Arch Pathol Lab Med. 2006;130(2):209-212. doi:10.1043/1543-2165(2006)130[209:MIINFD]2.0.CO:2PubMedGoogle Scholar
3.
Mendoza  FA, Artlett  CM, Sandorfi  N, Latinis  K, Piera-Velazquez  S, Jimenez  SA.  Description of 12 cases of nephrogenic fibrosing dermopathy and review of the literature.  Semin Arthritis Rheum. 2006;35(4):238-249. doi:10.1016/j.semarthrit.2005.08.002PubMedGoogle ScholarCrossref
4.
Zhang  B, Liang  L, Chen  W, Liang  C, Zhang  S.  An updated study to determine association between gadolinium-based contrast agents and nephrogenic systemic fibrosis.  PLoS One. 2015;10(6):e0129720. doi:10.1371/journal.pone.0129720PubMedGoogle Scholar
5.
Kitajima  K, Maeda  T, Watanabe  S, Ueno  Y, Sugimura  K.  Recent topics related to nephrogenic systemic fibrosis associated with gadolinium-based contrast agents.  Int J Urol. 2012;19(9):806-811. doi:10.1111/j.1442-2042.2012.03042.xPubMedGoogle ScholarCrossref
6.
US Food and Drug Administration. FDA safety communication: new warnings for using gadolinium-based contrast agents in patients with kidney dysfunction. https://www.fda.gov/Drugs/DrugSafety/ucm223966.htm. Updated February 6, 2018. Accessed May 9, 2019.
7.
Altun  E, Martin  DR, Wertman  R, Lugo-Somolinos  A, Fuller  ER  III, Semelka  RC.  Nephrogenic systemic fibrosis: change in incidence following a switch in gadolinium agents and adoption of a gadolinium policy—report from two US universities.  Radiology. 2009;253(3):689-696. doi:10.1148/radiol.2533090649PubMedGoogle ScholarCrossref
8.
Martin  DR, Krishnamoorthy  SK, Kalb  B,  et al.  Decreased incidence of NSF in patients on dialysis after changing gadolinium contrast-enhanced MRI protocols.  J Magn Reson Imaging. 2010;31(2):440-446. doi:10.1002/jmri.22024PubMedGoogle ScholarCrossref
9.
Wang  Y, Alkasab  TK, Narin  O,  et al.  Incidence of nephrogenic systemic fibrosis after adoption of restrictive gadolinium-based contrast agent guidelines.  Radiology. 2011;260(1):105-111. doi:10.1148/radiol.11102340PubMedGoogle ScholarCrossref
10.
Balogh  EPMB, Ball  JR; Committee on Diagnostic Error in Health Care; Board on Health Care Services; Institute of Medicine; The National Academies of Sciences, Engineering, and Medicine.  Improving Diagnosis in Health Care. Washington, DC: National Academies Press; 2015.
11.
Abraham  JL, Thakral  C, Skov  L, Rossen  K, Marckmann  P.  Dermal inorganic gadolinium concentrations: evidence for in vivo transmetallation and long-term persistence in nephrogenic systemic fibrosis.  Br J Dermatol. 2008;158(2):273-280. doi:10.1111/j.1365-2133.2007.08335.xPubMedGoogle ScholarCrossref
12.
Morcos  SK.  Extracellular gadolinium contrast agents: differences in stability.  Eur J Radiol. 2008;66(2):175-179. doi:10.1016/j.ejrad.2008.01.025PubMedGoogle ScholarCrossref
13.
Abujudeh  HH, Rolls  H, Kaewlai  R,  et al.  Retrospective assessment of prevalence of nephrogenic systemic fibrosis (NSF) after implementation of a new guideline for the use of gadobenate dimeglumine as a sole contrast agent for magnetic resonance examination in renally impaired patients.  J Magn Reson Imaging. 2009;30(6):1335-1340. doi:10.1002/jmri.21976PubMedGoogle ScholarCrossref
14.
Alhadad  A, Sterner  G, Svensson  Å, Alhadad  H, Leander  P.  Incidence of nephrogenic systemic fibrosis at a large university hospital in Sweden.  Scand J Urol Nephrol. 2012;46(1):48-53. doi:10.3109/00365599.2011.621142PubMedGoogle ScholarCrossref
15.
Amet  S, Launay-Vacher  V, Clément  O,  et al.  Incidence of nephrogenic systemic fibrosis in patients undergoing dialysis after contrast-enhanced magnetic resonance imaging with gadolinium-based contrast agents: the Prospective Fibrose Nephrogénique Systémique study.  Invest Radiol. 2014;49(2):109-115. doi:10.1097/RLI.0000000000000000PubMedGoogle ScholarCrossref
16.
Bruce  R, Wentland  AL, Haemel  AK,  et al.  Incidence of nephrogenic systemic fibrosis using gadobenate dimeglumine in 1423 patients with renal insufficiency compared with gadodiamide.  Invest Radiol. 2016;51(11):701-705. doi:10.1097/RLI.0000000000000259PubMedGoogle ScholarCrossref
17.
Chrysochou  C, Power  A, Shurrab  AE,  et al.  Low risk for nephrogenic systemic fibrosis in nondialysis patients who have chronic kidney disease and are investigated with gadolinium-enhanced magnetic resonance imaging.  Clin J Am Soc Nephrol. 2010;5(3):484-489. doi:10.2215/CJN.06580909PubMedGoogle ScholarCrossref
18.
Heinz-Peer  G, Neruda  A, Watschinger  B,  et al.  Prevalence of NSF following intravenous gadolinium-contrast media administration in dialysis patients with endstage renal disease.  Eur J Radiol. 2010;76(1):129-134. doi:10.1016/j.ejrad.2009.06.028PubMedGoogle ScholarCrossref
19.
