Renal Adventures in Imaging

8/9/2019 Renal Adventures in Imaging

1/32

ADVENTURES IN RENALIMAGING

JOEL M. TOPF, M.D.

Clinical Nephrologist248.470.8163

N O O N

C O N F E R E N C EYou gotta eat lunch any ways, might as well learn how reduce the

number of patients you kill

2 2 2 0 1 M o r o s s R o a d , D e t r o i t t e l e p h o n e : 3 1 3 . 8 8 6 . 8 7 8 7

f a x : 3 1 3 . 8 8 6 . 4 1 0 3 w w w . p b f l u i d s . c o m


2/32

Introduction

The pattern of modern medical care:

The patient gets sick and presents with a set of symptoms.

Chest pain, shortness of breath

The physician determines a differential of possible diagnosis

Cardiac failure, cardiac ischemia, PE, pneumonia

The physician orders and performs a slate of diagnostic studies

Stress test, CT angio, Chest x-ray, troponins, D-dimers, cultures

After establishing a diagnosis drugs are administered and procedures performed

to cure or alleviate symptoms

antibiotics, thrombolytics, angiogram

A common calculation in determining the treatment is to weigh the risks inherent inthe therapy and compare them to the risks of the disease itself. The more severe the

disease the more toxicity we are wiling to accept in the therapy. Contemporary hospi-

talized patients are so ill that physicians routinely order therapies with huge degrees

of risk and associated morbidity: bypass surgery, chemotherapy, liver transplantation,

radiation therapy; there is nearly no limit to the amount of risk we will take to battle disease.

I believe that physicians have become so used to exposing patients to risk in thetreatment phase that they do not adjust the assessment of risk-benefit when exposing

patients to the risks inherent in diagnostic tests.

The benefit that patients receive from undergoing a diagnostic tests are more difficult

to pin-down than the benefit from treatment. What benefit does a patient gain from

accurately diagnosing a self-limited, spontaneously resolving condition like post-

strep GN? What benefit does a patient gain from accurately diagnosing a fatal and

untreatable condition like metastatic pancreatic cancer? Obviously patients have little

to gain from these procedures when they are successful. So, we should accept

very little risk in the diagnostic work-up of these conditions.

J o e l M. To pf , M.D.! N o o n C o n f e r e n c eA d v e n t u r e s i n R e n a l I m a g i n g! 2


3/32

When determining acceptable risk to patients in the diagnostic phase, you need to

have less tolerance for risk than in the treatment phase.

There are three different diagnostic situations where renal issues figure prominently:

1. The risk of contrast nephropathy with iodinated contrast agents

2. The risk of nephrogenic systemic sclerosis with gadolinium containing

paramagnetics for MRI/MRA

3. The risk of acute renal failure with sodium phosphate for bowel prep prior

to colonoscopy.

Acute phosphate nephropathy! 4Sodium phosphorous and nephrocalcinosis! 4Epidemiology! 6Conclusion! 10Key teaching points! 11

Gadolinium and Nephrogenic Fibrosing Dermopathy! 12Nephrogenic Fibrosing Dermopathy! 13Gadolinium! 13Epidemiology! 15Treatment! 16Conclusion! 17Key teaching points! 18

Strategies to Avoid Contrast Nephropathy: Case report! 19Definition of contrast nephropathy! 20Epidemiology! 20Consequences of contrast nephropathy! 22Preventing contrast nephropathy ! 24

The Contrast Agents! 24Hydration Strategies! 25Acetylcysteine! 29

Key Teaching points! 31



4/32

Acute phosphate nephropathyCC: Im still on dialysis

A 61 y.o. Caucasian female presented to my office after being told that she needed temporary

dialysis for acute renal failure. After 6 weeks of peritoneal dialysis there are no signs that her

kidneys are healing. Four months prior, she had started feeling tired and losing weight. Her

primary care doctor did some tests and told her that she had renal failure and sent her to a

nephrologist. This nephrologist performed a kidney biopsy (See below) and told her that she

had acute tubular necrosis, with calcium deposits in her kidneys, but that the glomeruli looked

great and he expected full recovery.

Past Medical History

No history of hypertension, diabetes or heart disease. No history of primary hyperparathyroid-

ism or renal tubular acidosis. Patient has a recto-vaginal fistula from Crohns disease that was

surgically repaired 4 months ago.

Patients who receive phosphate containing cathartics occasionally develop severe

acute renal failure. As the biopsy shows, unlike most renal disease the glomeruli are

spared with calcified debris in the tubules.

