Keywords

Kidney injury; Nephrotoxicity; Diagnostic; Radiology

Introduction

Contrast medium-induced nephropathy (CIN), recently renamed contrast medium-induced acute kidney injury (CIAKI) [1] has become a highly controversial subject. As late as 2002 it was considered a leading cause of hospital-acquired acute kidney injury (AKI) [2]. However, recent propensity score (PS) matched controlled studies comparing patients undergoing computed tomography (CT) with and without intravenous (IV) iodine contrast media (CM) found no association between AKI and CM administration and have questioned the existence of CI-AKI [3-5]. The studies have been criticized due to their retrospective nature, which makes selection bias a serious concern [6,7]. Thus, there is still a great need for proper prospective studies to establish to what extent CI-AKI exists, since this will have a great impact on whether to use CM or not with subsequent implications on diagnostic accuracy. The aim of the present paper is to give a historical review, to discuss the studies questioning the existence of CIAKI, to present two cases of possible CI-AKI and to propose some layouts for future prospective studies.

Literature Review

Historical review

1920s to 1950s: Organic iodine contrast media (CM) for intravascular use have been employed in diagnostic radiology since the late 1920s [8,9]. Occasional cases of uraemia after intravenous pyelography (IVP) were reported during the 1940s and early 1950s [10,11]. The rare publications of AKI following IVP was probably due to the fact that CM was seldom used in patients with impaired renal function simply because IVP yielded too low CM concentration with insufficient diagnostic information. During the 1950s AKI following aortography and renal angiography was reported [12] and was found to be the most common serious complication following abdominal aortography in a 1957 published survey [13]. From the mid-1950s new less toxic tri-iodinated salts of benzoic acid were gradually introduced e.g. diatrizoate, iothalamate, ioxithalamate and metrizoate [14], today classified as high osmolar CM (HOCM).

Increased contrast-nephropathy

1960-1970s: In the early 1960s it was demonstrated that the urinary tract could be visualized in patients with renal failure when IVP was performed using larger doses of the new triiodinated CM and did not seem to cause any further deterioration in renal function [15]. However, with the increasing use of large doses of CM for IVP, angiography and CT an increasing number of AKI were reported during the 1970s [16,17]. Byrd and Sherman [16] reported a mean dose of about 1.2 gram of iodine per kg among those who developed AKI corresponding to roughly 300 mL of 300 mg I/mL in an 80 kg individual. The possible pathogenetic significance of CM hyperosmolality (1500-2100 mOsm/kg H2O at 300-370 mg I/mL) was outlined and apart from large CM doses postulated risk factors were prior renal insufficiency, diabetes mellitus, advanced age and dehydration [16,17].

“Low osmolar” contrast media

1980s: Today’s "low osmolar" CM (LOCM; 520 - 900 mOsm/kg H₂O at 300-370 mg I/mL) was introduced during the first half of the 1980s [18]. Randomized studies including a meta-analysis [19,20] showed that LOCM caused a lower incidence for AKI than HOCM in patients with impaired renal function undergoing intra-arterial (IA) CM injections, mainly cardiac examinations. This supports a casual association with AKI for HOCM. The presence of diabetes in the now classical "Iohexol Cooperative Study" in coronary angiography [19] tripled and quadrupled the risk of AKI in patients with renal insufficiency when LOCM and HOCM, respectively, was used. Today diabetes is ignored as a risk factor for CI-AKI in recently revised international guidelines [1,6,21,22] with the motivation that it may also cause AKI in patients not subjected to CM. However, it seems unlikely that all cases of AKI in diabetics in the above cited study should have been solely caused by uncontrolled diabetic disease, but rather that the presence of diabetes potentiated the nephrotoxic effect of CM. In the meta-analysis by Barrett and Carlisle [20] LOCM was, however, less likely to prevent AKI after IV CM injections [20].

"Iso-osmolar" contrast media

1990s: During the mid-1990s the iso-osmolar CM (IOCM) iodixanol, iso-tonic to plasma (290 mOsm/kg H2O) at all concentrations, was marketed [18]. Meta-analyses of prospective randomized studies have shown a beneficial renal effect of iodixanol relative to LOCM in general [23-25] or only when compared with the LOCM ioxaglate or iohexol [26-32] following IA injections, most commonly coronary angiography/ interventions. In one meta-analyses no difference in the incidence of AKI was found among various LOCM [33]. Metaanalyses of IV studies have not disclosed any significant difference between IOCM and LOCM regarding renal toxicity [24,25]. The lack of difference in IV studies may partly be due to the fact that many high-risk patients such as those with unstable renal function, heart failure, hemodynamic instability, uncontrolled diabetes, recent CM examinations, etc. are often excluded in randomized CT studies [34-37].