Janus  N, Launay-Vacher  V, Karie  S,  et al.  Prevalence of nephrogenic systemic fibrosis in renal insufficiency patients: results of the FINEST study.  Eur J Radiol. 2010;73(2):357-359. doi:10.1016/j.ejrad.2008.11.021PubMedGoogle ScholarCrossref
20.
Michaely  HJ, Aschauer  M, Deutschmann  H,  et al.  Gadobutrol in renally impaired patients: results of the GRIP study.  Invest Radiol. 2017;52(1):55-60. doi:10.1097/RLI.0000000000000307PubMedGoogle ScholarCrossref
21.
Nandwana  SB, Moreno  CC, Osipow  MT, Sekhar  A, Cox  KL.  Gadobenate dimeglumine administration and nephrogenic systemic fibrosis: is there a real risk in patients with impaired renal function?  Radiology. 2015;276(3):741-747. doi:10.1148/radiol.2015142423PubMedGoogle ScholarCrossref
22.
Reilly  RF.  Risk for nephrogenic systemic fibrosis with gadoteridol (ProHance) in patients who are on long-term hemodialysis.  Clin J Am Soc Nephrol. 2008;3(3):747-751. doi:10.2215/CJN.05721207PubMedGoogle ScholarCrossref
23.
Smorodinsky  E, Ansdell  DS, Foster  ZW,  et al.  Risk of nephrogenic systemic fibrosis is low in patients with chronic liver disease exposed to gadolinium-based contrast agents.  J Magn Reson Imaging. 2015;41(5):1259-1267. doi:10.1002/jmri.24650PubMedGoogle ScholarCrossref
24.
Soulez  G, Bloomgarden  DC, Rofsky  NM,  et al.  Prospective cohort study of nephrogenic systemic fibrosis in patients with stage 3-5 chronic kidney disease undergoing MRI with injected gadobenate dimeglumine or gadoteridol.  AJR Am J Roentgenol. 2015;205(3):469-478. doi:10.2214/AJR.14.14268PubMedGoogle ScholarCrossref
25.
Soyer  P, Dohan  A, Patkar  D, Gottschalk  A.  Observational study on the safety profile of gadoterate meglumine in 35,499 patients: the SECURE study.  J Magn Reson Imaging. 2017;45(4):988-997. doi:10.1002/jmri.25486PubMedGoogle ScholarCrossref
26.
Tsushima  Y, Awai  K, Shinoda  G,  et al.  Post-marketing surveillance of gadobutrol for contrast-enhanced magnetic resonance imaging in Japan.  Jpn J Radiol. 2018;36(11):676-685. doi:10.1007/s11604-018-0778-4PubMedGoogle ScholarCrossref
27.
Young  LK, Matthew  SZ, Houston  JG.  Absence of potential gadolinium toxicity symptoms following 22,897 gadoteric acid (Dotarem®) examinations, including 3,209 performed on renally insufficient individuals.  Eur Radiol. 2019;29(4):1922-1930. doi:10.1007/s00330-018-5737-zPubMedGoogle ScholarCrossref
28.
ACR Committee on Drugs and Contrast Media. ACR manual on contrast media, version 10.3. https://www.acr.org/-/media/ACR/Files/Clinical-Resources/Contrast_Media.pdf. Published June 2018. Accessed May 9, 2019.
29.
European Society of Urogenital Radiology. ESUR guidelines on contrast agents, version 10.0. http://www.esur-cm.org/index.php/a-general-adverse-reactions-2. Published March 2018. Accessed May 9, 2019.
30.
Moher  D, Liberati  A, Tetzlaff  J, Altman  DG.  Preferred Reporting Items for Systematic Reviews and Meta-analyses: the PRISMA statement.  Int J Surg. 2010;8(5):336-341. doi:10.1371/journal.pmed.1000097Google ScholarCrossref
31.
Viera  AJ, Garrett  JM.  Understanding interobserver agreement: the kappa statistic.  Fam Med. 2005;37(5):360-363.PubMedGoogle Scholar
32.
Gagnier  JJ, Morgenstern  H, Chess  L.  Interventions designed to prevent anterior cruciate ligament injuries in adolescents and adults: a systematic review and meta-analysis.  Am J Sports Med. 2013;41(8):1952-1962. doi:10.1177/0363546512458227PubMedGoogle ScholarCrossref
33.
Katayama  H, Yamaguchi  K, Kozuka  T, Takashima  T, Seez  P, Matsuura  K.  Adverse reactions to ionic and nonionic contrast media: a report from the Japanese Committee on the Safety of Contrast Media.  Radiology. 1990;175(3):621-628. doi:10.1148/radiology.175.3.2343107PubMedGoogle ScholarCrossref
34.
Behzadi  AH, Zhao  Y, Farooq  Z, Prince  MR.  Immediate allergic reactions to gadolinium-based contrast agents: a systematic review and meta-analysis.  Radiology. 2018;286(2):471-482. doi:10.1148/radiol.2017162740PubMedGoogle ScholarCrossref
35.
Schieda  N, Blaichman  JI, Costa  AF,  et al.  Gadolinium-based contrast agents in kidney disease: a comprehensive review and clinical practice guideline issued by the Canadian Association of Radiologists.  Can J Kidney Health Dis. 2018;5:2054358118778573. doi:10.1177/2054358118778573PubMedGoogle Scholar
36.
Davenport  MS, Cohan  RH, Ellis  JH.  Contrast media controversies in 2015: imaging patients with renal impairment or risk of contrast reaction.  AJR Am J Roentgenol. 2015;204(6):1174-1181. doi:10.2214/AJR.14.14259PubMedGoogle ScholarCrossref
37.
Dekkers  IA, van der Molen  AJ.  Propensity score matching as a substitute for randomized controlled trials on acute kidney injury after contrast media administration: a systematic review.  AJR Am J Roentgenol. 2018;211(4):822-826. doi:10.2214/AJR.17.19499PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    December 9, 2019

