- Email: [email protected]
Overestimating the Risk of Intravenous Contrast Medium-Induced Nephropathy: A Pitfall in Imaging the Genitourinary System
Author's Accepted Manuscript
Overestimating the Risk of Intravenous Contrast Medium-Induced Nephropathy: A Pitfall in Imaging the Genitourinary System Lyndon Luk M.D., Jeffrey H. Newhouse M.D.
www.elsevier.com/locate/enganabound
PII: DOI: Reference:
S0037-198X(16)00004-3 http://dx.doi.org/10.1053/j.ro.2016.01.003 YSROE50535
To appear in:
Seminar in Roentgenology
Cite this article as: Lyndon Luk M.D., Jeffrey H. Newhouse M.D., Overestimating the Risk of Intravenous Contrast Medium-Induced Nephropathy: A Pitfall in Imaging the Genitourinary System, Seminar in Roentgenology, http://dx.doi.org/10.1053/j. ro.2016.01.003 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Overestimating the Risk of Intravenous Contrast Medium-Induced Nephropathy: A Pitfall in Imaging the Genitourinary System
Lyndon Luk1, M.D. Jeffrey H. Newhouse1, M.D. 1
Department of Radiology, Columbia University Medical Center, New York, New York
Address for Correspondence
Lyndon Luk, MD Department of Radiology Columbia University Medical Center 622 West 168th Street, PB-1-301 New York, NY 10032 Tel: 201-562-2739 Email: [email protected]
Abstract The overestimation of overall risk and consequences of intravenous contrast media-induced nephropathy has perpetuated an unwarranted avoidance of contrast-enhanced studies with consequent reduction in effectiveness of management of a number of genitourinary diseases. Properly assessing the risk of contrast nephropathy and avoiding this common imaging pitfall is critical in avoiding or misdiagnosing otherwise treatable genitourinary pathology.
History Iodinated contrast media (CM) were first discovered and used in diagnostic radiology nearly a century ago [1], leading to the first clinical pyelogram performed by Osborne, Southerland, Scholl and Roundtree in 1923 [2] and the first femoral arteriogram by Berberich and Hersch in 1924 [3]. Not long after, Wallingford, a chemist, created the first iodine-containing benzoic acid ring. [4]. The contributions of Wallingford, Swick [5] and Hoppe et al. [6] in the 1950s led to markedly improved contrast opacification as well as patient tolerance.
By the 1970s, annual consumption of contrast media surpassed 2,000 metric tons, fueled by the discovery of computer tomography (CT) [7]. At first, there was little thought given to the idea that iodinated CM were potentially nephrotoxic. Multiple publications documenting this phenomenon were published starting in the 1960s; Manitz and Matthes [8] and Ansell [9] documented transient anuria in renal failure patients after intravenous pyelography. The concept of acute renal dysfunction caused by intravenous CM has since become fundamental
and axiomatic in both practice and literature of modern medicine [10], with articles numbering in the thousands. Definition of Contrast-induced nephropathy (CIN) Contrast-induced nephropathy (CIN) is typically defined as an absolute or percentage increase in serum creatinine (SCr) level over baseline; 0.5 mg/dl is the commonest threshold for an absolute rise; 25%, 50% and 100% have all been used in published series dealing with CIN. With these thresholds, the risk of CIN in a cohort of patients receiving contrast ostensibly determined by clinical series has ranged from 0% to nearly 50%, and the decades-long flow of publications purporting to document CIN has led to a widespread conviction that the risk is considerable, both among practitioners who refer patients for various contrast-requiring examinations and among radiologists and interventional cardiologists. One of the most widely-cited quotes from papers dealing with CIN risk [11] states that contrast is the third-most-common cause of acute kidney injury (AKI) in hospitalized patients. Although the commonest course of CIN consists of a transient rise in serum creatinine [12], it has been claimed that in some cases renal function does not return to baseline, and may even require chronic dialysis [13]. It has also been found that patients who experience CIN are at risk for longer hospital stays than patients without it, and even run a higher risk of death [14]. Given these concerns, radiology departments and practices usually have established policies which preclude administering intravenous contrast in patients deemed to be at risk. Since the risk is felt to rise in proportion to degrees of chronic renal failure [15], thresholds of serum creatinine or estimated glomerular filtration rate (eGFR) are usually used to identify patients for whom contrast administration is precluded. Patients may also have contrast withheld if serum
creatinine is below the threshold but is rising, despite lack of rigorous evidence that contrast is deleterious in this circumstance. Radiology practices and departments may establish policies requiring a recent creatinine or eGFR to be measured even for patients whose risk of renal disease is very small. Many also require that informed consent be formally obtained and documented [16, 17]. All of these practices (not to mention fear of litigation) have led to great reluctance to administer contrast to patients whose management requires information only available from its use. Discussion Our thesis is that the perception of risk of nephropathy from intravenous CM throughout the medical community is much higher than the real risk warrants. There are two major sources of this misconception: the overestimation of risk in nearly all publications due to the serious error of omitting controls, and the conflation of the risk of procedures requiring intravenous contrast administration with the administration of contrast during cardiac angiography. We will deal with each of these in order.
From the initial recognition that contrast may cause renal dysfunction, there has been an increasing rate of publications regarding the risk of CIN; literally hundreds of clinical series on the topics have appeared. The experiments detailed within these publications were performed with varying degrees of scientific rigor, but nearly all assumed that any renal dysfunction which was found after administration of contrast was caused by the contrast; control series of patients to estimate the incidence of renal dysfunction without contrast were, for decades, never provided. Studies performed by Cramer et al. in 1985 and Heller et al. in 1991 did include
such controls [18, 19]. Each found rates of post-contrast nephropathy within the ranges reported in previous publications; however, each also found that these rates did not exceed those assessed in their control groups, and concluded that, at least in the circumstances they reported, contrast-induced changes might not, in fact, occur.
Despite their potential importance, these two publications were rarely cited in subsequent reported experiments. In 2006, they were stressed in a critical review of the literature [20]. This review was shortly followed in 2009 by an experiment in which the incidence of short-term creatinine increases were assessed in a very large group of patients who had had no contrast [21]; these figures were strikingly similar to the rates of post-contrast nephropathy found in all of the previous literature involving intravenous contrast administration, further calling into question the validity of these experiments.