Venous versus arterial injection

2000s: The next step in the discussion about CI-AKI was sparked by a critical literature analysis in the mid-late 2000s where it was concluded that “controlled series that support the hypothesis that IV administered CM is potentially nephrotoxic are conspicuously absent” [38] and one study demonstrated that serum creatinine (s-creatinine) in hospitalised patients not exposed to CM increased about as often as in published series of patients receiving IV CM [39]. This lead to editorials [40,41] and guidelines [42] claiming that the risk of CI-AKI is lower with IV than IA CM injections. This has been questioned simply because the vast majority of IA injections, i.e. selective arterial and infrarenal aortic, are IV relatively to the kidneys since the CM has to pass the local capillaries, draining veins and pulmonary circulation before reaching the systemic circulation and the kidneys [43], so called "second pass renal exposure" [21]. This also includes coronary arteriography and interventions where only a minor portion of the CM will reach the kidney directly from regurgitation into the aorta during each coronary injection or if a left ventriculogram is performed [43]. Subsequent studies comparing the incidence of AKI following IV and IA injections have not been able to demonstrate any significant differences [44-48]. In another retrospective study IV CM was associated with higher 30-day and overall mortality than IA CM administration after adjustment for comorbidities, CM dose, medications and total hydration [49]. One explanation for this difference might be that IV injection of the total CM dose for 30 seconds or less at CT results in a much higher injected dose rate than e.g. multiple small IA injections during a prolonged coronary procedure [43].

Supra- and juxta-renal aortic and selective renal injections as well as left ventriculograms are true IA CM exposures of the kidneys, so called "first pass renal exposure" [21] and should pose a greater risk of CI-AKI than injections that are IV relative to the kidneys. Higher plasma CM concentrations with higher nephrotoxic potential will strike the kidneys, especially if plasma hypertonic solutions are used with possible vascular endothelial injuries [50] and crenation of red blood cells [51] that may affect microcirculation.

Present Controversy

Does contrast-nephropathy exist?

2010s: The last step in the CI-AKI history now includes an increasing number of retrospective controlled observational studies comparing the incidence of AKI in patients undergoing CM-enhanced CT with controls subjected to unenhanced CT. A recent meta-analysis of 28 such studies [52] including those with PS-matching to control for background risk factors [3-5,53-55] found no association between AKI, need for renal replacement therapy or mortality. This has lead to the conclusion that iodine CM may not be the causative agent of AKI following IV CM administration [3-5]. The evidence for such a conclusion has been questioned [7] and graded as low in a systematic review [6] based on the retrospective nature of the studies, which makes selection bias a serious concern. The vast majority of controls were inpatients that may have suffered from diseases with extra high risk of AKI motivating repeated s-creatinine analysis as well as steering them into non-enhanced CT or no CT at all [4,56]. There may also exist an indication bias for CM-enhanced CT if the physician feels that the patient is at low risk of AKI. The CM cohort may have been more likely to have received intravenous hydration or other preventive measures. PS-matching including comorbidities extracted based on ICD-9 codes is not perfect and cannot account for unmeasured confounders, may not be able to discriminate severity of a disease and may have been entered at any time of hospital admission, even after AKI developed. Some of the methods used to analyze data has also been criticised including the use of relative instead of absolute glomerular filtration rate (GFR) values, insufficient attention being paid to risk stratification according to CM dose or rather CM dose/GFR ratio as well as non-renal risk factors [57,58]. Finally, no attention in the meta-analysis [52] was paid to subgroup analysis which in certain studies indicated an increased risk of AKI following CM-enhanced CT in patients with impaired renal function [53,59] or in intensive care unit patients compared with controls [60]. Thus, we are concerned regarding the recently revised European and American guidelines lowering the CI-AKI risk threshold to 30 mL/min/ 1.73 m2 and omitting non-renal risk factors based on studies with low grade evidence and since the apparently low incidence of post-CT CI-AKI [61,62] may simply be a result of adherence to previous guidelines with proper patient selection and preventive measures.

A recent study found a similar risk of AKI among patients with ST-segment-elevation myocardial infarction who underwent percutaneous coronary intervention compared with patients receiving fibrinolysis or no reperfusion and who were not exposed to CM [63].

The authors emphasized the risk of selection bias though PS-matching was used, and the results should only be viewed as exploratory and hypothesis generating.

Analysis of Individual Cases

CI-AKI is a condition generally defined as a decrease in renal function occurring within 3 days after intravascular CM administration in the absence of an alternative aetiology [42]. In the clear majority of IV CM studies on CI-AKI, controls not receiving CM are lacking [38,64]. The lack of analyzing the cause of AKI in uncontrolled CI-AKI studies and ascribing any increase in s-creatinine beyond a certain threshold (most commonly 44 μmol/L or 25%) to the effects of CM have quite rightly been criticised [38,56]. On the other hand, it has not been analyzed on an individual patient basis in the controlled studies among those receiving CM [3-5]. Thus, statistics in large cohort studies, only focusing on the group risk, may hide a true individual risk (“terror of mean”) but in today’s era of personalized medicine one must focus on the individual risk (https: //en.wikipedia.org/wiki/Personalized_medicine). Thus, we may have to start all over again with carefully analysis of possible etiological factors of AKI in each individual case to find out whether CI-AKI still exists or not. This is illustrated by two cases (Box 1 and 2) encountered in the authors’ clinical practice.