    Risk of Nephrogenic Systemic Fibrosis in Patients With Stage 4 or 5 Chronic Kidney Disease Receiving a Group II Gadolinium-Based Contrast Agent: A Systematic Review and Meta-analysis

    Author Affiliations
    • 1Department of Radiology, University of Michigan, Ann Arbor
    • 2Michigan Radiology Quality Collaborative, University of Michigan, Ann Arbor
    • 3Department of Orthopaedic Surgery, University of Michigan, Ann Arbor
    • 4Department of Epidemiology, University of Michigan, Ann Arbor
    • 5Taubman Health Sciences Library, University of Michigan, Ann Arbor
    • 6Dana-Farber Cancer Institute, Department of Radiation Oncology, Brigham and Women’s Hospital, Boston, Massachusetts
    • 7Department of Urology, Michigan Medicine, Ann Arbor
    JAMA Intern Med. 2020;180(2):223-230. doi:10.1001/jamainternmed.2019.5284
    Key Points

    Question  What is the risk of nephrogenic systemic fibrosis in patients with stage 4 or 5 chronic kidney disease receiving a group II gadolinium-based contrast agent?

    Findings  In this systematic review and meta-analysis of 16 unique studies and 4931 patients, the pooled incidence of nephrogenic systemic fibrosis after administration of a group II gadolinium-based contrast agent in patients with stage 4 or 5 chronic kidney disease was 0%; the upper bound of the 95% CI was 0.07%.

    Meaning  Findings suggest that the risk of nephrogenic systemic fibrosis from group II gadolinium-based contrast agent administration in stage 4 or 5 chronic kidney disease is likely less than 0.07%; potential diagnostic harms of withholding group II gadolinium-based contrast agents for indicated examinations may outweigh the risk of nephrogenic systemic fibrosis in this population.

    Abstract

    Importance  Risk of nephrogenic systemic fibrosis (NSF) to individual patients with stage 4 or 5 chronic kidney disease (CKD; defined as estimated glomerular filtration rate of <30 mL/min/1.73 m2) who receive a group II gadolinium-based contrast agent (GBCA) is not well understood or summarized in the literature.

    Objective  To assess the pooled risk of NSF in patients with stage 4 or 5 CKD receiving a group II GBCA.

    Data Sources  A health sciences informationist searched the Ovid (MEDLINE and MEDLINE Epub Ahead of Print, In-Process & Other Non-Indexed Citation, and Daily and Versions), Embase, Cochrane Central Register of Controlled Trials, Web of Science, and Open Grey databases from inception to January 29, 2019, yielding 2700 citations.

    Study Selection  Citations were screened for inclusion in a multistep process. Agreement for final cohort inclusion was determined by 2 blinded screeners using Cohen κ. Inclusion criteria consisted of stage 4 or 5 CKD with or without dialysis, administration of an unconfounded American College of Radiology classification group II GBCA (gadobenate dimeglumine, gadobutrol, gadoterate meglumine, or gadoteridol), and incident NSF as an outcome. Conference abstracts, retracted manuscripts, narrative reviews, editorials, case reports, and manuscripts not reporting total group II GBCA administrations were excluded.

    Data Extraction and Synthesis  Data extraction was performed for all studies by a single investigator, including publication details, study design and time frame, patient characteristics, group II GBCA(s) administered, total exposures for patients with stage 4 or stage 5 CKD, total cases of unconfounded NSF, reason for GBCA administration, follow-up duration, loss to follow-up, basis for NSF screening, and diagnosis.

    Main Outcomes and Measures  Pooled incidence of NSF and the associated upper bound of a 2-sided 95% CI (risk estimate) for the pooled data and each of the 4 group II GBCAs.

    Results  Sixteen unique studies with 4931 patients were included (κ = 0.68) in this systematic review and meta-analysis. The pooled incidence of NSF was 0 of 4931 (0%; upper bound of 95% CI, 0.07%). The upper bound varied owing to different sample sizes for gadobenate dimeglumine (0 of 3167; upper bound of 95% CI, 0.12%), gadoterate meglumine (0 of 1204; upper bound of 95% CI, 0.31%), gadobutrol (0 of 330; upper bound of 95% CI, 1.11%), and gadoteridol (0 of 230; upper bound of 95% CI, 1.59%).

    Conclusions and Relevance  This study’s findings suggest that the risk of NSF from group II GBCA administration in stage 4 or 5 CKD is likely less than 0.07%. The potential diagnostic harms of withholding group II GBCA for indicated examinations may outweigh the risk of NSF in this population.