In the few years following this series, several additional clinical series appeared which did include control groups [22-26]. All but one found no excess cases of nephropathy after contrast beyond those which appeared in control patients; the one which did reported results from two contrast agents, with no difference from control groups found for one agent and a slight increase in nephropathy rates for the other [27].
Subsequently, these controlled studies were criticized since control patients were those receiving non-contrast CT scans, and the data from the two groups were compared retrospectively. Since many of the control patients were steered to non-contrast scans because
their referring physicians felt their renal function to be particularly at risk, a selection bias could well have arisen; a real tendency for contrast to cause renal dysfunction might have been masked by the control patients’ excess tendency to experience renal dysfunction for noncontrast-related reasons.
Multiple articles addressed possible selection bias by performing 1:1 propensity matching and propensity score analysis [15, 28, 29], a statistical technique intended to reduce the effects of differences between experimental and control groups in retrospective studies by evaluating variables that may predict either increased or decreased likelihood of receiving a particular treatment. One study identified low-contrast contrast material (LOCM) as a risk for CIN in patients with GFR less than 30 ml/min/1.73 m2 [15] after propensity score adjustment. Other studies found no significant difference in AKI risk between patients undergoing non-contrast or contrast studies in any risk subgroup after propensity score adjustment [28, 29] and identified AKI risk as independent of CM exposure, even in patients with eGFR less than 30 ml/min/1.73 m2 [30].
Angiocardiography has long been recognized as a procedure that can lead to AKI [31, 32]. In the past two decades, most of the literature regarding CIN reports studies involves angiocardiography, and much of the current consensus regarding increased morbidity and mortality of CIN has arisen from these publications. As an example, a study by Gruberg et al. of patients undergoing percutaneous coronary intervention with baseline creatinine ≥1.8 mg/dL found a 37.7% rate of CIN, a 7.1% rate of CIN requiring hemodialysis and a 22.6% mortality rate
in patients requiring dialysis [33]. In the discussion sessions of these papers, and those dealing with intravenous contrast, CIN risk tends to be considered as a single entity, with little attention given to the differences in risk between the two types of procedures. Conflation of these study results has led to a serious overestimate of risk of intravenous contrast, as detailed by Katzberg and Newhouse in a detailed literature review published in 2010 [34].
The incidence of CIN with IV contrast media has been overstated not only because of extrapolation of angiocardiography experience despite the literature detailing the significantly safer profile of IV CM in comparison to contrast-enhanced cardiac studies which dates back to as early as 1979 [35]. In 1992, Moore et al. found a greater than two-fold increase in the rate of nephrotoxicity in patients undergoing angiocardiography versus those undergoing CE CT [36]. Review of more contemporary prospective studies investigating the use of LOCM and isoosmolar CM show an overall CIN rate of approximately 5.4 % [15, 37-43], including a post IV CM CIN rate of 5.2% in patients with renal insufficiency and diabetes mellitus [41]. In comparison, overall CIN rate in patients with chronic kidney disease and diabetes as depicted by the cardiology literature are upwards of 33%, noted by Rudnick et al. in the Iohexol Cooperative Study [44].
The overall difference in the rate of morbidity and mortality ostensibly caused by CIN in patients receiving IV CM and those receiving intraarterial CM during cardiac angiography and intervention is even more pronounced. The same CIN literature review in 2006 evaluating CE CT in patients with renal insufficiency [20] found no documented cases of CIN requiring dialysis or
death out of a total of 1,175 subjects. A 2015 study by McDonald evaluating rates of AKI, emergent dialysis and mortality in a large, propensity matched cohort with stage III-V CKD revealed no significant differences in morbidity or mortality between the noncontrast and contrast groups [45]. A recent systematic review and meta-analysis of over 25,000 patients demonstrated similar rates of AKI, dialysis and death between CE and control groups [46]
There is no doubt that nephropathy can be a serious condition; after all, it constitutes failure of an important organ system, which may in turn have effects on other systems. And nephropathy may not only be a primary event: it may be the result of, and act as a marker for, failure of other organs. These events would be expected to lengthen hospital stays, occasionally require dialysis and even increase mortality rates, but if they occur for reasons other than intravenous contrast, their temporal association with contrast administration does not imply causation.
In sum, recent investigations show that CIN, and any serious consequence of it, is much less of a risk than most physicians have believed for decades. The debate about its exact incidence may continue, but it is now evident that it is low. Use of contrast, therefore, must be reevaluated since the added value of CM in the cross-sectional imaging evaluation of numerous diseases affecting the genitourinary (GU) system is irrefutable.
The American College of Radiology (ACR) Appropriateness Criteria can serve as a good indicator of the fraction of cases dealing with GU issues which do or do not need contrast enhancement
for CT. In the urologic, womens' imaging and vascular sections of these criteria there are 43 variants of GU conditions for which CT is or may be indicated [47]. Among these, in 38 variants, CT with contrast is rated more appropriate than CT without contrast; in only 5 is CT without contrast more appropriate. Some conditions where contrast administration may either clinch or clarify the diagnosis include acute flank pain which may be due to pyelonephritis (figure 1) or acute renal ischemia (figure 2), identifying vasculitis, arteriovenous fistulae and venous thrombosis, demonstrating active bleeding and pseudo-aneurysms due to trauma, diagnosing enhancing urothelial tumors, screening for renal tumors in syndromes which may include them and assessing endometrial morphology.
It should be immediately obvious from the above that in the conditions which may be missed or mischaracterized if contrast is withheld, appropriate patient management can be severely impeded in a number of ways. And the problem is wider: a great deal of GU disease is discovered incidentally on CT scans initially performed for other problems. If contrast is unnecessarily withheld from abdominal CT performed for any reason, significant errors in GU diagnoses may ensue. Detecting renal cell carcinoma in patients in whom it is not suspected is only one example. As abdominal CT has become more frequently performed, an increasing fraction of these tumors are discovered serendipitously, and more often at a stage which permits cure [48-51]. And although rigorous demonstration that contrast is necessary for this diagnosis is lacking, any experienced abdominal radiologist would agree that small intrarenal masses are easy to miss on an examination performed without contrast. Furthermore, nearly all ACR Appropriateness Criteria that include CT as a possible imaging modality in addressing
various gastrointestinal issues advocate CM; of the 61 listed variants of gastrointestinal conditions with ACR imaging recommendations that include CT, all but 2 rate CT with contrast as more appropriate than CT without contrast.