H & E Stain: Strange intratubular

deposits

Van Kossa Stain: Deposits are

calcium deposits

Hanging out on peritoneal dialysis

waiting for my kidneys to get

better.

SODIUM PHOSPHOROUS AND NEPHROCALCINOSISColonoscopy for cancer screening is increasing in the United States. In 2002 over 14

million colonoscopies were performed for cancer screening. Two prep agents are used

in almost all of these procedures: polyethylene glycol (PEG, Golytely) and oral so-

dium phosphate solution (OSPS, Fleet EZ-Prep). Sodium phosphate produces the

superior prep with better patient tolerance according to most analysis. An inadequite

prep increases the likelihood of aborting the procedure or missing a neoplasm.



5/32

While generally safe, OSPS has some theoretical disadvantages compared to PEG.

OSPS causes fluid shifts, allows phosphorous absorption, and has numerous contra-

indications for use in various populations.

In June 2004 Markowitz et al described 5 patients who developed serious renal com-

plications from the OSPS and ushered in a new era of caution regarding colonic prep.(Markowitz GS, Nasr SH, Klein P, Anderson H, Et al. Hum Pathol 35: 675-684; 2004.)

1 2 3 4 5

demo-

graphics

69 whitemale

82 AfricanAmericanmale

55 hispanicfemale

64 AfricanAmericanfemale

76 whitefemale

PMH

HTN, Pros-tate Ca1 HyperparaRad. proctitis

HTNCOPDcolon perf.

HTNDMCAD/MI

HTNDMobesity

HTNspastic colon

Meds

ARB ARBHCTZglitazone,

ASA

ACEiHCTZglitazone

ASA

ARBASA

Cr (GFR)before Na-Phos

1.2 mg/dl (64mL/min)2 weeksbefore

0.9 (104 mL/min)1 day before

0.6 (110 mL/min)3.5 monthsbefore

0.9 (81 mL/min)4 monthsbefore

0.9 (65 mL/min)1 yearbefore

Crafter Na-Phos

6.71 month after

5.21 week after

4.52 weeks after

2.38 weeks later

63 days after

Cr at Bi-opsy

6.37 weeks

4.94 weeks

4.016 weeks

3.028 weeks

8.01 week

Final Cr 8.5 (7 mL/min)6 weeks afterthe bx

4.3 (17 mL/min)3 monthsafter bx

3.5 (14 mL/min)5 weeks afterbx

3.3 (18 mL/min)3 weeks afterbx

3.7 (15 mL/min)3 wks afterbx

All the patients were biopsied and had normal glomeruli (except patient 4 who hadconcurrent nodular diabetic glomerulopathy) but significant acute and chronic

changes to the tubules. There were prominent tubule lesions of calcium phosphate.

Patients also had variable amounts of interstitial inflammation.



6/32

OSPS have been used as a cathartic since the 1893. In the early 90s their use as a

bowel prep agent was recognized. Prior to Markowitz, Fleets was thought of as a be-

nign agent as recognized in its other-the-counter status. The company quotes a rate of

serious adverse events of 1 per 1.4

million doses. Most adverse events

are related to inappropriate dosesor use in inappropriate patients.

Despite this enviable safety record

there were some warning signals

that there maybe problems with

OSPS.

Transient hyperphosphatemia isubiquitous and, previously, had

been thought to be without clinical

significance. It is likely that in pa-

tients with volume depletion and

decreased nephron mass (i.e.

chronic kidney disease) the in-

creased phosphorous load com-

bined with increased water reabsorption in the proximal tubules causes a pathologi-

cally elevated intratubular phosphorous concentration. This results in a spontaneousprecipitation of calcium phosphate and permanently damages the tubule and causes

loss of the nephron.

EPIDEMIOLOGY

Considerable caution should be ad-vised before OsmoPrep Tablets are used

in patients with the following illnesses:

severe renal insufficiency (creatinine

clearance less than 30 mL/minute),

congestive heart failure, ascites, unsta-

ble angina, gastric retention, ileus,

acute bowel obstruction, pseudo-

obstruction of the bowel, severe chronic

constipation, bowel perforation, acute

colitis, toxic megacolon, gastric bypass

or stapling surgery, or hypomotility

syndrome.