Discussion

Both illustrated patients were outpatients in stable condition at the time of the CT examination without any signs of unstable hemodynamic, renal function or other unstable background conditions that may have caused AKI. There was no sudden change in medication. The marked increase in screatinine, 806% and 238% from baseline, respectively, certainly excludes the possibility of normal s-creatinine fluctuations [65]. In our opinion this makes it difficult to neglect iodine CM as a major etiological factor. Noticeably both cases had an estimated GFR (eGFR) well above the threshold value for the risk of CI-AKI in the recently revised European [6,21,22] and American [1] guidelines, i.e. 30 mL/min/1.73 m2. In addition, this threshold does not seem to take the notorious unreliability of eGFR into consideration with 20-30% of estimates having an error exceeding 30% of measured GFR [66,67].

Chronic heart failure and diabetes mellitus have consistently been identified as risk factors for the development of AKI following IV CM injections based on multivariate analysis in uncontrolled studies [6] as well as non-steroidal antiinflammatory drugs (NSAID) in one meta-analyses [62]. However, international guidelines no longer regard them as specific for CI-AKI due to lack of rigorous testing [1,6,21,22]. Both our illustrated cases had such risk factors, but with no acute deterioration of these conditions that may have caused the AKI. In our opinion there is still no evidence against the possibility that e.g. diabetes and chronic heart failure may potentiate a nephrotoxic effect of CM as already discussed [19] and are therefore still included among as risk factors for CI-AKI in the 2017 revised guidelines of the Swedish Society of Uroradiology CM committee [68].

It has been argued that the exaggerated fear of CI-AKI and withholding CM with resultant lessening of diagnostic information may be considerable [56]. On the other hand, CM are according to our experience many times used indiscriminately at CT for convenience with the argument "not risk missing anything" and without considering e.g. the very low prevalence of disease for a certain indication. Case 2 is a typical example of such unnecessary use of contrast media when the only issue was to measure the diameter of the aorta in a screening situation.

Though the European Society Urogenital Radiology guidelines states that "there is insufficient evidence that dose is a problem with intravenous CM" [21], the CM dose should nevertheless be as low as reasonable achievable for a diagnostic study considering that CI-AKI is a toxic effect. In case 1 mapping of the central venous system may have been possible to obtain by direct CT venography following injection of diluted CM, e.g. 50 mL of 30 mg I/mL through a peripherally inserted venous cannula in the symptomatic arm. Such a concentration corresponds to about 750 HU at 120 kVp (25 HU per mg I/mL) [69] and would have resulted in a total dose of 3 instead of 45 grams of iodine. Another option to reduce the CM dose at CT is to apply low kilovoltage technique [70-72].

LOCM nephrotoxic but not IOCM?

Both patients in the present report received LOCM. It should be noted that in one PS-matched controlled study IV LOCM was found to be a nephrotoxic risk factor in patients with eGFR <30 mL/min/1.73 m2 with a trend toward significance at 30-44 mL/min/1.73 m2. Bruce et al. [59] found that the LOCM iohexol caused a higher risk of AKI after CM-enhanced CT than the IOCM iodixanol at s-creatinine levels >1.8 mg/dL (160 μmol/L) while there was no difference between iodixanol and the control group above that level. Also Tong et al. found similar incidences of AKI when using iodixanol for CT and IA cardiac catheterizations compared with patients undergoing non-enhanced CT as controls [48]. Interestingly McDonald et al. [3] used iodixanol at creatinine levels above 2.0 mg/dL (175 μmol/L) when concluded that CM may not be the causative factor of AKI. In a later controlled study [73] they found no differences in the incidence of AKI, dialysis, or mortality between PS-matched patients with the highest perceived risk of AKI undergoing iodixanol enhanced CT and non-contrast CT. They also reported that patients in their prior studies [3,4,55,74], with predominantly LOCM, had fewer illnesses, fewer CKD and ICU patients, and were therefore at lower risk of developing CIN than patients in the iodixanol cohort. This may then explain why no difference in the rate of AKI after LOCM compared with controls was found in the studies by McDonald et al.

In an ischemic kidney model in pig’s renal artery injections of CM with the highest osmolality and lowest viscosity caused hemorrhagic congestions, necrosis and markedly decreased renal function while the iso-osmolar CM with the highest viscosity (iodixanol) affected renal function no different than saline [75-77]. These results and the fact that porcine kidneys are more like human kidneys than kidneys of most other species [78] also contradicts the view that the high viscosity of IOCM should be a significant pathophysiological factor in CIAKI [79].

Thus, instead of concluding that CI-AKI following CT may not occur one may as well conclude from these studies that it may occur when using LOCM but may not occur when using iodixanol and adequate prophylaxis. Reasons for the lack of difference in the incidence of AKI between iodixanol and LOCM in meta-analyses of randomized CT studies have already been discussed.