    Trial Registration  PROSPERO identifier: CRD42019123284

    Introduction

    Nephrogenic systemic fibrosis (NSF) is a rare, potentially fatal condition caused by iatrogenic gadolinium administration in patients with acute kidney injury or stage 4 or 5 chronic kidney disease (CKD), defined as an estimated glomerular filtration rate (eGFR) of less than 30 mL/min/1.73 m2.1-4 After more than 500 cases of NSF were reported from 1997 to 2007, regulations were adopted to prevent NSF.5 In 2007, the US Food and Drug Administration mandated a black box warning advising avoidance of all gadolinium-containing contrast agents (GBCA) in at-risk patients.5 The label was updated in 2010 to contain recommendations for health care professionals regarding kidney function screening, use of lower-risk GBCAs, and decreasing GBCA dose.6 Such recommendations informed hospital policies and were successful in effectively eliminating the disease.7-9 However, they also resulted in denial or delay of clinically indicated, contrast-enhanced magnetic resonance imaging (MRI) in patients with severe kidney disease, resulting in the undermeasured indirect harms of misdiagnosis and delayed diagnosis.10 In addition, the guidelines were applied to all GBCAs regardless of gadolinium-chelate lability or association with NSF.11,12

    Accumulating literature8,13-27 and newer guidelines28,29 have recognized that not all GBCAs have the same risk of NSF. The American College of Radiology (ACR) manual on contrast media, version 10.328 and the European Society of Urogenital Radiology guidelines on contrast agents, version 10.029 recognize differences in risk of NSF between GBCAs and classify GBCAs into 3 distinct (albeit slightly different) groups.28,29Quiz Ref ID The ACR terms the lowest-risk GBCAs as group II agents (gadobenate dimeglumine, gadoteridol, gadoterate meglumine, and gadobutrol), representing those GBCAs with “very low, if any, risk of NSF development.”28(p85) Both guidelines have been updated recently to indicate that, for the lowest-risk GBCAs, kidney function measurement is not obligatory and that indicated contrast-enhanced MRI with a low-risk GBCA should not be denied on the basis of NSF risk alone.28,29

    Unfortunately, the specific risk to individual patients is not well understood or summarized in the literature. Knowledge of this risk is important for counseling and risk-benefit decision-making in individual patients. Establishing these risk estimates may provide an evidence basis for policy makers and physicians who otherwise may hesitate to administer these agents to patients with stage 4 or 5 CKD. The purpose of this systematic review and meta-analysis is to assess the pooled risk of NSF in patients with stage 4 or 5 CKD receiving a group II GBCA.

    Methods

    This systematic review and meta-analysis was compliant with the Health Insurance Portability and Accountability Act and was exempt based on University of Michigan institutional review board exemption self-regulated status owing to the use of published data with no new study participants. We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline.30

    Eligibility Criteria

    Quiz Ref IDIncluded studies evaluated human participants with stage 4 or 5 CKD (eGFR, <30 mL/min/1.73 m2) and/or receiving dialysis who underwent administration of a group II GBCA (gadobenate dimeglumine, gadobutrol, gadoterate meglumine, or gadoteridol). The outcome measure required for inclusion was assessment of unconfounded incidence of NSF. All follow-up interval lengths were included. Conference abstracts, retracted manuscripts, narrative reviews, editorials, case reports, and manuscripts not reporting total group II GBCA administrations were excluded (Figure 1).

    Data Sources and Searches

    Comprehensive searches were performed by an expert health sciences informationist (M.P.M.) from inception to January 29, 2019, in the following databases: Ovid (MEDLINE and MEDLINE Epub Ahead of Print, In-Process & Other Non-Indexed Citation, and Daily and Versions), Embase, Cochrane Central Register of Controlled Trials (Wiley), Web of Science (Clarivate), and Open Grey. Each search consisted of CKD and group II GBCA concept blocks, with combinations of controlled headings (when possible) and title, abstract, and keyword terms. No date, language, or other restriction were incorporated into the searches. Duplicate citations were removed in Endnote X8 (Clarivate Analytics). Complete search strategies are available in eMethods 1 in the Supplement.

    Study Selection

    Studies were screened for inclusion using a multistep process summarized in the study flow diagram (Figure 1). Search results returned 2700 citations, which were screened at the title and abstract level by 2 study team members (1400 by P.R.S. and 1300 by S.A.W.). A sample set of 100 citations was randomly cross-reviewed by the study team member with more years of experience (P.R.S.) to ensure consistency.

    After the initial screening, the remaining citations (n = 62) were all reviewed at the manuscript level by 2 blinded study team members (S.A.W. and P.R.S.) (eMethods 2 in the Supplement). Agreement for inclusion was calculated using the Cohen κ with the following scale31: 0.01 to 0.20 indicates slight; 0.21 to 0.40, fair; 0.41 to 0.60, moderate; 0.61 to 0.80, substantial; and 0.81 to 0.99, almost perfect. Disagreements were resolved by consensus discussion. A third study team member (M.S.D.) adjudicated when needed.

    Data Collection and Data Items

    Data extraction was performed for all studies by a single study team member (P.R.S.). Extracted data included publication details, study design and time frame, patient characteristics, group II GBCA(s) administered, total exposures for patients with stage 4 or 5 CKD, total cases of unconfounded NSF, reason for GBCA administration, follow-up duration, loss to follow-up, basis for NSF screening, and diagnosis.

    Risk of Bias Analysis

    Risk of bias analysis of the included studies was performed by a single study team member (P.R.S.). Criteria used for assessment were based on previously described measures for nonrandomized cohort studies.32 Each possible source of bias was assessed as being fulfilled (yes, meaning bias is unlikely to be present), unfulfilled (no, meaning bias is likely to be present), or unknown (meaning information is inadequate or inapplicable to study design). Certain components of the risk of bias assessment were scored as not applicable when inapplicable to study design or results.

    Statistical Analysis

    The principal summary measure is the pooled incidence of NSF and associated upper bound of the 95% CI (risk estimate) in patients with stage 4 or 5 CKD receiving a group II GBCA. Subanalyses were performed to assess risk estimates on a per-study basis and for each of the 4 individual group II GBCAs. Data analysis was performed with Stata, version 15.2 (StataCorp LLC).

    Results

    Initial database searches returned 2700 unique citations (Figure 1). After title and abstract review, 62 potential citations remained. After full text review, a final cohort of 16 citations including 4931 patients was available for analysis. Interrater agreement in determining the final study cohort from the 62 screened citations was substantial (κ = 0.68; 95% CI, 0.49-0.87).