The frequency with which contrast is unnecessarily withheld in CT examinations is not known, but the authors' experiences suggest that the problem occurs with regularity. Factors incorrectly felt to confer risk of CIN include mild chronic renal failure, diabetes without renal failure, congenital or surgical absence of one kidney with normal creatinine, transplanted kidneys and recent administration of contrast. (Mild prior systemic contrast reactions are also often cited as reasons to withhold contrast, but these circumstances are beyond the scope of this paper.) Discussions between radiologists and referring physicians intended to compare the risks of CIN with the risks of missing diagnoses which require contrast, are also frequently foregone at a rate which is also not known, but probably considerable.
We conclude, therefore, that many essential CT examinations requiring contrast are withheld due to an unwarranted fear of nephropathy and its associated risks with consequent reduction in effectiveness of management of a number of GU and other diseases. This practice constitutes a sufficient threat to health that serious attempts to correct it are warranted. Within radiology departments, policy constraints for intravenous contrast administration should be reviewed and liberalized where possible, and policies which strictly forbid contrast administration should be replaced by requirements for realistic comparisons of the risk of nephropathy with the risk
of incorrect diagnosis. Radiologists should also address the need to educate our referring colleagues about the real level of risk. We owe our patients nothing less.
References 1. Haschek E, Lindenthal O: A contribution to the practical use of the photography according to Röntgen. Wien Klin Wochenschr 9:63, 1896. 2. Osborne ED, Sutherland CG, Scholl AJ et al: Roentgenography of urinary tract during excretion of sodium iodide. Journal of the American Medical Association 80.6:368-373, 1923. 3. Berberich J, Hirsch S: Die röntgenographische Darstellung der Arterien und Venen am lebenden Menschen. Journal of Molecular Medicine2.49: 2226-2228, 1923. 4. Wallingford VH: The development of organic iodide compounds as x-ray contrast media. J Am Pharmacol Assoc 42:721-728, 1953. 5. Swick M: Excretion urography by means of the intravenous and oral administration of sodium orthiodohippurate with some physiological considerations. Surg Gynecol Obstet 56: 62-65, 1933. 6. Hoppe JO, Larsen HA, Coulston FJ: Observations on the toxicity of a new urographic contrast medium, sodium 3, 5-diacetamido-2, 4, 6, triiodobenzoate (Hypaque sodium) and related compounds. J Pharmacol Exp Ther 116:394-403, 1956
7. Sovak, M. Contrast media: a journey almost sentimental. Investigative radiology 29: S4-S14, 1994. 8. Manitz G, Matthes K: Technique of renal biopsy. The Lancet 290.7514: 511-512, 1967. 9. Ansell, G: A national survey of radiological complications: interim report. Clinical Radiology 19.2:175-191, 1968. 10. Morcos SK, Thomsen HS , Webb JA: Contrast media induced nephrotoxicity: a consensus report—Contrast Media Safety Committee, European Society of Urogenital Radiology (ESUR). Eur Radiol 9(8):1602–1613, 1999 11. Hou, SH, Bushinsky DA, Wish JB et al: Hospital-acquired renal insufficiency: a prospective study. The American journal of medicine 74.2:243-248, 1983 12. Anderson RJ, Linas SL, Berns AS, et al: Nonoliguric acute renal failure. N Engl J Med 296 (20): 1134–1138, 1977 13. Gruberg L, Mintz GS, Mehran R, et al: The prognostic implications of further renal function deterioration within 48 h of interventional coronary procedures in patients with preexistent chronic renal insufficiency. J Am Coll Cardiol 36 (5):1542–1548, 2000 14. McCullough PA, Adam A, Becker CR, et al: Epidemiology and prognostic implications of contrast-induced nephropathy. The American journal of cardiology 98.6:5-13, 2006 15. Davenport MS, Khalatbari S, Cohan RH et al: Contrast material–induced nephrotoxicity and intravenous low-osmolality iodinated contrast material: risk stratification by using estimated glomerular filtration rate. Radiology 268.3:719-728, 2013
16. Sprung CL, and Winick BJ: Informed consent in theory and practice: Legal and medical perspectives on the informed consent doctrine and a proposed reconceptualization. Critical care medicine 17.12: 1346-1354, 1989 17. Lambe HA, Hopper KD, Matthews YL: Use of informed consent for ionic and nonionic contrast media. Radiology 184.1:145-148, 1992 18. Cramer BC, Parfrey PS, Hutchinson TA, et al: Renal function following infusion of radiologic contrast material: a prospective controlled study. Arch Intern Med: 145 (1):87–89, 1985 19. Heller CA, Knapp J, Halliday J et al: Failure to demonstrate contrast nephrotoxicity. Med J Aust 155(5):329–332, 1991 20. Rao QA, and Newhouse JH: Risk of Nephropathy after Intravenous Administration of Contrast Material: A Critical Literature Analysis 1. Radiology 239.2:392-397, 2006 21. Newhouse JH, Kho D, Rao QA et al: Frequency of serum creatinine changes in the absence of iodinated contrast material: implications for studies of contrast nephrotoxicity. American Journal of Roentgenology 191.2:376-382, 2008 22. Bansal GJ, Darby M: Measurement of change in estimated glomerular filtration rate in patients with renal insufficiency after contrast-enhanced computed tomography: a casecontrol study. J Comput Assist Tomogr 33(3):455–459, 2009 23. Aulicky P, Mikulík R, Goldemund D et al: Safety of performing CT angiography in stroke patients treated with intravenous thrombolysis. J Neurol Neurosurg Psychiatry 81(7):783– 787, 2010