Future Studies

How should we then move forward to obtain better evidence regarding whether CI-AKI exists or not? As indicated by the present case examples one option would be to prospectively perform post examination creatinine measurements in patients with certain defined risk factors and then carefully analyze individual cases of AKI to find out if CM is the only reasonably cause of AKI.

Instead of using patients undergoing non-enhanced CT as controls with the inherited risk of selection bias, a second option has been proposed based on the ratio between the gram-iodine CM dose and GFR [57], a crucial measure regarding toxicity of drugs excreted by glomerular filtration like CM [80,81]. The analysis may then be restricted to the CMenhanced CT group and to propensity match patients with different dose/GFR ratios at various GFR stages [57]. Such PSmatching most likely reduces the problem of selection bias, since all included patients have received CM but at different CM dose/GFR ratios.

A prospective controlled trial is a third option, though randomization of patients to receive either CM or saline would in general be unethical for obvious reasons; CM may be crucial for adequate diagnosis, control patients will be subjected to two examinations with a week interval, one without and one with contrast medium with doubling the radiation exposure, and double blinding would be impossible [82]. However, considering the magnitude of importance of the CI-AKI issue in diagnostic radiology there might be at least one group of patients where a prospective controlled trial might be ethically acceptable, e.g. elderly (60 years) patients with malignant diseases undergoing regular surveillance with CM-enhanced CT. In many instances both non-enhanced and CM-enhanced CT is performed in this patient group. By performing the two phases with a week interval the patient would also be its own control. This group may also be of special interest since they may be on nephrotoxic chemotherapeutics and the rate of AKI has been correlated significantly with IV CM [83] and appear higher in patients with recent chemotherapy [84].

Conclusion

Recent PS-matched retrospective controlled observational studies questioning the existence of CM nephrotoxicity following CM-enhanced CT suffer from several methodological concerns. Lowering the GFR threshold of CM-nephrotoxicity to 30 mL/min/1.73 m2 in international guidelines seems premature considering the low level of evidence of such studies and the unreliability of estimated GFR. Our two patient reports indicate that CM is an etiological factor to AKI following CM-enhanced CT. Optional approaches to study the possible existence of CI-AKI may include careful prospective analysis of the aetiology of PC-AKI in individual cases, restricting PS-matching to patients receiving CM with various CM dose/GFR ratios at different GFR levels or performing a prospective controlled study in patients with malignant disease scheduled for routine tumour surveillance without and with CM-enhanced CT, but with a week interval.

Box 1

Case 1: A 60-year old overweight woman with diabetes mellitus, hypertension and macroproteinuria (u-albumin/ucreatinine ratio 62 g/mol) was admitted to the emergency department because of a swollen arm. She was on antidiabetic medication, loop diuretic, amlodipine and ibuprofen (800 mg x 2) but none was initiated recently. There was no treatment with renin-angiotensin system (RAS) inhibitors. Her general condition was stable with no breathing difficulties, chest pain, nausea, vomiting or hemodynamically instability. A deep venous thrombosis was suspected. S-creatinine at admission was 109 μmol/L with an estimated GFR (eGFR) of 45 mL/min (45 mL/min/1.73 m2). Ultrasonography was inconclusive why phlebography was performed with a LOCM and a total dose of 90 mL 300 mg I/mL. No thrombosis was diagnosed. The patient left the hospital in stable condition and was informed to discontinue her metformin and ibuprofen medication. Screatinine prior to re-starting metformin two days later had now increased to 439 μmol/L. She had also developed nausea and vomiting, noticed decreased urine output and was subsequently hospitalized. S-creatinine peaked at 987 μmol/L at day seven after which renal function started to improve with no need for haemodialysis. She was discharged after fully two weeks with a s-creatinine of 139 μmol/L. Two months later it was 121 μmol/L. The hospital cost was estimated to about 7000 Euro.

Box 2

Case 2: A 65-year old man underwent ultrasonography as part of a screening program for abdominal aortic aneurysms. The examination was inconclusive (BMI 39 kg/cm2). Instead CT of the aorta was requested. Based on a s-creatinine of 122 μmol/L three days prior to CT, GFR was estimated to 63 mL/min (48 mL/min/1.73 m2). A CM-enhanced CT was performed using 93 mL 350 mg I/mL of a LOCM. Later the same evening the patient experienced dark coloured urine with decreasing amounts. During the following days he became increasingly tired and contacted the renal ward day five. S-creatinine was now 412 μmol/L and he was admitted. He had had no signs of acute heart failure or hemodynamic instability. There were no signs of any pulmonary oedema at the CT examination and no chest x-ray was performed before or during hospitalization query cardiac failure/pulmonary oedema. He was discharged a week later with a s-creatinine of 128 μmol/L and it remained stable at one month, 140 μmol/L. The patient had type 2 diabetes mellitus with multiple complications, chronic kidney disease, restrictive cardiomyopathy and chronic heart failure with fluid restriction, none of which was mentioned on the CT request or checked by the radiology department. He was on treatment with insulin, loop-diuretic, clindamycin, paracetamol and codeine, none of which had been recently started. He had no treatment with RAS-inhibitors. No hospital cost was calculated.