    Characteristics of included studies are provided in Table 1.8,13-27 Studies were published from May 2008 through April 2019. The time frame of investigation across all studies spanned 1997 through 2017. The included studies were a mix of retrospective cohort (11 of 16 [69%]) and prospective cohort (5 of 16 [31%]) designs. Study representation was international, with most of the studies performed in Europe, including 2 multiple-country studies (8 of 16 [50%]) and the United States (7 of 16 [44%]). Multicenter studies constituted 7 of 16 (44%) of the included cohort.

    The incidence of NSF in patients with stage 4 or 5 CKD across all 16 studies was 0 of 4931 (0%). The upper bound of the 2-sided 95% CI (1-sided 97.5% CI) for this pooled estimate was 0.07% (Figure 2). Study-specific details regarding characteristics of GBCA exposure, number of GBCA exposures, and reference standard for NSF assessment are provided in Table 2.

    Upper bounds of 95% CIs varied on a study-specific basis (0.26%-52.2%) owing to differences in study-specific eligible sample sizes (Figure 2). Follow-up intervals for NSF detection ranged from 3 to 72 months; follow-up interval was unknown in 2 of 16 studies. The reference standard for NSF was most commonly a retrospective medical review (11 of 16 [68.8%]).

    The pooled risk of NSF stratified by group II GBCA is provided in Figure 3. The greatest safety margin (ie, largest sample size) was for gadobenate dimeglumine (upper bound 95% CI, 0.12% [0 of 3167]). Upper bound 95% CIs for the other group II GBCAs were 1.11% (0 of 330) for gadobutrol, 0.31% (0 of 1204) for gadoterate meglumine, and 1.59% (0 of 230) for gadoteridol.

    The risk-of-bias assessment is summarized in the eTable in the Supplement. Because the incidence of NSF across all studies was 0%, certain factors in the risk-of-bias assessment related to clustering of outcomes at analysis and adjustment for analysis were not applicable. The most common methodological limitation across the included studies was the unblinding of assessors of NSF to the intervention of GBCA administration (15 of 16 studies [94%]). Strengths of all 16 studies included uniformity in the absence of the outcome (NSF) at the start of investigations, consistency in administration of the intervention (GBCA administration) across all groups, and absence of bias between any potential groups in GBCA administration. A potential for funding bias related to industry support was reported in 7 of 16 studies (44%).

    Discussion

    Across 16 studies and 4931 administrations, we found the pooled risk of NSF from group II GBCAs in patients with stage 4 or 5 CKD to be 0% (upper bound of 95% CI, 0.07%). Quiz Ref IDThis finding indicates the per-patient risk of NSF from group II GBCA administration in stage 4 or 5 CKD is likely less than 0.07%. This risk can be compared with the risk of a severe allergiclike contrast reaction, which has been estimated to be approximately 0.04% for modern low-osmolality iodinated contrast agents33 and approximately 0.006% to 0.02% for group II GBCAs.34 Despite existing US Food and Drug Administration guidelines indicating that all GBCAs are contraindicated if the eGFR is less than 30 mL/min/1.73 m2, these data suggest that group II GBCAs are relatively safe in patients with severe CKD, and their benefits may exceed their risks for indicated examinations. Consistent with our results, recent updates to the ACR,28 European Society of Urogenital Radiology,29 and Canadian Association of Radiologists35 guidelines support use of indicated low-risk GBCAs in this setting.

    In comparison with the risk of contrast-induced acute kidney injury, these data indicate that, in patients with stage 4 or 5 CKD who are not receiving dialysis, there is a clearer safety profile for contrast-enhanced MRI using a single-dose group II GBCA than there is for contrast-enhanced computed tomography using a single-dose low-osmolality iodinated contrast agent.28,29,36,37 The number needed to harm from low-osmolality iodinated contrast agents (ie, contrast-induced acute kidney injury) has been estimated to be between 1 in 6 and no harm evident (ie, indicating substantial uncertainty) based on recent large, propensity score–adjusted retrospective cohort studies.36,37 In both cases (contrast-enhanced computed tomography and contrast-enhanced MRI), the harms of delayed diagnosis and misdiagnosis resulting from the withholding of contrast material in at-risk patients are incompletely measured but likely real.10 For many disease states, unenhanced imaging has poorer diagnostic accuracy than contrast-enhanced imaging, increasing the risk of diagnostic error and iatrogenic morbidity and mortality.10

    Group II GBCAs include 3 macrocyclic agents with 100% renal excretion (gadoteridol, gadoterate meglumine, and gadobutrol) and 1 linear ionic agent with approximately 95% renal and 5% hepatobiliary excretion (gadobenate dimeglumine).28Quiz Ref ID Of the 4931 administrations we evaluated, 3167 were exposed to gadobenate dimeglumine. Therefore, most of the safety evidence in this setting is for a specific linear ionic group II GBCA with partial hepatobiliary excretion. Hepatobiliary excretion may offer a protective advantage against NSF by providing an alternative clearance mechanism in patients with severely impaired kidney function.

    Previous meta-analyses on GBCA and NSF risk have been heavily weighted by group I GBCAs.1,4 Agarwal et al1 analyzed 7 studies published from 2006 to 2007 with 4276 patients and found an odds ratio of 26.7 (95% CI, 10.3-69.4) for the risk of NSF. Six of the studies evaluated majority or sole group I GBCA exposure, and 1 study had an unknown exposure history. Zhang et al4 performed an updated meta-analysis of 11 studies published from 2006 to 2012 with 5405 patients and found an odds ratio of 16.5 (95% CI, 7.5-36.5) for the risk of NSF, suggesting a decline in risk since preventive strategies were introduced. However, because their inclusion criteria required patients diagnosed with NSF, 3 studies of GBCA without evidence of NSF published in 2010, 2013, and 2014 were excluded. Therefore, the existing published meta-analyses1,4 are likely not directly relevant to the risk of NSF from group II GBCA.