24. Langner S, Stumpe S, Kirsch M et al: No increased risk for contrast-induced nephropathy after multiple CT perfusion studies of the brain with a nonionic, dimeric, iso-osmolal contrast medium. AJNR Am J Neuroradiol 29(8):1525–1529, 2008 25. Oleinik A, Romero JM, Schwab K et al: CT angiography for intracerebral hemorrhage does not increase risk of acute nephropathy. Stroke 40(7):2393–2397, 2009. 26. Tremblay LN, Tien H, Hamilton P et al: Risk and benefit of intravenous contrast in trauma patients with an elevated serum creatinine. J Trauma 59(5):1162–1166, 2005 27. Bruce RJ, Djamali A, Shinki K et al: Background fluctuation of kidney function versus contrast-induced nephrotoxicity. American Journal of Roentgenology 192.3:711-718, 2009. 28. McDonald RJ, McDonald JS, Bida JP et al: Intravenous contrast material–induced nephropathy: causal or coincident phenomenon? Radiology 267.1:106-118, 2013. 29. McDonald RJ, McDonald JS, Carter RE et al: Intravenous contrast material exposure is not an independent risk factor for dialysis or mortality. Radiology 273.3: 714-725, 2014. 30. McDonald JS, McDonald RJ, Carter RE et al: Risk of intravenous contrast material–mediated acute kidney injury: a propensity score–matched study stratified by baseline-estimated glomerular filtration rate. Radiology 271.1: 65-73, 2014. 31. Sagy M, Aladjem M, Shem-Tov A et al: The renal effects of radiocontrast administration during cardioangiography in two different groups with congenital heart disease. Eur J Pediatr 141(4):236–239, 1984. 32. Taliercio CP, Vlietstra RE, Fisher LD et al: Risks for renal dysfunction with cardiac angiography. Ann Intern Med 104(4):501–504, 1986
33. Gruberg L, Mintz GS, Mehran R et al: The prognostic implications of further renal function deterioration within 48 h of interventional coronary procedures in patients with preexistent chronic renal insufficiency . J Am Coll Cardiol 36(5):1542–1548, 2000 34. Katzberg RW, Newhouse JH. Intravenous contrast medium-induced nephropathy: is the medical risk really as great as we have come to believe? Radiology 256:21-28, 2010 35. Byrd L, Sherman RL: Radiocontrast- induced acute renal failure: a clinical and pathophysiologic review. Medicine (Baltimore) 58(3):270–279, 1979 36. Moore RD, Steinberg EP, Powe NR et al: Nephrotoxicity of high-osmolality versus lowosmolality contrast media: randomized clinical trial. Radiology 182:649-655, 1992 37. Tepel M, van der Giet M, Schwarzfeld C et al: Prevention of radiographic-contrast-agentinduced reductions in renal function by acetylcysteine. N Engl J Med 343(3):180–184, 2000. 38. Becker CR, Reiser MF: Use of iso-osmolar nonionic dimeric contrast media in multidetector row computed tomography angiography for patients with renal impairment. Invest Radiol 40(10):672–675, 2005. 39. Barrett BJ, Katzberg RW, Thomsen HS et al: Contrast-induced nephropathy in patients with chronic kidney disease undergoing computed tomography: a double-blind comparison of iodixanol and iopamidol. Invest Radiol 41(11):815–821, 2006. 40. Thomsen HS, Morcos SK, Erley CM et al: The ACTIVE Trial: comparison of the effects on renal function of iomeprol-400 and iodixanol-320 in patients with chronic kidney disease undergoing abdominal computed tomography. Invest Radiol 43(3):170–178, 2008.
41. Kuhn MJ, Chen N, Sahani DV et al: The PREDICT study: a randomized double-blind comparison of contrast-induced nephropathy after low or iso-osmolar contrast agent exposure. AJR Am J Roentgenol 191(1):151–157, 2008 . 42. Nguyen SA, Suranyi P, Ravenel JG et al: Isoosmolality versus low-osmolality iodinated contrast medium at intravenous contrast-enhanced CT: effect on kidney function. Radiology 248(1):97–105, 2008 . 43. Weisbord SD, Mor MK, Resnick AL et al: Incidence and outcomes of contrast-induced AKI following computed tomography. Clin J Am Soc Nephrol 3(5):1274–1281, 2008 . 44. Rudnick MR, Goldfarb S, Wexler L et al: For the Iohexol Cooperative Study. Nephrotoxicity of ionic and nonionic contrast media in 1196 patients: a randomized trial. Kidney Int 47(1): 254–261, 1995. 45. McDonald, JS, McDonald RJ, Lieske JC et al: Risk of acute kidney injury, dialysis, and mortality in patients with chronic kidney disease after intravenous contrast material exposure. Mayo Clinic Proceedings 90(8):1046-1053, 2015. 46. McDonald JS, McDonald RJ, Comin J et al: Frequency of acute kidney injury following intravenous contrast medium administration: a systematic review and metaanalysis. Radiology 267.1:119-128, 2013. 47. Appropriateness Criteria. American College of Radiology. Web. 12 Jan 2016. 48. Skinner DG, Colvin RB, Vermillion CD et al: Diagnosis and management of renal cell carcinoma: a clinical and pathologic study of 309 cases. Cancer 28:1165-1177, 1971. 49. Konnak JW, Grossman HB: Renal cell carcinoma as an incidental finding. J Urol 134:10941096, 1985.
50. Jayson M, Sanders H: Increased incidence of serendipitously discovered renal cell carcinoma. Urology 51.2: 203-205, 1998. 51. Chow WH, Devesa SS, Warren JL et al: Rising incidence of renal cell cancer in the United States. JAMA 281:1628–31, 1999.
Figure Legends Fig. 1. Pyelonephritis. A: Non-contrast axial CT of the abdomen of a young patient presenting with abdominal pain. The kidneys are normal. B: Contrast-enhanced axial CT of the abdomen demonstrates a focal wedge-shaped region of decreased enhancement in the anterior right kidney extending to the periphery of the cortex.
Fig. 2. Multifocal renal infarct. A: Non-contrast axial CT of the abdomen in a patient with history of abdominal pain. The kidneys are normal. B: Contrast-enhanced axial CT demonstrates multiple wedgeshaped regions of decreased enhancement in the left kidney.