Competing Interests

U.N. and P.A has received speaker’s fees from GE Healthcare AB, Danderyd, Sweden.

References

1. ACR (2017) American College of Radiology.(2017) ACR Manual on contrast media. Version 10.3.
2. Nash K, Hafeez A, Hou S(2002) Hospital-acquired renal insufficiency. Am J Kidney Dis 39:930-936.
3. McDonald RJ, McDonald JS, Bida JP, Carter RE, Fleming CJ, et al. (2013) Intravenous contrast material-induced nephropathy: causal or coincident phenomenon? Radiology 267:106-118.
4. McDonald JS, McDonald RJ, Carter RE, Katzberg RW, Kallmes DF, et al. (2014) Risk of intravenous contrast material-mediated acute kidney injury: a propensity score-matched study stratified by baseline-estimated glomerular filtration rate. Radiology 271:65-73.
5. Hinson JS, Ehmann MR, Fine DM, Fishman EK, Toerper MF, et al. (2017) Risk of Acute Kidney Injury After Intravenous Contrast Media Administration. Ann Emerg Med 69:577-586.
6. Dutch (2017) Radiological Society of the Netherlands. Guideline safe use of contrast media - Part 1.
7. Weisbord SD, du Cheryon D (2018) Contrast-associated acute kidney injury is a myth: No. Intensive Care Med 44:107-109.
8. Swick M (1929) Darstellung der Niere und Harnwege in Röntgenbild durch intravenöse Einbringung eines neuen Kontraststoffes: des Uroselectans. Klinische Wochenschrift 8:2087-2089.
9. Grainger RG (1982) Intravascular contrast media. British Journal of Radiology 55:251.
10. Pendergrass EP, Chamberlin GW, Godfrey EW, Burdick ED (1942) A survey of deaths and unfavorable suquelae following the administration of contrast media. Am J Roentgenol  48:741-762.
11.       Bartels ED, Brun GC, Gammeltoft A, Gjorup PA (1954) Acute anuria following intravenous pyelography in a patient with myelomatosis. Acta Med Scand 150:297-302.
12. Alwall N, Johnsson S, Tornberg A, Werko L (1955) Acute renal failure following angiography especially the risk of repeated examination, revealed by eight cases (two deaths). Acta Chir Scand 109:11-19.
13. McAfee JG (1957) A survey of complications of abdominal aortography. Radiology 68:825-838.
14. Hoppe JO (1959) Some pharmacological aspects of radiopaque compounds. Ann N Y Acad Sci 78:727-739.
15. 15. Schwartz WB, Hurwit A, Ettinger A (1963) Intravenous urography in the patient with renal insufficiency. N Engl J Med  269:277-283.
16. Byrd L, Sherman RL (1979) Radiocontrast-induced acute renal failure: a clinical and pathophysiologic review. Medicine (Baltimore) 58:270-279.
17. Mudge GH (1980) Nephrotoxicity of urographic radiocontrast drugs. Kidney Int  18:540-552.
18. Nyman U, Ekberg O, Aspelin P (2016) Torsten Almen (1931-2016): the father of non-ionic iodine contrast media. Acta Radiol 57:1072-1078.
19. Rudnick MR, Goldfarb S, Wexler L, Ludbrook PA, Murphy MJ,  et al. (1995)Nephrotoxicity of ionic and nonionic contrast media in 1196 patients: a randomized trial. The Iohexol Cooperative Study. Kidney Int  47:254-261.
20. Barrett BJ, Carlisle EJ (1993) Metaanalysis of the relative nephrotoxicity of high- and low-osmolality iodinated contrast media. Radiology 188:171-178.
21. van der Molen AJ, Reimer P, Dekkers IA, Bongartz G, Bellin MF, et al. (2018) Post-contrast acute kidney injury - Part 1: Definition, clinical features, incidence, role of contrast medium and risk factors : Recommendations for updated ESUR Contrast Medium Safety Committee guidelines. Eur Radiol 
22. van der Molen AJ, Reimer P, Dekkers IA, Bongartz G, Bellin MF,  et al. (2018) Post-contrast acute kidney injury. Part 2: risk stratification, role of hydration and other prophylactic measures, patients taking metformin and chronic dialysis patients : Recommendations for updated ESUR Contrast Medium Safety Committee guidelines. Eur Radiol .
23. McCullough PA, Bertrand ME, Brinker JA, Stacul F (2006) A meta-analysis of the renal safety of isosmolar iodixanol compared with low-osmolar contrast media. J Am Coll Cardiol 48:692-699.
24. McCullough PA, Brown JR (2011) Effects of Intra-Arterial and Intravenous Iso-Osmolar Contrast Medium (Iodixanol) on the Risk of Contrast-Induced Acute Kidney Injury: A Meta-Analysis. Cardiorenal Med  1:220-234.