0

2

4

6

8

Vanner, 1990 Cohen, 1994

SerumPhos(mg/dL)

pre-OSPS

2 hr after 2nd dose

http://www.rxlist.com/script/main/art.asp?articlekey=3896http://www.rxlist.com/script/main/art.asp?articlekey=3896http://www.rxlist.com/script/main/art.asp?articlekey=2362http://www.rxlist.com/script/main/art.asp?articlekey=2362


7/32

Markowitzs paper

resulted in a wave of

case reports and

small case series

showing similar find-

ings. Markowitzhimself published a

definitive article with

21 cases. (Markowitz

GS et al. J Am Soc

Nephrol 2005; 16:3389.) The proposed risk factors of

decreased renal function, ACEi/ARB and diuretics were all verified. Case

reports and case-control studies are incapable of estimating the incidence of this prob-lem. There have been numerous studies that tried to retrospectively determine if this

was a common but unrecognized cause of renal failure or was rare.

Most use a similar methodology of combing a database of colonoscopies and looking

for patients who incidentally had pre- and post-procedure creatinines.

The Bronx VA

Singal et. al. reviewed 311 patients who underwent colonoscopy at the Peters Veter-

ans Hospital in the Bronx New York who incidentally had a creatinine in the 12

months preceding the procedure and the 12 months after the procedure. They found a

tiny difference in the change in the average serum creatinine (0.09 mg/dL lower with

PEG, p=0.005) and a large difference in the number of patients with a 25% increase in

serum creatinine. However, none of the patients had severe irreversible renal failure.

In the three patientswhose creatinines

rose over 2, all re-

turned to pre-

colonoscopy levels

within 6 months. No

cases of irreversible

severe renal failure in150+ administrations

of OSPS.


CONTRAINDICATIONS TO NA PHOS

SOLUTIONS FOR COLONOSCOPY PREP

Renal failure

Hyperparathyroidism

Congestive heart failure

Liver failure

Intestinal obstruction

0%

5%

10%

15%

20%

8%

20%

25%increaseinCr

p


8/32

Henry Ford Hospital

Researchers at Henry Ford Hospital looked at colonoscopies done from 1999-2005.

They excluded pre-existing renal disease, patients without at least one creatinine

within 12 months prior and 6 months after the procedure. This lead to a study popu-

lation of 2,352 patients. OSPS was used in 2,083 and PEG in 269. Eighty-eight of these

patients dropped their GFR below 60 cc/min in the 6 months after the colonoscopy

without an obvious explanation following chart review. There was no association be-

tween using OSPS and developing renal failure (RR 1.14 95% CI: 0.55-2.39). They did find

some general risk factors for developing renal insufficiency after a colonoscopy. See

table.

FACTOR RR 95% CI

Age > 65 3.23 2.08-5.03

African American Race 1.70 1.02-2.83

Baseline GFR 60-90 mL/min 5.60 3.18-9.86

Hypertension 2.99 1.73-5.19

ACEi 1.90 1.23-2.95

ARB 2.69 1.43-5.05

Thiazide diuretics 2.26 1.47-3.45

Russmann S, Lamerato L, Marfatia A, Motsko SP. Am J Gastroenterol 2007; 102: 2655-

2663.

University of Pennsylvania Health System

Bruneli performed a case control study of 141 cases of incident ARF following endo-

scopy (25% or 0.5 mg/dL increase in creatinine) and found no association with OSPS.

(Brunelli SM, Lewis JD, Gupta M, et al. J Am Soc Nephrol 18: 3199-3205, 2007)

Archives of Internal Medicine

Case control study of 286 patients with normal kidney function who underwent out-

patient colonoscopy with OSPS compared to 125 age and sex matched controls. Un-

fortunately since nearly everyone who underwent colonoscopy at this institution

used OSPS the investigators had to use control patients who did not get colonosco-



9/32

pies. The control group does allow one to look at the affect of the prep without being

confused by natural age-related loss of renal function.

70

72

74

76

78

80

Baseline 6 mo 12 mo

GFR(mL/min) =0.20

=0.002


10/32

CONCLUSION

In December 2008, the FDA required a boxed warning for OSPS regarding phosphate

nephropathy and identified the following risk groups:

Individuals who appear to have an increased risk of acute phosphate neph-

ropathy following the use of OSPs include persons: who are over age 55;who are hypovolemic or have decreased intravascular volume; who have

baseline kidney disease, bowel obstruction, or activecolitis; and

who are using medications that affect renal perfusion or function (such as

diuretics, angiotensin converting enzymeinhibitors, angiotensin

receptor blockers, and possibly nonsteroidal anti-inflammatorydrugs).

OSPS can have devastating renal complications. Unfortunately we have no idea who

is at risk of developing those complications. The risk profile we develop from case

series (ACEi, diuretics, elderly) looks almost exactly like a population that regularly

receives OSPS without any complications.

Though the risk factors have not been conclusively determined it seems prudent to

avoid using OSPS in patients with any degree of renal insufficiency (even with a

normal creatinine) and to stop ACEi, ARB and diuretics prior to the prep.