Overestimating the Risk of Intravenous Contrast Medium-Induced Nephropathy: A Pitfall in Imaging the Genitourinary System Lyndon Luk M.D., Jeffrey H. Newhouse M.D.
www.elsevier.com/locate/enganabound
PII: DOI: Reference:
S0037-198X(16)00004-3 http://dx.doi.org/10.1053/j.ro.2016.01.003 YSROE50535
To appear in:
Seminar in Roentgenology
Cite this article as: Lyndon Luk M.D., Jeffrey H. Newhouse M.D., Overestimating the Risk of Intravenous Contrast Medium-Induced Nephropathy: A Pitfall in Imaging the Genitourinary System, Seminar in Roentgenology, http://dx.doi.org/10.1053/j. ro.2016.01.003 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Overestimating the Risk of Intravenous Contrast Medium-Induced Nephropathy: A Pitfall in Imaging the Genitourinary System
Lyndon Luk1, M.D. Jeffrey H. Newhouse1, M.D. 1
Department of Radiology, Columbia University Medical Center, New York, New York
Address for Correspondence
Lyndon Luk, MD Department of Radiology Columbia University Medical Center 622 West 168th Street, PB-1-301 New York, NY 10032 Tel: 201-562-2739 Email: [email protected]
Abstract The overestimation of overall risk and consequences of intravenous contrast media-induced nephropathy has perpetuated an unwarranted avoidance of contrast-enhanced studies with consequent reduction in effectiveness of management of a number of genitourinary diseases. Properly assessing the risk of contrast nephropathy and avoiding this common imaging pitfall is critical in avoiding or misdiagnosing otherwise treatable genitourinary pathology.
History Iodinated contrast media (CM) were first discovered and used in diagnostic radiology nearly a century ago [1], leading to the first clinical pyelogram performed by Osborne, Southerland, Scholl and Roundtree in 1923 [2] and the first femoral arteriogram by Berberich and Hersch in 1924 [3]. Not long after, Wallingford, a chemist, created the first iodine-containing benzoic acid ring. [4]. The contributions of Wallingford, Swick [5] and Hoppe et al. [6] in the 1950s led to markedly improved contrast opacification as well as patient tolerance.
By the 1970s, annual consumption of contrast media surpassed 2,000 metric tons, fueled by the discovery of computer tomography (CT) [7]. At first, there was little thought given to the idea that iodinated CM were potentially nephrotoxic. Multiple publications documenting this phenomenon were published starting in the 1960s; Manitz and Matthes [8] and Ansell [9] documented transient anuria in renal failure patients after intravenous pyelography. The concept of acute renal dysfunction caused by intravenous CM has since become fundamental
and axiomatic in both practice and literature of modern medicine [10], with articles numbering in the thousands. Definition of Contrast-induced nephropathy (CIN) Contrast-induced nephropathy (CIN) is typically defined as an absolute or percentage increase in serum creatinine (SCr) level over baseline; 0.5 mg/dl is the commonest threshold for an absolute rise; 25%, 50% and 100% have all been used in published series dealing with CIN. With these thresholds, the risk of CIN in a cohort of patients receiving contrast ostensibly determined by clinical series has ranged from 0% to nearly 50%, and the decades-long flow of publications purporting to document CIN has led to a widespread conviction that the risk is considerable, both among practitioners who refer patients for various contrast-requiring examinations and among radiologists and interventional cardiologists. One of the most widely-cited quotes from papers dealing with CIN risk [11] states that contrast is the third-most-common cause of acute kidney injury (AKI) in hospitalized patients. Although the commonest course of CIN consists of a transient rise in serum creatinine [12], it has been claimed that in some cases renal function does not return to baseline, and may even require chronic dialysis [13]. It has also been found that patients who experience CIN are at risk for longer hospital stays than patients without it, and even run a higher risk of death [14]. Given these concerns, radiology departments and practices usually have established policies which preclude administering intravenous contrast in patients deemed to be at risk. Since the risk is felt to rise in proportion to degrees of chronic renal failure [15], thresholds of serum creatinine or estimated glomerular filtration rate (eGFR) are usually used to identify patients for whom contrast administration is precluded. Patients may also have contrast withheld if serum
creatinine is below the threshold but is rising, despite lack of rigorous evidence that contrast is deleterious in this circumstance. Radiology practices and departments may establish policies requiring a recent creatinine or eGFR to be measured even for patients whose risk of renal disease is very small. Many also require that informed consent be formally obtained and documented [16, 17]. All of these practices (not to mention fear of litigation) have led to great reluctance to administer contrast to patients whose management requires information only available from its use. Discussion Our thesis is that the perception of risk of nephropathy from intravenous CM throughout the medical community is much higher than the real risk warrants. There are two major sources of this misconception: the overestimation of risk in nearly all publications due to the serious error of omitting controls, and the conflation of the risk of procedures requiring intravenous contrast administration with the administration of contrast during cardiac angiography. We will deal with each of these in order.
From the initial recognition that contrast may cause renal dysfunction, there has been an increasing rate of publications regarding the risk of CIN; literally hundreds of clinical series on the topics have appeared. The experiments detailed within these publications were performed with varying degrees of scientific rigor, but nearly all assumed that any renal dysfunction which was found after administration of contrast was caused by the contrast; control series of patients to estimate the incidence of renal dysfunction without contrast were, for decades, never provided. Studies performed by Cramer et al. in 1985 and Heller et al. in 1991 did include
such controls [18, 19]. Each found rates of post-contrast nephropathy within the ranges reported in previous publications; however, each also found that these rates did not exceed those assessed in their control groups, and concluded that, at least in the circumstances they reported, contrast-induced changes might not, in fact, occur.
Despite their potential importance, these two publications were rarely cited in subsequent reported experiments. In 2006, they were stressed in a critical review of the literature [20]. This review was shortly followed in 2009 by an experiment in which the incidence of short-term creatinine increases were assessed in a very large group of patients who had had no contrast [21]; these figures were strikingly similar to the rates of post-contrast nephropathy found in all of the previous literature involving intravenous contrast administration, further calling into question the validity of these experiments.