25. Dong M, Jiao Z, Liu T, Guo F, Li G et.al. (2012) Effect of administration route on the renal safety of contrast agents: a meta-analysis of randomized controlled trials. J Nephrol 25:290-301.
26. Sharma SK, Kini A (2005) Effect of nonionic radiocontrast agents on the occurrence of contrast-induced nephropathy in patients with mild-moderate chronic renal insufficiency: pooled analysis of the randomized trials. Catheter Cardiovasc Interv 65:386-393.
27. Solomon R (2005) The role of osmolality in the incidence of contrast-induced nephropathy: a systematic review of angiographic contrast media in high risk patients. Kidney Int 68:2256-2263.
28. Heinrich MC, Haberle L, Muller V, Bautz W, Uder M (2009) Nephrotoxicity of iso-osmolar iodixanol compared with nonionic low-osmolar contrast media: meta-analysis of randomized controlled trials. Radiology 250:68-86.
29. Reed M, Meier P, Tamhane UU, Welch KB, Moscucci M, et.al. (2009)  The relative renal safety of iodixanol compared with low-osmolar contrast media: a meta-analysis of randomized controlled trials. JACC Cardiovasc Interv 2:645-654.
30. From AM, Al Badarin FJ, McDonald FS, Bartholmai BJ, Cha SS, et.al. (2010) Iodixanol versus low-osmolar contrast media for prevention of contrast induced nephropathy: meta-analysis of randomized, controlled trials. Circ Cardiovasc Interv 3:351-358.
31. Biondi-Zoccai G, Lotrionte M, Thomsen HS, Romagnoli E, D'Ascenzo F, et al. (2014) Nephropathy after administration of iso-osmolar and low-osmolar contrast media: evidence from a network meta-analysis. Int J Cardiol 172:375-380.
32. Pandya B, Chalhoub JM, Parikh V, Gaddam S, Spagnola J, et al. (2017) Contrast media use in patients with chronic kidney disease undergoing coronary angiography: A systematic review and meta-analysis of randomized trials. Int J Cardiol 228:137-144.
33. Eng J, Wilson RF, Subramaniam RM, Zhang A, Suarez-Cuervo C, et al.  (2016) Comparative Effect of Contrast Media Type on the Incidence of Contrast-Induced Nephropathy: A Systematic Review and Meta-analysis. Ann Intern Med  164:417-424.
34. Barrett BJ, Katzberg RW, Thomsen HS, Chen N, Sahani D, et al. (2006) Contrast-induced nephropathy in patients with chronic kidney disease undergoing computed tomography: a double-blind comparison of iodixanol and iopamidol. Invest Radiol 41:815-281.
35. Kuhn MJ, Chen N, Sahani DV, Reimer D, van Beek EJ, et al. (2008) The PREDICT study: a randomized double-blind comparison of contrast-induced nephropathy after low- or isoosmolar contrast agent exposure. AJR Am J Roentgenol  191:151-157.
36. Nguyen SA, Suranyi P, Ravenel JG, Randall PK, Romano PB, et al. (2008)  Iso-osmolality versus low-osmolality iodinated contrast medium at intravenous contrast-enhanced CT: effect on kidney function. Radiology 248:97-105.
37. Thomsen HS, Morcos SK, Erley CM, Grazioli L, Bonomo L, et al. (2008) The ACTIVE Trial: Comparison on the effects on renal function of iomeprol-400 and iodixanol-320 in patients with chronic kidney disease undergoing abdominal computed tomography. Invest Radiol ;43:170-178.
38. Rao QA, Newhouse JH (2006)  Risk of nephropathy after intravenous administration of contrast material: a critical literature analysis. Radiology 239:392-397.
39. Newhouse JH, Kho D, Rao QA, Starren J (2008) Frequency of serum creatinine changes in the absence of iodinated contrast material: implications for studies of contrast nephrotoxicity. AJR Am J Roentgenol 191:376-382.
40. Katzberg RW, Barrett BJ (2007) Risk of iodinated contrast material--induced nephropathy with intravenous administration. Radiology 243:622-628.
41. Katzberg RW, Newhouse JH (2010) Intravenous contrast medium-induced nephrotoxicity: is the medical risk really as great as we have come to believe? Radiology 256:21-28.
42. Stacul F, van der Molen AJ, Reimer P, Webb JA, Thomsen HS, et al. (2011) Contrast induced nephropathy: updated ESUR Contrast Media Safety Committee guidelines. Eur Radiol 21:2527-2541.
43. Nyman U, Almén T, Jacobsson B, Aspelin P (2012)  Are intravenous injections of contrast media really less nephrotoxic than intra-arterial injections? Eur Radiol 22:1366-1371.
44. Karlsberg RP, Dohad SY, Sheng R (2011)  Contrast medium-induced acute kidney injury: comparison of intravenous and intraarterial administration of iodinated contrast medium. J Vasc Interv Radiol 22:1159-1165.