The most important next step would be to perform high quality prospective studies

to help determine the true incidence and risk factors for this devastating but rare

complication.



11/32

KEY TEACHING POINTS



12/32

Gadolinium and

Nephrogenic Fibrosing DermopathyChief Complaint: My legs are made of wood.

38 year old dialysis patient complains

of lower extremity edema. The patient

is poorly compliant with fluid restric-

tions and regularly comes in for di-

alysis with 5 to 6 kg of weight gain

since last dialysis. Additionally the

patient has resistant hypertensionwith systolic blood pressures running

from 180 to 200.

Past medical history

Diabetic nephropathy resulting in

ESRD, OSA with non-compliance

with CPAP, obesity and recent fever

associated with neck pain. The patient

had an MRI with gadolinium to rule

out a paraspinal abscess.

Physical exam

Remarkable for a pleasant, obese man.

His blood pressure is 180/79 with

lower extremities that have wood-like

skin changes, erythema, and indurated

plaque-like lesions.

Patient course:

The patient is aggressively ultra-filtered with additional treatments and prolonged nocturnal

dialysis. Unfortunately no improvement in the lower extremity edema or wood-like skin oc-

curs. He develops increased erythema resistant to IV antibiotics to treat a presumed cellulitis.Vascular surgery and infectious diseases suspect inadequate lower extremity circulation. An-

giography shows adequate perfusion and no venous insufficiency.



13/32

A skin biopsy is ultimately performed and shows:

The dermis demonstrates haphazardly arranged collagen bundles and a strikingly increased

number of spindled and plump fibroblast-like cells. There is a distinct absence of inflammatory

cells. The diagnosis is Nephrogenic Fibrosing Dermopathy.

NEPHROGENIC FIBROSING DERMOPATHY

Nephrogenic fibrosing dermopathy (NFD) is now more often referred by the more gen-

eral term nephrogenic systemic fibrosis (NSF) due to pathologic going beyond the skin.

In 2000, Cowper et al. published a case series of 14 hemodialysis patients who had

thickened, hardened skin, features suggestive ofscleromyxedema, however, none of the patients had

monoclonal gammopathy. In a later paper the same

group characterized the histology and suggested that

this was a new disease entity that had not previously

been recognized. (Lancet 2000;356: 1000-1)

NFD is an acquired condition that usually occurs in

patients on dialysis. The condition is characterized by

swelling of the distal extremities (legs and arms)

which then moves proximally (trunk), sparing the

face. Skin changes begin with swelling and progresses

to induration and stiffening joints. Pruritis is a vari-

able complaint. The speed of progression varies from case to case.

GADOLINIUM

Numerous agents and factors were suspected of causing NSF including ACEi, epo-

etin, clotting abnormalities, antiphospholipid antibodies, and acidosis. In January

2006, a case series of 5 patients with NFD was reported in which every patient had

received gadodiamide 2-4 weeks before the initial skin changes. They noted that all



14/32

the patients were acidotic and that 4 control patients that were on dialysis and ex-

posed to gadodiamide but were not acidotic did not develop NFD. The average dose

of contrast was 35 mL. (Grobner T. Nephro Dial Transplant 2006; 21: 1104-1108.)

This was substantiated by a second report of 13 hemodialysis patients who received

gadodiamide 2-75 days (mean 25 days) prior to development of skin changes. Theaverage dose was 38 mL. However; this group did not find any association with aci-

dosis. (Marckmann P, Skov L, Rossen K, et al. J Am Soc Nephrol 2006; 17: 2359-2362.)

Further corroborating the link between gadolinium and NFD were two reports that

demonstrated gadolinium deposition in the affected areas of the skin biopsies. (Boyd

A, Zic JA, Abraham JL. J Am Acad Dermotol 2007; 56: 27-30.)

Gadolinium contrast agents for MRI/MRA are rapidly cleared with a half

life of 1.3 hours in healthy volunteers.

In chronic kidney disease the half-life

can be extended from 30 to 120 hours.

Dialysis effectively clears gadolinium.

See graph. (study of 70 dialysis pa-

tients, each dialysis session was 4hours long. Okada S, Katagiri K, Et al.

Acta Radiol 2001; 42: 339-341.)

With the current hypothesis that gado-

linium is the cause of NSF the FDA

recommends avoiding gadolinium in

any patient with a GFR < 30 mL/min,any degree of acute renal failure, and

particularly with patients in hepatore-

nal syndrome. The FDA also recom-

mends that if at-risk patients are exposed to gadolinium that they receive prompt

and repeated (up to 4) dialysis sessions.