In the few years following this series, several additional clinical series appeared which did include control groups [22-26]. All but one found no excess cases of nephropathy after contrast beyond those which appeared in control patients; the one which did reported results from two contrast agents, with no difference from control groups found for one agent and a slight increase in nephropathy rates for the other [27].
Subsequently, these controlled studies were criticized since control patients were those receiving non-contrast CT scans, and the data from the two groups were compared retrospectively. Since many of the control patients were steered to non-contrast scans because
their referring physicians felt their renal function to be particularly at risk, a selection bias could well have arisen; a real tendency for contrast to cause renal dysfunction might have been masked by the control patients’ excess tendency to experience renal dysfunction for noncontrast-related reasons.
Multiple articles addressed possible selection bias by performing 1:1 propensity matching and propensity score analysis [15, 28, 29], a statistical technique intended to reduce the effects of differences between experimental and control groups in retrospective studies by evaluating variables that may predict either increased or decreased likelihood of receiving a particular treatment. One study identified low-contrast contrast material (LOCM) as a risk for CIN in patients with GFR less than 30 ml/min/1.73 m2 [15] after propensity score adjustment. Other studies found no significant difference in AKI risk between patients undergoing non-contrast or contrast studies in any risk subgroup after propensity score adjustment [28, 29] and identified AKI risk as independent of CM exposure, even in patients with eGFR less than 30 ml/min/1.73 m2 [30].
Angiocardiography has long been recognized as a procedure that can lead to AKI [31, 32]. In the past two decades, most of the literature regarding CIN reports studies involves angiocardiography, and much of the current consensus regarding increased morbidity and mortality of CIN has arisen from these publications. As an example, a study by Gruberg et al. of patients undergoing percutaneous coronary intervention with baseline creatinine ≥1.8 mg/dL found a 37.7% rate of CIN, a 7.1% rate of CIN requiring hemodialysis and a 22.6% mortality rate
in patients requiring dialysis [33]. In the discussion sessions of these papers, and those dealing with intravenous contrast, CIN risk tends to be considered as a single entity, with little attention given to the differences in risk between the two types of procedures. Conflation of these study results has led to a serious overestimate of risk of intravenous contrast, as detailed by Katzberg and Newhouse in a detailed literature review published in 2010 [34].
The incidence of CIN with IV contrast media has been overstated not only because of extrapolation of angiocardiography experience despite the literature detailing the significantly safer profile of IV CM in comparison to contrast-enhanced cardiac studies which dates back to as early as 1979 [35]. In 1992, Moore et al. found a greater than two-fold increase in the rate of nephrotoxicity in patients undergoing angiocardiography versus those undergoing CE CT [36]. Review of more contemporary prospective studies investigating the use of LOCM and isoosmolar CM show an overall CIN rate of approximately 5.4 % [15, 37-43], including a post IV CM CIN rate of 5.2% in patients with renal insufficiency and diabetes mellitus [41]. In comparison, overall CIN rate in patients with chronic kidney disease and diabetes as depicted by the cardiology literature are upwards of 33%, noted by Rudnick et al. in the Iohexol Cooperative Study [44].
The overall difference in the rate of morbidity and mortality ostensibly caused by CIN in patients receiving IV CM and those receiving intraarterial CM during cardiac angiography and intervention is even more pronounced. The same CIN literature review in 2006 evaluating CE CT in patients with renal insufficiency [20] found no documented cases of CIN requiring dialysis or
death out of a total of 1,175 subjects. A 2015 study by McDonald evaluating rates of AKI, emergent dialysis and mortality in a large, propensity matched cohort with stage III-V CKD revealed no significant differences in morbidity or mortality between the noncontrast and contrast groups [45]. A recent systematic review and meta-analysis of over 25,000 patients demonstrated similar rates of AKI, dialysis and death between CE and control groups [46]
There is no doubt that nephropathy can be a serious condition; after all, it constitutes failure of an important organ system, which may in turn have effects on other systems. And nephropathy may not only be a primary event: it may be the result of, and act as a marker for, failure of other organs. These events would be expected to lengthen hospital stays, occasionally require dialysis and even increase mortality rates, but if they occur for reasons other than intravenous contrast, their temporal association with contrast administration does not imply causation.
In sum, recent investigations show that CIN, and any serious consequence of it, is much less of a risk than most physicians have believed for decades. The debate about its exact incidence may continue, but it is now evident that it is low. Use of contrast, therefore, must be reevaluated since the added value of CM in the cross-sectional imaging evaluation of numerous diseases affecting the genitourinary (GU) system is irrefutable.
The American College of Radiology (ACR) Appropriateness Criteria can serve as a good indicator of the fraction of cases dealing with GU issues which do or do not need contrast enhancement
for CT. In the urologic, womens' imaging and vascular sections of these criteria there are 43 variants of GU conditions for which CT is or may be indicated [47]. Among these, in 38 variants, CT with contrast is rated more appropriate than CT without contrast; in only 5 is CT without contrast more appropriate. Some conditions where contrast administration may either clinch or clarify the diagnosis include acute flank pain which may be due to pyelonephritis (figure 1) or acute renal ischemia (figure 2), identifying vasculitis, arteriovenous fistulae and venous thrombosis, demonstrating active bleeding and pseudo-aneurysms due to trauma, diagnosing enhancing urothelial tumors, screening for renal tumors in syndromes which may include them and assessing endometrial morphology.
It should be immediately obvious from the above that in the conditions which may be missed or mischaracterized if contrast is withheld, appropriate patient management can be severely impeded in a number of ways. And the problem is wider: a great deal of GU disease is discovered incidentally on CT scans initially performed for other problems. If contrast is unnecessarily withheld from abdominal CT performed for any reason, significant errors in GU diagnoses may ensue. Detecting renal cell carcinoma in patients in whom it is not suspected is only one example. As abdominal CT has become more frequently performed, an increasing fraction of these tumors are discovered serendipitously, and more often at a stage which permits cure [48-51]. And although rigorous demonstration that contrast is necessary for this diagnosis is lacking, any experienced abdominal radiologist would agree that small intrarenal masses are easy to miss on an examination performed without contrast. Furthermore, nearly all ACR Appropriateness Criteria that include CT as a possible imaging modality in addressing
various gastrointestinal issues advocate CM; of the 61 listed variants of gastrointestinal conditions with ACR imaging recommendations that include CT, all but 2 rate CT with contrast as more appropriate than CT without contrast.