    Strengths and Limitations

    Some strengths of our analysis include its focus on a specific and clinically important question, narrow inclusion criteria, a comprehensive search strategy, dual inclusion methods with high interrater agreement, and a low risk of bias for most domains. Quiz Ref IDCommon weaknesses included general lack of blinding in the included studies, no universal reference standard for the diagnosis of NSF, and insufficient sample size for specific GBCAs. Most of the studies in our cohort (69%) performed retrospective evaluations of the medical records to identify potential cases of NSF, raising the possibility that cases could be missed in situations in which individuals were no longer patients in the system where the MRI was performed. However, most studies provided a minimum follow-up interval for study inclusion and provided mean times for record review following GBCA administration (Table 2). In 7 of the studies included in our analysis, development of NSF also was evaluated in patients who received non–group II GBCAs.8,14-18,23 In 3 of these studies, NSF was observed.8,16,18 These positive controls suggest the lack of NSF detection within our analysis of group II GBCAs was not due solely to methodological biases.

    Although our sample size was large (n = 4931), no NSF events occurred. Therefore, the true risk of NSF in this cohort is unknown. The upper bound of the 95% CI was 0.07%, but this result depended on sample size. The absolute risk could be (for example) nonexistent, 1 in a million, or 1 in 2000. With a larger sample, a more precise estimate would be possible. In addition, the analysis reflects studies performed before and after changes to practice guidelines designed to mitigate NSF risk. Therefore, our results are not a pure reflection of either era. There have been single-digit numbers of reports of unconfounded NSF resulting after exposure to a group II GBCA,35 suggesting that the risk of NSF in high-risk patients receiving a group II GBCA is not zero. Larger series are needed to determine what that risk is. Our analysis is unable to determine the risk of sequential group II GBCA exposures or the risk from group II GBCA administration in the setting of acute kidney injury. The studies we analyzed did not comprehensively or universally address those issues, and this is an area for future investigation. Our analysis was designed to evaluate harms specifically related to development of NSF. It is not a comprehensive assessment of all potential GBCA-related risk (eg, allergiclike reactions, gadolinium retention).

    Current ACR guidelines do not require informed consent before group II GBCA administration.28 If a practice wishes to do so, we would suggest the following: “Current evidence does not support withholding group II GBCAs on the basis of NSF risk alone in patients with stage 4 or 5 CKD. Although there is likely a very small risk of developing NSF (likely less than 0.07%) in this population, if the diagnostic question necessitates the use of a GBCA, use of a group II GBCA is recommended.”

    Conclusions

    The risk of NSF from group II GBCA administration in patients with stage 4 or 5 CKD is likely less than 0.07%. The harms of withholding group II GBCA for indicated examinations may outweigh the risk of NSF in this population. These data support recent updates to ACR and European Society of Urogenital Radiology guidelines28,29 liberalizing use of low-risk GBCAs for indicated examinations in this setting.

    Back to top
    Article Information

    Corresponding Author: Prasad R. Shankar, MD, Department of Radiology, University of Michigan, 1500 E Medical Center Dr, Room B2 A209P, Ann Arbor, MI 48109 (pshankar@med.umich.edu).

    Published Online: December 9, 2019. doi:10.1001/jamainternmed.2019.5284

    Author Contributions: Dr Shankar had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Woolen, Shankar, Gagnier, Singer, Davenport.

    Acquisition, analysis, or interpretation of data: Woolen, Shankar, Gagnier, MacEachern, Davenport.

    Drafting of the manuscript: Woolen, Shankar, Gagnier, MacEachern, Davenport.

    Critical revision of the manuscript for important intellectual content: All authors.

    Statistical analysis: Shankar, Gagnier.

    Administrative, technical, or material support: Shankar, Gagnier, Singer, Davenport.

    Supervision: Shankar, Gagnier, Singer, Davenport.

    Conflict of Interest Disclosures: Dr Gagnier reported receiving personal fees from Rubin Anders Scientific and Bartimus Frickleton Robertson Rader PC outside the submitted work. Dr Singer reported receiving grants from Brigham Research Institute outside the submitted work. Dr Davenport reported receiving royalties from Wolters Kluwer (Genitourinary Imaging: A Core Review) and uptodate.com for articles related to iodinated contrast media. No other disclosures were reported.