There has been an interest (especially from the manufacturers at the number of cases

fractionated by the specific gadolinium compound used. The Dallas VA uses Pro-

Hance and in a retrospective analysis of 198 exposures in hemodialysis patients they

found no cases of NFD. (Clin J Am Soc Nephrol 3: 747-751, 2008). In an animal study,

only rats exposed to gadodiamide developed NFD like symptoms. (Eur Radiol. 2008;

18: 2164-73.)

0

25

50

75

100

Infusion 1 2 3 4

0.0

78.2

95.6 98.7 99.5

Fractionofgadoliniumcleared

Hemodialysis Treatment



15/32

GENERIC

NAME

BRAND

NAME

DISSOCIATION

HALF-LIFE

NSF CASES: FDA

MEDWATCH

Gadobenate-

dimeglumineMultiHance N/A 10

Gadodiamide Omniscan 30 seconds 283

Gadopentate

dimeglumineMagnevist 10 minutes 125

Gadoteridol ProHance 3 hours 9

Gadoverse-

tamide Optimark N/A 20

EPIDEMIOLOGY

80% of reported cases have been in patients on dialysis and no patients have been

reported with CKD stages 1-3.

Two recent analysis have focused on patients with advanced CKD, not yet on dialysis.

Chrysochou looked at 2,278 Gd-MRI scans in patients not on dialysis and retrospec-

tively looked for evidence of subsequent

NFD over a ten year period in England.

(Clin J Am Soc Nephrol 5: 484-489, 2010)

They found no cases of NFD. One con-

cern is that 65% of the scans used Multi-

hance, a Gd compound with an appar-ently lower rate of NFD and only 2%

0

250

500

750

1,000

CKD1

CKD2

CKD3

CKD4

CKD5

Number of scans by CKD stage


Multihance

Magnevist

OmniscanDotarem

Gadovist

Vasovist


16/32

used gadodiamide, a frequent offender.

A few studies have attempted to look at the rate of NFD when following exposure to

gadolinium. Collidge and colleagues found an attack rate of 3.1% when they looked

at a cohort of dialysis patients who underwent MRI with gadolinium. They also

found a tight relationship of gadolinium exposure and increasing diagnosis of NFD.(Collidge TA,Thomson PC, et al. Radiology 2007: 245; 168.)

Deo and Cowper looked at NSF in and around Bridgeport, Connecticut. They found a

2.4% attack rate per MRI. (Deo A, Fogel M, Cowper SE. Clin J Am Soc Nephrol 2: 264

267, 2007.)

TREATMENT

Unfortunately there are no proven therapies for NSF. Some attempted therapies in-

clude: oral and topical steroids, cytoxin, extracorporeal photopheresis (ECP), tacro-

limus, IV IG, plasmapheresis, alpha interferon.

In one series from Stanford 5 patients underwent ECP. ECP is an immune modulating

treatment.

1. The patient ingests or is infused with a light-activatable drug, called psoralen.



17/32

2. Blood is temporarily removed from the patients. White blood cells are separated

out and then mixed outside the body with the patient's plasma, saline, heparin,

and 8-methoxyposoralen.

3. The preparation is then passed through a device where it is exposed to ultraviolet

light.

4. The treated mixture and untreated blood is combined and returned to the patient.

The procedure takes approximately 4 hours.

In the Stanford series 3 of the 5 patients reported marked improvement in self-

ambulation with the therapy (after 45 treatments, 6 and 20 weeks) A mean of 20

treatments were required to see an affect. Objective improvement in joint mobilitywas modest in the three responders and 2 patients had no improvement. The 60%

response rate is greater than the 20% improvement rate found in most series. (Rich-

mond H, Zwerner J, et al. Arch Dermatol.2007;143:1025-1030.)CONCLUSION

NSF is a new disease entity that occurs in dialysis patients. It was first recognized in

2000. NSF is characterized by thickened skin that causes joint contractures. It cancause fibrosis of the heart, diaphragm and skeletal muscle. It has been associated with

the use of gadolinium contrast used in MRI/MRA. There is no evidence that dialysis

reduces the risk of NSF after gadolinium exposure but since dialysis effectively re-

moves gadolium it is recommended that patients get dialysis immediatley after any

gadolinium exposure. J o e l M. To pf , M.D.! N o o n C o n f e r e n c eA d v e n t u r e s i n R e n a l I m a g i n g! 17


18/32

KEY TEACHING POINTS



19/32

Strategies to Avoid Contrast Nephropathy:

Case report

Chief Complaint: chest pain

70 year old African American presents to the ER with typi-

cal chest pain.