The frequency with which contrast is unnecessarily withheld in CT examinations is not known, but the authors' experiences suggest that the problem occurs with regularity. Factors incorrectly felt to confer risk of CIN include mild chronic renal failure, diabetes without renal failure, congenital or surgical absence of one kidney with normal creatinine, transplanted kidneys and recent administration of contrast. (Mild prior systemic contrast reactions are also often cited as reasons to withhold contrast, but these circumstances are beyond the scope of this paper.) Discussions between radiologists and referring physicians intended to compare the risks of CIN with the risks of missing diagnoses which require contrast, are also frequently foregone at a rate which is also not known, but probably considerable.
We conclude, therefore, that many essential CT examinations requiring contrast are withheld due to an unwarranted fear of nephropathy and its associated risks with consequent reduction in effectiveness of management of a number of GU and other diseases. This practice constitutes a sufficient threat to health that serious attempts to correct it are warranted. Within radiology departments, policy constraints for intravenous contrast administration should be reviewed and liberalized where possible, and policies which strictly forbid contrast administration should be replaced by requirements for realistic comparisons of the risk of nephropathy with the risk
of incorrect diagnosis. Radiologists should also address the need to educate our referring colleagues about the real level of risk. We owe our patients nothing less.
References 1. Haschek E, Lindenthal O: A contribution to the practical use of the photography according to Röntgen. Wien Klin Wochenschr 9:63, 1896. 2. Osborne ED, Sutherland CG, Scholl AJ et al: Roentgenography of urinary tract during excretion of sodium iodide. Journal of the American Medical Association 80.6:368-373, 1923. 3. Berberich J, Hirsch S: Die röntgenographische Darstellung der Arterien und Venen am lebenden Menschen. Journal of Molecular Medicine2.49: 2226-2228, 1923. 4. Wallingford VH: The development of organic iodide compounds as x-ray contrast media. J Am Pharmacol Assoc 42:721-728, 1953. 5. Swick M: Excretion urography by means of the intravenous and oral administration of sodium orthiodohippurate with some physiological considerations. Surg Gynecol Obstet 56: 62-65, 1933. 6. Hoppe JO, Larsen HA, Coulston FJ: Observations on the toxicity of a new urographic contrast medium, sodium 3, 5-diacetamido-2, 4, 6, triiodobenzoate (Hypaque sodium) and related compounds. J Pharmacol Exp Ther 116:394-403, 1956
7. Sovak, M. Contrast media: a journey almost sentimental. Investigative radiology 29: S4-S14, 1994. 8. Manitz G, Matthes K: Technique of renal biopsy. The Lancet 290.7514: 511-512, 1967. 9. Ansell, G: A national survey of radiological complications: interim report. Clinical Radiology 19.2:175-191, 1968. 10. Morcos SK, Thomsen HS , Webb JA: Contrast media induced nephrotoxicity: a consensus report—Contrast Media Safety Committee, European Society of Urogenital Radiology (ESUR). Eur Radiol 9(8):1602–1613, 1999 11. Hou, SH, Bushinsky DA, Wish JB et al: Hospital-acquired renal insufficiency: a prospective study. The American journal of medicine 74.2:243-248, 1983 12. Anderson RJ, Linas SL, Berns AS, et al: Nonoliguric acute renal failure. N Engl J Med 296 (20): 1134–1138, 1977 13. Gruberg L, Mintz GS, Mehran R, et al: The prognostic implications of further renal function deterioration within 48 h of interventional coronary procedures in patients with preexistent chronic renal insufficiency. J Am Coll Cardiol 36 (5):1542–1548, 2000 14. McCullough PA, Adam A, Becker CR, et al: Epidemiology and prognostic implications of contrast-induced nephropathy. The American journal of cardiology 98.6:5-13, 2006 15. Davenport MS, Khalatbari S, Cohan RH et al: Contrast material–induced nephrotoxicity and intravenous low-osmolality iodinated contrast material: risk stratification by using estimated glomerular filtration rate. Radiology 268.3:719-728, 2013
16. Sprung CL, and Winick BJ: Informed consent in theory and practice: Legal and medical perspectives on the informed consent doctrine and a proposed reconceptualization. Critical care medicine 17.12: 1346-1354, 1989 17. Lambe HA, Hopper KD, Matthews YL: Use of informed consent for ionic and nonionic contrast media. Radiology 184.1:145-148, 1992 18. Cramer BC, Parfrey PS, Hutchinson TA, et al: Renal function following infusion of radiologic contrast material: a prospective controlled study. Arch Intern Med: 145 (1):87–89, 1985 19. Heller CA, Knapp J, Halliday J et al: Failure to demonstrate contrast nephrotoxicity. Med J Aust 155(5):329–332, 1991 20. Rao QA, and Newhouse JH: Risk of Nephropathy after Intravenous Administration of Contrast Material: A Critical Literature Analysis 1. Radiology 239.2:392-397, 2006 21. Newhouse JH, Kho D, Rao QA et al: Frequency of serum creatinine changes in the absence of iodinated contrast material: implications for studies of contrast nephrotoxicity. American Journal of Roentgenology 191.2:376-382, 2008 22. Bansal GJ, Darby M: Measurement of change in estimated glomerular filtration rate in patients with renal insufficiency after contrast-enhanced computed tomography: a casecontrol study. J Comput Assist Tomogr 33(3):455–459, 2009 23. Aulicky P, Mikulík R, Goldemund D et al: Safety of performing CT angiography in stroke patients treated with intravenous thrombolysis. J Neurol Neurosurg Psychiatry 81(7):783– 787, 2010