    References
    1.
    Agarwal  R, Brunelli  SM, Williams  K, Mitchell  MD, Feldman  HI, Umscheid  CA.  Gadolinium-based contrast agents and nephrogenic systemic fibrosis: a systematic review and meta-analysis.  Nephrol Dial Transplant. 2009;24(3):856-863. doi:10.1093/ndt/gfn593PubMedGoogle ScholarCrossref
    2.
    Gibson  SE, Farver  CF, Prayson  RA.  Multiorgan involvement in nephrogenic fibrosing dermopathy: an autopsy case and review of the literature.  Arch Pathol Lab Med. 2006;130(2):209-212. doi:10.1043/1543-2165(2006)130[209:MIINFD]2.0.CO:2PubMedGoogle Scholar
    3.
    Mendoza  FA, Artlett  CM, Sandorfi  N, Latinis  K, Piera-Velazquez  S, Jimenez  SA.  Description of 12 cases of nephrogenic fibrosing dermopathy and review of the literature.  Semin Arthritis Rheum. 2006;35(4):238-249. doi:10.1016/j.semarthrit.2005.08.002PubMedGoogle ScholarCrossref
    4.
    Zhang  B, Liang  L, Chen  W, Liang  C, Zhang  S.  An updated study to determine association between gadolinium-based contrast agents and nephrogenic systemic fibrosis.  PLoS One. 2015;10(6):e0129720. doi:10.1371/journal.pone.0129720PubMedGoogle Scholar
    5.
    Kitajima  K, Maeda  T, Watanabe  S, Ueno  Y, Sugimura  K.  Recent topics related to nephrogenic systemic fibrosis associated with gadolinium-based contrast agents.  Int J Urol. 2012;19(9):806-811. doi:10.1111/j.1442-2042.2012.03042.xPubMedGoogle ScholarCrossref
    6.
    US Food and Drug Administration. FDA safety communication: new warnings for using gadolinium-based contrast agents in patients with kidney dysfunction. https://www.fda.gov/Drugs/DrugSafety/ucm223966.htm. Updated February 6, 2018. Accessed May 9, 2019.
    7.
    Altun  E, Martin  DR, Wertman  R, Lugo-Somolinos  A, Fuller  ER  III, Semelka  RC.  Nephrogenic systemic fibrosis: change in incidence following a switch in gadolinium agents and adoption of a gadolinium policy—report from two US universities.  Radiology. 2009;253(3):689-696. doi:10.1148/radiol.2533090649PubMedGoogle ScholarCrossref
    8.
    Martin  DR, Krishnamoorthy  SK, Kalb  B,  et al.  Decreased incidence of NSF in patients on dialysis after changing gadolinium contrast-enhanced MRI protocols.  J Magn Reson Imaging. 2010;31(2):440-446. doi:10.1002/jmri.22024PubMedGoogle ScholarCrossref
    9.
    Wang  Y, Alkasab  TK, Narin  O,  et al.  Incidence of nephrogenic systemic fibrosis after adoption of restrictive gadolinium-based contrast agent guidelines.  Radiology. 2011;260(1):105-111. doi:10.1148/radiol.11102340PubMedGoogle ScholarCrossref
    10.
    Balogh  EPMB, Ball  JR; Committee on Diagnostic Error in Health Care; Board on Health Care Services; Institute of Medicine; The National Academies of Sciences, Engineering, and Medicine.  Improving Diagnosis in Health Care. Washington, DC: National Academies Press; 2015.
    11.
    Abraham  JL, Thakral  C, Skov  L, Rossen  K, Marckmann  P.  Dermal inorganic gadolinium concentrations: evidence for in vivo transmetallation and long-term persistence in nephrogenic systemic fibrosis.  Br J Dermatol. 2008;158(2):273-280. doi:10.1111/j.1365-2133.2007.08335.xPubMedGoogle ScholarCrossref
    12.
    Morcos  SK.  Extracellular gadolinium contrast agents: differences in stability.  Eur J Radiol. 2008;66(2):175-179. doi:10.1016/j.ejrad.2008.01.025PubMedGoogle ScholarCrossref
    13.
    Abujudeh  HH, Rolls  H, Kaewlai  R,  et al.  Retrospective assessment of prevalence of nephrogenic systemic fibrosis (NSF) after implementation of a new guideline for the use of gadobenate dimeglumine as a sole contrast agent for magnetic resonance examination in renally impaired patients.  J Magn Reson Imaging. 2009;30(6):1335-1340. doi:10.1002/jmri.21976PubMedGoogle ScholarCrossref
    14.
    Alhadad  A, Sterner  G, Svensson  Å, Alhadad  H, Leander  P.  Incidence of nephrogenic systemic fibrosis at a large university hospital in Sweden.  Scand J Urol Nephrol. 2012;46(1):48-53. doi:10.3109/00365599.2011.621142PubMedGoogle ScholarCrossref
    15.
    Amet  S, Launay-Vacher  V, Clément  O,  et al.  Incidence of nephrogenic systemic fibrosis in patients undergoing dialysis after contrast-enhanced magnetic resonance imaging with gadolinium-based contrast agents: the Prospective Fibrose Nephrogénique Systémique study.  Invest Radiol. 2014;49(2):109-115. doi:10.1097/RLI.0000000000000000PubMedGoogle ScholarCrossref
    16.
    Bruce  R, Wentland  AL, Haemel  AK,  et al.  Incidence of nephrogenic systemic fibrosis using gadobenate dimeglumine in 1423 patients with renal insufficiency compared with gadodiamide.  Invest Radiol. 2016;51(11):701-705. doi:10.1097/RLI.0000000000000259PubMedGoogle ScholarCrossref
    17.
    Chrysochou  C, Power  A, Shurrab  AE,  et al.  Low risk for nephrogenic systemic fibrosis in nondialysis patients who have chronic kidney disease and are investigated with gadolinium-enhanced magnetic resonance imaging.  Clin J Am Soc Nephrol. 2010;5(3):484-489. doi:10.2215/CJN.06580909PubMedGoogle ScholarCrossref
    18.