Past medical history:

Hypertension, hyperlipidemia diabetes and chronic kidney

disease with a baseline Cr of 2.5 mg/dL.

Patient course:

He is admitted to the CCU for unstable angina and rules out

for an MI with serial negative enzymes. Has undergoes

stress test for risk stratification and this shows reversible

ischemia with intact ejection fraction.

He goes for a cardiac catheterization following NAC and generous hydration with isotonicbicarbonate fluid. During the cath, a high grade lesion in the LAD is stented with a sirolimus

stent. Total dose of low osmolar, non-ionic dye is 220 cc.

On the second day after the catheterization his creatinine rises to 4.5 mg/dL. He remains non-

oliguric. The primary team consults neph-

rology. The next day the creatinine is 4.3

and nephrology clears the patient for dis-

charge with follow-up.

At the follow-up visit in one week the

creatine is 3.2 and 2 weeks after the cathe-

terization the creatinine is back baseline.

Summary

A patient at high risk for contrast

nephropathy gets exposed to contrast for a solid indication, gets sate of the art pro-

phylaxis, but develops contrast nephropathy anyways. Fortunately the patient gets

better without requiring dialysis.

0

1

2

3

4

5

Admit

Cath

PCD1

PCD2

PCD3

D/C

PCD7

P

CD14

SerumCrmg/dL



20/32

DEFINITION OF CONTRASTNEPHROPATHY

Radio contrast nephropathy (RCN)

is one of the leading causes of hospi-

tal acquired acute renal failure (12-

14% of ARF). It follows administra-tion of iodinated contrast for vascu-

lar visualization in CT scans, angi-

ography and venography. In pa-

tients who develop contrast nephro-

pathy the creatinine begins rising

immediately after exposure to the

contrast and typically peaks on the fourth day. The creatinine usually takes more thana week to return to pre-hydration levels.

RCN typically causes non-oliguric acute renal failure.

While over two thirds of patients (68%) will manifest some increase in creatinine, only

increases in creatinine of 0.5 mg/dL or 25% are considered clinically significant and

labelled as contrast nephropathy.

The 0.5 mg/dL comes from a questionnaire sent to attending physicians which asked:

What is the smallest change in creatinine that would cause you to delay a follow-up pro-

cedure or discharge?

Eighty percent of the 84 respondents responded an increase of 0.5 mg/dL or lower.

The 25% increase in creatininecomes from empiric data on 1

year mortality data based on

change in creatinine following

contrast.

EPIDEMIOLOGY

The incidence is generally about1-5% in unselected patients but

can rise to a much higher level in

high risk patients. The numerous

risk factors have been identified

0%

10%

20%

30%

40%

0-10%

11-25%

26-50%

>50%

1yearmortality

rise in creatinine



21/32

in the proliferation of studies on RCN

but two risk factors have been significant

in every study of sufficient power:

chronic kidney disease and diabetes.

(Rihal CS et al. Circulation 2002; 105:

2259-2264.)

0%

10%

20%

30%

40%

0-1.1

1.2-1.9

2.0-2.9

3.0

Mayo Clinic cardiac cath data

Frequencyof

RCN

Pre-Catheterization Cr

DM No DM



22/32

CONSEQUENCES OF CONTRAST NEPHROPATHY

Retrospective studies have looked at

the consequences of these modest

increases in creatinine and com-pared them to patients who did not

develop contrast nephropathy. Levy

found nearly a 5-fold increase in

death among patients with contrast

nephropathy regtardless if they

needed dialysis. (Levy EM, Horwitz

RI, Et al. JAMA 1996;275(19):1489-1494.)

The Mayo clinic has a registry of all the

cardiac catheterizations done at that insti-

tution. (Rihal CS, Et al. Circulation 2002;

105: 2259-2264.) They tracked their pa-

tients for five years following the expo-

sure and looked at mortality based on thedevelopment of contrast nephropathy.

The data is striking:

0%

15%

30%

45%

60%

75%

NoCN

CN

CN,noHD

CNandHD

62%

31%34%

7%

Levy, Horwitz, Et. al.

HospitalMortality


0%

10%

20%

30%

40%

50%

6 mo 12 mo 60 mo

15%

4%2%

45%

12%10%

Mayo Clinic Cath Registry

Mortality

Contrast Nephropathy No Contrast Nephropathy


23/32

The retrospective data is weak and few successful trials have been done with

intension-to-treat analysis. The few that have been done however have supported the

contention that contrast nephropathy itself is harmful and not just a marker of sus-

ceptible patients.