24. Langner S, Stumpe S, Kirsch M et al: No increased risk for contrast-induced nephropathy after multiple CT perfusion studies of the brain with a nonionic, dimeric, iso-osmolal contrast medium. AJNR Am J Neuroradiol 29(8):1525–1529, 2008 25. Oleinik A, Romero JM, Schwab K et al: CT angiography for intracerebral hemorrhage does not increase risk of acute nephropathy. Stroke 40(7):2393–2397, 2009. 26. Tremblay LN, Tien H, Hamilton P et al: Risk and benefit of intravenous contrast in trauma patients with an elevated serum creatinine. J Trauma 59(5):1162–1166, 2005 27. Bruce RJ, Djamali A, Shinki K et al: Background fluctuation of kidney function versus contrast-induced nephrotoxicity. American Journal of Roentgenology 192.3:711-718, 2009. 28. McDonald RJ, McDonald JS, Bida JP et al: Intravenous contrast material–induced nephropathy: causal or coincident phenomenon? Radiology 267.1:106-118, 2013. 29. McDonald RJ, McDonald JS, Carter RE et al: Intravenous contrast material exposure is not an independent risk factor for dialysis or mortality. Radiology 273.3: 714-725, 2014. 30. McDonald JS, McDonald RJ, Carter RE et al: Risk of intravenous contrast material–mediated acute kidney injury: a propensity score–matched study stratified by baseline-estimated glomerular filtration rate. Radiology 271.1: 65-73, 2014. 31. Sagy M, Aladjem M, Shem-Tov A et al: The renal effects of radiocontrast administration during cardioangiography in two different groups with congenital heart disease. Eur J Pediatr 141(4):236–239, 1984. 32. Taliercio CP, Vlietstra RE, Fisher LD et al: Risks for renal dysfunction with cardiac angiography. Ann Intern Med 104(4):501–504, 1986
33. Gruberg L, Mintz GS, Mehran R et al: The prognostic implications of further renal function deterioration within 48 h of interventional coronary procedures in patients with preexistent chronic renal insufficiency . J Am Coll Cardiol 36(5):1542–1548, 2000 34. Katzberg RW, Newhouse JH. Intravenous contrast medium-induced nephropathy: is the medical risk really as great as we have come to believe? Radiology 256:21-28, 2010 35. Byrd L, Sherman RL: Radiocontrast- induced acute renal failure: a clinical and pathophysiologic review. Medicine (Baltimore) 58(3):270–279, 1979 36. Moore RD, Steinberg EP, Powe NR et al: Nephrotoxicity of high-osmolality versus lowosmolality contrast media: randomized clinical trial. Radiology 182:649-655, 1992 37. Tepel M, van der Giet M, Schwarzfeld C et al: Prevention of radiographic-contrast-agentinduced reductions in renal function by acetylcysteine. N Engl J Med 343(3):180–184, 2000. 38. Becker CR, Reiser MF: Use of iso-osmolar nonionic dimeric contrast media in multidetector row computed tomography angiography for patients with renal impairment. Invest Radiol 40(10):672–675, 2005. 39. Barrett BJ, Katzberg RW, Thomsen HS et al: Contrast-induced nephropathy in patients with chronic kidney disease undergoing computed tomography: a double-blind comparison of iodixanol and iopamidol. Invest Radiol 41(11):815–821, 2006. 40. Thomsen HS, Morcos SK, Erley CM et al: The ACTIVE Trial: comparison of the effects on renal function of iomeprol-400 and iodixanol-320 in patients with chronic kidney disease undergoing abdominal computed tomography. Invest Radiol 43(3):170–178, 2008.
41. Kuhn MJ, Chen N, Sahani DV et al: The PREDICT study: a randomized double-blind comparison of contrast-induced nephropathy after low or iso-osmolar contrast agent exposure. AJR Am J Roentgenol 191(1):151–157, 2008 . 42. Nguyen SA, Suranyi P, Ravenel JG et al: Isoosmolality versus low-osmolality iodinated contrast medium at intravenous contrast-enhanced CT: effect on kidney function. Radiology 248(1):97–105, 2008 . 43. Weisbord SD, Mor MK, Resnick AL et al: Incidence and outcomes of contrast-induced AKI following computed tomography. Clin J Am Soc Nephrol 3(5):1274–1281, 2008 . 44. Rudnick MR, Goldfarb S, Wexler L et al: For the Iohexol Cooperative Study. Nephrotoxicity of ionic and nonionic contrast media in 1196 patients: a randomized trial. Kidney Int 47(1): 254–261, 1995. 45. McDonald, JS, McDonald RJ, Lieske JC et al: Risk of acute kidney injury, dialysis, and mortality in patients with chronic kidney disease after intravenous contrast material exposure. Mayo Clinic Proceedings 90(8):1046-1053, 2015. 46. McDonald JS, McDonald RJ, Comin J et al: Frequency of acute kidney injury following intravenous contrast medium administration: a systematic review and metaanalysis. Radiology 267.1:119-128, 2013. 47. Appropriateness Criteria. American College of Radiology. Web. 12 Jan 2016. 48. Skinner DG, Colvin RB, Vermillion CD et al: Diagnosis and management of renal cell carcinoma: a clinical and pathologic study of 309 cases. Cancer 28:1165-1177, 1971. 49. Konnak JW, Grossman HB: Renal cell carcinoma as an incidental finding. J Urol 134:10941096, 1985.
50. Jayson M, Sanders H: Increased incidence of serendipitously discovered renal cell carcinoma. Urology 51.2: 203-205, 1998. 51. Chow WH, Devesa SS, Warren JL et al: Rising incidence of renal cell cancer in the United States. JAMA 281:1628–31, 1999.
Figure Legends Fig. 1. Pyelonephritis. A: Non-contrast axial CT of the abdomen of a young patient presenting with abdominal pain. The kidneys are normal. B: Contrast-enhanced axial CT of the abdomen demonstrates a focal wedge-shaped region of decreased enhancement in the anterior right kidney extending to the periphery of the cortex.
Fig. 2. Multifocal renal infarct. A: Non-contrast axial CT of the abdomen in a patient with history of abdominal pain. The kidneys are normal. B: Contrast-enhanced axial CT demonstrates multiple wedgeshaped regions of decreased enhancement in the left kidney.