45. Kooiman J, Le Haen PA, Gezgin G, de Vries JP, Boersma D, et al. (2013) Contrast-induced acute kidney injury and clinical outcomes after intra-arterial and intravenous contrast administration: risk comparison adjusted for patient characteristics by design. Am Heart J 165:793-799, 9 e1.
46. McDonald JS, Leake CB, McDonald RJ, Gulati R, Katzberg RW, et al. (2016) Acute Kidney Injury After Intravenous Versus Intra-Arterial Contrast Material Administration in a Paired Cohort. Invest Radiol  51:804-1809.
47. Nijssen EC, Rennenberg RJ, Nelemans PJ, Essers BA, Janssen MM, et al. (2017) Prophylactic hydration to protect renal function from intravascular iodinated contrast material in patients at high risk of contrast-induced nephropathy (AMACING): a prospective, randomised, phase 3, controlled, open-label, non-inferiority trial. Lancet 389:1312-1322.
48. Tong GE, Kumar S, Chong KC, Shah N, Wong MJ, et al. (2016) Risk of contrast-induced nephropathy for patients receiving intravenous vs. intra-arterial iodixanol administration. Abdom Radiol (NY) 41:91-99.
49. From AM, Bartholmai BJ, Williams AW, Cha SS, McDonald FS (2008)  Mortality associated with nephropathy after radiographic contrast exposure. Mayo Clin Proc  83:1095-1100.
50. Nyman U, Almen T (1980) Effects of contrast media on aortic endothelium. Experiments in the rat with non-ionic monomeric and monoacidic dimeric contrast media. Acta Radiol Suppl 362:65-71.
51. Aspelin P, Stacul F, Thomsen HS, Morcos SK, van der Molen AJ (2006) Effects of iodinated contrast media on blood and endothelium. Eur Radiol 16:1041-1049.
52. Aycock RD, Westafer LM, Boxen JL, Majlesi N, Schoenfeld EM, et.al. (2018)  Acute Kidney Injury After Computed Tomography: A Meta-analysis. Ann Emerg Med  71:44-53 e4.
53. Davenport MS, Khalatbari S, Cohan RH, Dillman JR, Myles JD, et.al. (2013) Contrast material-induced nephrotoxicity and intravenous low-osmolality iodinated contrast material: risk stratification by using estimated glomerular filtration rate. Radiology 268:719-728.
54. Davenport MS, Khalatbari S, Dillman JR, Cohan RH, Caoili EM, et.al. (2013) Contrast material-induced nephrotoxicity and intravenous low-osmolality iodinated contrast material. Radiology 267:94-105.
55. McDonald RJ, McDonald JS, Carter RE, Hartman RP, Katzberg RW, et al. (2014)  Intravenous contrast material exposure is not an independent risk factor for dialysis or mortality. Radiology 273:714-725.
56. Newhouse JH, RoyChoudhury A (2013)  Quantitating contrast medium-induced nephropathy: controlling the controls. Radiology 267:4-8.
57. Nyman U, Aspelin P, Jakobsen J, Bjork J (2015) Controversies in Contrast Material-induced Acute Kidney Injury: Propensity Score Matching of Patients with Different Dose/Absolute Glomerular Filtration Rate Ratios. Radiology 277:633-637.
58. Nyman U, Aspelin P, Jakobsen J, Bjork J (2016) Some Clarifying Points Regarding Controversies in Contrast Material-induced Acute Kidney Injury. Radiology 279:982-984.
59. Bruce RJ, Djamali A, Shinki K, Michel SJ, Fine JP, et.al. (2009)  Background fluctuation of kidney function versus contrast-induced nephrotoxicity. AJR Am J Roentgenol 192:711-718.
60. McDonald JS, McDonald RJ, Williamson EE, Kallmes DF, Kashani K (2017) Post-contrast acute kidney injury in intensive care unit patients: a propensity score-adjusted study. Intensive Care Med 43:774-784.
61. Kooiman J, Pasha SM, Zondag W, Sijpkens YW, van der MJ, et al. (2012)  Meta-analysis: serum creatinine changes following contrast enhanced CT imaging. Eur J Radiol 81:2554-2561.
62. Moos SI, van Vemde DN, Stoker J, Bipat S (2013)  Contrast induced nephropathy in patients undergoing intravenous (IV) contrast enhanced computed tomography (CECT) and the relationship with risk factors: a meta-analysis. Eur J Radiol 82:e387-399.
63. Caspi O, Habib M, Cohen Y, Kerner A, Roguin A, et al. (2017) Acute Kidney Injury After Primary Angioplasty: Is Contrast-Induced Nephropathy the Culprit? J Am Heart .
64. McDonald JS, McDonald RJ, Comin J, Williamson EE, Katzberg RW, et al. (2013) Frequency of acute kidney injury following intravenous contrast medium administration: a systematic review and meta-analysis. Radiology 267:119-128.