0%

2%

4%

6%

8%

10%

MAC

E

30-D

ayMort.

Fractionofpatients

0%

6%

12%

18%

24%

30%

Hospit

almo

rt.

12mon

thmo

rt.

Mueller C, Et al. Arch Intern Med

2002; 162: 329-336

Marenzi G, Marana I, Lauri G, et al. N

En Med 2003; 349: 1333-40.

Saline CVVH0.9% 0.45%

contrast


24/32

PREVENTING CONTRAST NEPHROPATHY

The Contrast Agents

While the dose of contrast is an important risk factor for RCN, the nature of the con-

trast agent itself is also a risk factor.

The first generation iodinatedcontrast agents were high os-

molar. These agents were tested

in a large trial of over 1000 pa-

tients and found to have a low

risk of contrast nephropathy,

approximately 2%. In the mid

90s low osmolar agentsemerged with two advantages:

better quality imaging and pur-

ported lower incidence of RCN.

Randomized trials of unselected

patients showed no advantage

to low osmolar agents (RCN

2.8% with high osmolar vs.

2.7% with low osmolar agents,

P=0.85, Harris KG, et al Radiol-

ogy 1991; 179:849-852.), how-

ever there was a trend to im-

proved safety in patients at

higher inherent risk. Follow-up studies restricted to high risk patients confirmed this

trend. These low osmolar agents were only low osmolar compared to the first genera-

tion agents. Low osmolar agents were still hypertonic to plasma with an osmolarity of

600-915 mOsm/kg H2O.

There is one contrast agent which is isosmolar to blood, iodixanol. In a pair of well

done randomized controlled trials (one in CT scans and the other in cardiac caths)

iodixanol has been shown to be safer than iohexol in regards to RCN. (Chalmers N,

Jackson RW. Br J Radiol 1999; 72: 701-3. Aspelin P, Aubry P, Fransson SG, et al. N EngJ Med 2003; 348: 491-9.)

Blood

Iodixanol

Ioxaglate

Iomeprol

loversoliopromide

iohexol

iopamidol

iopentol

iobitridol

diatrizoate

iothalamate 2,130

2,000

915

810

796

780

770702

620

600

290

290

Osmolality



25/32

Two subsequent studies (ICON by Mehran and CARE by Solomon) have challenged

this conclusion while the RECOVER trial by Jo support the use of isosmolar agents.

Hydration Strategies

Half normal saline

The traditional standard of care for preventing RCN is hydration with half normalsaline one mL per Kg for 12 hours before and 12 hours after exposure. The interesting

thing is this has never been tested in a RCT except as the control group in a negative

trial.

The Solomon trial (Solomon R, et al. N Engl J Med 1994; 331: 1416-20) looked at fu-

rosemide vs. manitol vs. placebo. All three groups received hydration with half nor-

mal saline.

The control group did less

bad than either of the ex-

perimental groups, thus

crowning half normal sa-

line as the champion and

standard of care.

Iodixanol Iohexol

0%

10%

20%

30%

Cr rise >0.5 Cr rise >1.0

15%

25%

3%

Aspelin data on Cardiac Cath

FrequencyofCN

Isotonic saline



26/32

The best study on hydration is a German study by Christian Mueller who hypothe-

sized that If half normal saline was good, then full normal normal saline should be

twice as good. He randomized 1,620 patients undergoing coronary angioplasty to

either 0.9% Saline for 24-hours or 0.45% saline for 24-hours. The patient population

was relatively low risk: 20% CKD, 16% diabetes, 65% hypertension, 16% EF< 45%.

Despite the low risk profile there was less than half as much CN with the normal sa-

line (p=0.04). This advantage was extended in three prospectively determined sub-

groups: women, patients with diabetes and patients receiving high volumes of con-

trast (>250 mL). There was no advantage among patients with CKD (Cr > 1.3) P=0.36.

Isotonic bicarbonate

0%

2%

4%

6%

Wom

en

Diabe

tes

>250mL

CKD

Freq.ofRCN

0.9% Saline

0.45% Saline



27/32

With isotonic saline crowned as superior to hypotonic fluids the question progressed

to whether chloride was the ideal anion to associate with sodium. In 2004 Merten, et

al published a randomized control trial of isotonic bicarbonate versus isotonic saline.

(Merten, Et al. JAMA 2004; 291: 2328-2334.) The study looked at 120 high risk popula-

tion: average creatinine 1.8, 48% diabetics and 95% cardiac catheterizations.

They found a 14% rate of CN with NaCl and 2% with Na Bicarbonate.