    Heinz-Peer  G, Neruda  A, Watschinger  B,  et al.  Prevalence of NSF following intravenous gadolinium-contrast media administration in dialysis patients with endstage renal disease.  Eur J Radiol. 2010;76(1):129-134. doi:10.1016/j.ejrad.2009.06.028PubMedGoogle ScholarCrossref
    19.
    Janus  N, Launay-Vacher  V, Karie  S,  et al.  Prevalence of nephrogenic systemic fibrosis in renal insufficiency patients: results of the FINEST study.  Eur J Radiol. 2010;73(2):357-359. doi:10.1016/j.ejrad.2008.11.021PubMedGoogle ScholarCrossref
    20.
    Michaely  HJ, Aschauer  M, Deutschmann  H,  et al.  Gadobutrol in renally impaired patients: results of the GRIP study.  Invest Radiol. 2017;52(1):55-60. doi:10.1097/RLI.0000000000000307PubMedGoogle ScholarCrossref
    21.
    Nandwana  SB, Moreno  CC, Osipow  MT, Sekhar  A, Cox  KL.  Gadobenate dimeglumine administration and nephrogenic systemic fibrosis: is there a real risk in patients with impaired renal function?  Radiology. 2015;276(3):741-747. doi:10.1148/radiol.2015142423PubMedGoogle ScholarCrossref
    22.
    Reilly  RF.  Risk for nephrogenic systemic fibrosis with gadoteridol (ProHance) in patients who are on long-term hemodialysis.  Clin J Am Soc Nephrol. 2008;3(3):747-751. doi:10.2215/CJN.05721207PubMedGoogle ScholarCrossref
    23.
    Smorodinsky  E, Ansdell  DS, Foster  ZW,  et al.  Risk of nephrogenic systemic fibrosis is low in patients with chronic liver disease exposed to gadolinium-based contrast agents.  J Magn Reson Imaging. 2015;41(5):1259-1267. doi:10.1002/jmri.24650PubMedGoogle ScholarCrossref
    24.
    Soulez  G, Bloomgarden  DC, Rofsky  NM,  et al.  Prospective cohort study of nephrogenic systemic fibrosis in patients with stage 3-5 chronic kidney disease undergoing MRI with injected gadobenate dimeglumine or gadoteridol.  AJR Am J Roentgenol. 2015;205(3):469-478. doi:10.2214/AJR.14.14268PubMedGoogle ScholarCrossref
    25.
    Soyer  P, Dohan  A, Patkar  D, Gottschalk  A.  Observational study on the safety profile of gadoterate meglumine in 35,499 patients: the SECURE study.  J Magn Reson Imaging. 2017;45(4):988-997. doi:10.1002/jmri.25486PubMedGoogle ScholarCrossref
    26.
    Tsushima  Y, Awai  K, Shinoda  G,  et al.  Post-marketing surveillance of gadobutrol for contrast-enhanced magnetic resonance imaging in Japan.  Jpn J Radiol. 2018;36(11):676-685. doi:10.1007/s11604-018-0778-4PubMedGoogle ScholarCrossref
    27.
    Young  LK, Matthew  SZ, Houston  JG.  Absence of potential gadolinium toxicity symptoms following 22,897 gadoteric acid (Dotarem®) examinations, including 3,209 performed on renally insufficient individuals.  Eur Radiol. 2019;29(4):1922-1930. doi:10.1007/s00330-018-5737-zPubMedGoogle ScholarCrossref
    28.
    ACR Committee on Drugs and Contrast Media. ACR manual on contrast media, version 10.3. https://www.acr.org/-/media/ACR/Files/Clinical-Resources/Contrast_Media.pdf. Published June 2018. Accessed May 9, 2019.
    29.
    European Society of Urogenital Radiology. ESUR guidelines on contrast agents, version 10.0. http://www.esur-cm.org/index.php/a-general-adverse-reactions-2. Published March 2018. Accessed May 9, 2019.
    30.
    Moher  D, Liberati  A, Tetzlaff  J, Altman  DG.  Preferred Reporting Items for Systematic Reviews and Meta-analyses: the PRISMA statement.  Int J Surg. 2010;8(5):336-341. doi:10.1371/journal.pmed.1000097Google ScholarCrossref
    31.
    Viera  AJ, Garrett  JM.  Understanding interobserver agreement: the kappa statistic.  Fam Med. 2005;37(5):360-363.PubMedGoogle Scholar
    32.
    Gagnier  JJ, Morgenstern  H, Chess  L.  Interventions designed to prevent anterior cruciate ligament injuries in adolescents and adults: a systematic review and meta-analysis.  Am J Sports Med. 2013;41(8):1952-1962. doi:10.1177/0363546512458227PubMedGoogle ScholarCrossref
    33.
    Katayama  H, Yamaguchi  K, Kozuka  T, Takashima  T, Seez  P, Matsuura  K.  Adverse reactions to ionic and nonionic contrast media: a report from the Japanese Committee on the Safety of Contrast Media.  Radiology. 1990;175(3):621-628. doi:10.1148/radiology.175.3.2343107PubMedGoogle ScholarCrossref
    34.
    Behzadi  AH, Zhao  Y, Farooq  Z, Prince  MR.  Immediate allergic reactions to gadolinium-based contrast agents: a systematic review and meta-analysis.  Radiology. 2018;286(2):471-482. doi:10.1148/radiol.2017162740PubMedGoogle ScholarCrossref
    35.
    Schieda  N, Blaichman  JI, Costa  AF,  et al.  Gadolinium-based contrast agents in kidney disease: a comprehensive review and clinical practice guideline issued by the Canadian Association of Radiologists.  Can J Kidney Health Dis. 2018;5:2054358118778573. doi:10.1177/2054358118778573PubMedGoogle Scholar
    36.
    Davenport  MS, Cohan  RH, Ellis  JH.  Contrast media controversies in 2015: imaging patients with renal impairment or risk of contrast reaction.  AJR Am J Roentgenol. 2015;204(6):1174-1181. doi:10.2214/AJR.14.14259PubMedGoogle ScholarCrossref
    37.
    Dekkers  IA, van der Molen  AJ.  Propensity score matching as a substitute for randomized controlled trials on acute kidney injury after contrast media administration: a systematic review.  AJR Am J Roentgenol. 2018;211(4):822-826. doi:10.2214/AJR.17.19499PubMedGoogle ScholarCrossref
    ×