65. Thomsen HS, Stacul F (2014) CIN: can we forget it? Acta Radiol  55:1027-1030.
66. Nyman U, Björk J, Bäck SE, Sterner G, Grubb A (2016) Estimating GFR prior to contrast medium examinations - what the radiologist needs to know! Euro Radiol 26:425-435.
67. Björk J, Grubb A, Sterner G, Bäck SE, Nyman U (2017) Accuracy diagrams: a novel way to illustrate uncertainty of estimated GFR. Scand J Clin Lab Invest 77:199-204.
68. SURF (2017)  Swedish Society of Uroradiology. National guidelines iodine contrast media [Svensk uroradiologisk förenings kontrastmedelsgrupp. Rekommendationer jodkontrastmedel] v. 6.0 .
69. Nyman U, Elmståhl B, Geijer H, Leander P, Almén T,et.al. (2011) Iodine contrast iso-attenuating with diagnostic gadolinium doses in CTA and angiography results in ultra-low iodine doses. A way to avoid both CIN and NSF in azotemic patients? Eur Radiol 21:326-336.
70. Kristiansson M, Holmquist F, Nyman U (2010) Ultralow contrast medium doses at CT to diagnose pulmonary embolism in patients with moderate to severe renal impairment. A feasibility study. Eur Radiol 20:1321-1330.
71. Nakaura T, Awai K, Maruyama N, Takata N, Yoshinaka I,  et al. (2011) Abdominal dynamic CT in patients with renal dysfunction: contrast agent dose reduction with low tube voltage and high tube current-time product settings at 256-detector row CT. Radiology 261:467-476.
72. Taguchi N, Oda S, Utsunomiya D, Funama Y, Nakaura T, et al. (2017) Using 80 kVp on a 320-row scanner for hepatic multiphasic CT reduces the contrast dose by 50 % in patients at risk for contrast-induced nephropathy. Eur Radiol 27:812-820.
73. McDonald JS, McDonald RJ, Williamson EE, Kallmes DF (2017) Is Intravenous Administration of Iodixanol Associated with Increased Risk of Acute Kidney Injury, Dialysis, or Mortality? A Propensity Score-adjusted Study. Radiology 285:414-424.
74. McDonald JS, McDonald RJ, Lieske JC, Carter RE, Katzberg RW, et al. (2015) Risk of Acute Kidney Injury, Dialysis, and Mortality in Patients With Chronic Kidney Disease After Intravenous Contrast Material Exposure. Mayo Clin Proc 90:1046-1053.
75. Elmståhl B, Nyman U, Leander P, Chai CM, Golman K, et al. (2006) Gadolinium contrast media are more nephrotoxic than iodine media. The importance of osmolality in direct renal artery injections. Eur Radiol 16:2712-2720.
76. Elmståhl B, Leander P, Grant D, Doughty RW, Chai CM, et al. (2007) Histomorphological Changes after Renal X-Ray Arteriography Using Iodine and Gadolinium Contrast Media in an Ischemic Porcine Model. Acta Radiol :1-11.
77. Elmståhl B, Nyman U, Leander P, Golman K, Chai CM, et al. (2008) Iodixanol 320 results in better renal tolerance and radiodensity than do gadolinium-based contrast media: arteriography in ischemic porcine kidneys. Radiology 247:88-97.
78. Elmståhl B, Nyman U, Leander P, Chai CM, Frennby B, et al. (2004) Gadolinium contrast media are more nephrotoxic than a low osmolar iodine medium employing doses with equal X-ray attenuation in renal arteriography: an experimental study in pigs. Acad Radiol 11:1219-1228.
79. Persson PB, Hansell P, Liss P (2005) Pathophysiology of contrast medium-induced nephropathy. Kidney Int 68:14-22.
80. Nyman U, Björk J, Aspelin P, Marenzi G (2008) Contrast medium dose-to-GFR ratio: A measure of systemic exposure to predict contrast-induced nephropathy after percutaneous coronary intervention. Acta Radiol 49:658-667.
81. Gurm HS, Dixon SR, Smith DE, Share D, LaLonde T, et al. (2011) Renal function-based contrast dosing to define safe limits of radiographic contrast media in patients undergoing percutaneous coronary interventions. J Am Coll Cardiol 58:907-914.
82. Thomsen HS, Webb JA (2016) Control groups in studies of contrast media adverse events. Acta Radiol 57:903-905.
83. Salahudeen AK, Doshi SM, Pawar T, Nowshad G, Lahoti A, (2013) Incidence rate, clinical correlates, and outcomes of AKI in patients admitted to a comprehensive cancer center. Clin J Am Soc Nephrol 8:347-354.
84. Cicin I, Erdogan B, Gulsen E, Uzunoglu S, Sut N, et al. (2014) Incidence of contrast-induced nephropathy in hospitalised patients with cancer. Eur Radiol 24:184-190.