This study was criticized because the data safety monitoring board stopped the study

early and probably shouldnt have.

The RENO trial randomized patients going for emergency PCI. (Recio-Mayoral A,

Chaparro M, Prado B, et al. J Am Coll Cardiol 2007; 49: 1283-8) Sodium bicarbonate

plus NAC versus isotonic salinie. Unfortunately this study had three variables which

varied between the control and experimental groups making it impossible to tease

out the relative contributions of the individual factors



28/32

The study demonstrated a powerful treatment effect:

Masuda, Ozcan and Briguori all performed

additional trials of various sizes (n=59, 264,

326) that also supported the use of isotonic

bicarbonate.

Masuda M, Yamada T, Mine T, et al.

Am J Cardiol 2007; 100: 781-6.

Ozcan EE, Guneri S, Akdeniz A, et al. Am Heart J 2007; 154: 539-44.

Briguori C, Airoldi F, D-Andrea D, et al. Circulation. 2007; 115: 1211-7.

In 2008, Maioli and Brar published two larger trials (n=502, 353) of isotonic bicarbon-

ate that showed no benefit.

Maioli M, Toso A, Leoncini M, Et al. J Am Coll Cardiol, 2008; 52:599-604

Brar SS, Yuh-Jer Shen A, Jorgensen MB, Et al. JAMA, 2008; 300:1038-1046

Further confusing the issue were two meta-analysis published in 2009 with conflict-

ing conclusions. Navaneethan published a meta-analysis in AJKD of 12 trials repre-

senting 1,854 patients but did not include Maiolis data. He concluded:

Hydration with sodium bicarbonate decreases the incidence of contrast-

induced nephropathy in comparison to hydration with normal saline with-out a significant difference in need for renal replacement therapy and in-

hospital mortality.


0

8

16

24

32

Crrise>0.5

Crrise25%

Crrise50%

AnuricARF

RENO Trial

IncidenceofRCN(%)

Experimental ControlCONTROL EXPERIMENTAL

0.9% Saline Isotonic bicarbonate

1 mL/kg for

12 hours after

exposure

5mL/kg bolus over 1

hour then 1.5 mL/kg

for 12 hours

600 mg of

NAC q12h x2

following

exposure

2400 mg of IV NAC

with the initial bolus


29/32

Zoungas in the Annals of Internal Medicine analyzed 23 trials of 3,563 patients and

concluded:

The effectiveness of sodium bicarbonate treatment to prevent CIN in high-

risk patients remains uncertain. Earlier reports probably overestimated the

magnitude of any benefit, whereas larger, more recent trials have had neu-tral results.

Navaneethan SD, Singh S, Appasamy S, Et al. Am J Kidney Dis. 2009 53: 617-

27.

Zoungas S, Ninomiya T, Huxley R, Et al. Ann Intern Med. 2009; 151: 631-8.

Acetylcysteine

In 2000 Mark Tepel published a small study on contrast nephropathy in the New Eng-

land Journal of Medicine. It was a small under powered study that had business be-

ing significant, except it was. He randomized 41 patients to acetylcysteine and 43 to

placebo. All patients underwent a contrasted CT scan and 10 developed contrast

nephropathy, 1 from the acetylcysteine group and 9 from the placebo group. Over-

night this changed our approach to contrast nephropathy.

N-acetylcysteine was previously used to restore glutathione levels in Tylenol toxicity

and to break up mucous when given by nebulizer. It is cheap and safe. In rats the

LD50 is 7,888 mg/kg.

Despite the immediate

and widespread adop-

tion of acetylcysteine

there was a deep level of

skepticism in the neph-

rology community.

Within a year and a half

the journals were awash

in acetylcysteine studies

and shortly after that, aslew of meta-analysis

filled the land.

Despite variations in

technique the general



30/32

consensus of the meta-analysis was that acetylcysteine worked. It appeared to reduce

the incidence of contrast nephropathy by roughly 60%. The problem was it didnt

seem to prevent any patient oriented outcome. Acetylcysteine prevents the creatinine

from rising more than 25% or 0.5 mg/dL but it has not been able to prevent dialysis,

major adverse cardiovas-

cular events or death.

This maybe because ace-

tylcysteine alters creati-

nine handling in the

proximal tubule. Acetyl-

cysteine, actually accel-

erates the excretion ofcreatinine resulting in

decreased serum creati-

nine.



31/32

KEY TEACHING POINTS



32/32

Joel M. Topf, MDNephrology248.470.8163

http://pbfluids.com/

Renal Adventures in Imaging

Documents