Effect of viscosity and iodine concentration of nonionic radiographic contrast media on coronary arteriography in patients

Effect of viscosity and iodine concentration nonionic radiographic contrast media on coronary arteriography in patients

of

To assess the influence of viscosity and iodine concentration, three matched and standardized left coronary arteriograms were obtained in 20 patients using iopamidol (lsovue-370), ioversol (Optiray-320), and iohexol (Omnipaque-350). The order of contrast media was randomized and the administration of contrast was double-blinded. Quantitative densitometric angiographic evaluation of the coronary angiograms was performed in addition to independent operator qualitative assessment. The injection volume of iopamidol (5.4 k 1.0 ml) was slightly but significantly less than that of ioversol and iohexol (5.6 2 1.0 ml, 5.7 + 1.0 ml, both p < 0.05). The calculated iodine concentration was also lower for ioversol (1.7 + 0.32 gm) than for iopamidol (1.96 + 0.35 gm) and iohexhol (1.9 + 0.35 gm, both p < 0.05). There were significantly lower contrast syringe injection pressures for ioversol (6.6 + 0.8 atm) than for iopamidol (7.5 +_ 0.9 atm) and iohexol (7.2 f 1.1 atm, both p < 0.05). The quantitative densitometric analysis failed to demonstrate significant differences among the contrast media with respect to image density parameters for any individual agent. All coronary angiograms were deemed of diagnostic quality. The data in this study indicated that although differences in iodine concentration exist among the three agents, operator compensation with more rapid contrast delivery (higher volume) and lower viscosity (lower injection pressure) produced equivalent image opacification during coronary angiography. Given the same incidence of adverse hemodynamic and clinical effects, selection of a low viscosity media theoretically provides an advantage during procedures using small diameter catheters or interventional procedures requiring contrast visualization through reduced catheter lumina. (AM HEART J 1992;123:160.)

Morton J. Kern, MD, Robert A. Roth, RN, Frank V. Aguirre, MD, Glen Beauman, MS,a and Robert Vogel, MD.” St. Louis, MO., and Baltimore,

Optimal imaging during coronary angiography is thought to be best achieved with the highest iodine concentration contrast media formulations.1-4 Because small diameter (5F and 6F) coronary angiographic catheters requiring high flow rates for satisfactory images are now in common clinical use,5, 6 contrast media viscosity and iodine concentration become important factors in selection of radiographic contrast media for use during coronary arteriography. A recently developed nonionic angiographic contrast medium, ioversol (Optiray-320, Mallinckrodt Medical, Inc., St. Louis, MO.), is now available with lower viscosity and iodine concentration com-

From vision

the Cardiology of Cardiology,

Received

Division, University

for publication

Reprint requests: Morton Catheterization Laboratory, at Grand Blvd., St. Louis, 411133652

160

June

St. Louis University of Maryland. 3, 1991;

accepted

J. Kern, MD, Director, St. Louis University MO 63110-0250.

Hospital; July

and ‘the

Di-

17, 1991.

J. Gerard Hospital,

Mudd Cardiac 3635 Vista Ave.

Md.

pared with previously available nonionic formulations, iohexol (Omnipaque-350, Winthrop Pharmaceuticals, New York, N.Y.) and iopamidol (Isovue370, Squibb Pharmaceuticals, New Brunswick, N.J.).2, 7 The purpose of this study was to compare and assess the influence of viscosity and iodine concentration on routine diagnostic coronary angiography using the three clinically available nonionic radiographic contrast media. The hypotheses tested were that (1) diagnostic image opacification and quality obtained during coronary angiography in patients would be equivalent and independent of the small differences in iodine concentration and that (2) the lower the viscosity and iodine concentration, the higher the volume of contrast media needed to achieve the same diagnostic image. The data regarding the efficacy of low iodine, low viscosity contrast media would indicate whether significant differences exist among nonionic contrast media and whether a lower viscosity may be an adjunctive benefit for an-

Volume Number

Table

123 1

media

161

I. Comparative osmolality and viscosity Contrast

Iodine concentration

media

Diatrizoate meglumine or sodium (Renografin-76) Iopamidol (Isovue-370) Iohexol (Omnipaque-350) Ioxaglate Meglumine 39.3 5%and Ioxaglate Sodium 19X0; injection (Hexabrix) Ioversol (Optiray-320) From

of radiocontrast

Viscosity

package

inserts,

Fisher

H. Catalog

of intravascular

Osmolality (mOsm/kg HzO)

Viscosity at

37” C (CPS) 8.4

1940

796

20.9

844

20.4

10.4

600

15.7

7.5

702

11.6

5.8

37% 35%

(370 (350

32%

(370

mg/ml) mg/ml) mg/ml)

32%

(370

mg/ml)

contrast

media.

Radiology

METHODS

The three contrast media tested in the coronary arteriography study wereiopamidol (Isovue-370),iohexol (Omnipaque-350), and ioversol (Optiray-320). In addition, diatrizoate (Renografin-76, Squibb Pharmaceuticals) and ioxaglate (Hexabrix, Mallinckrodt Medical, Inc.) were included for evaluation in an in vitro flow phantom. The physical characteristicsof severalcommonly usedcontrast agents are provided in Table I. In vitro densitometry. To assess the influence of iodine

concentration and viscosity on radiographic imageopacification, an in vitro testing systemwasconstructed in which contrast media were injected through a radiographic arterial phantom containing 2.5, 1.5, and 1.0 mm diameter segments. Contrast agents were injected using a constant pressurethrough a mechanicalpressureinjector using a 12 ml syringe at 37” C through 5F and 6F diagnosticcatheters. The density wasset at 20 cm Hz0 with equivalent absorption, and digital radiographic acquisition in the maskmode subtraction wasemployed. The parameter measuredduring the in vitro study was the mean peak region of interest density. The iodine content of each injection was determined with the following conversion factors: 0.32 gm/ml for ioversol; 0.37 gm/ml for iopamidol; 0.35gm/ml for iohexol; 0.37gm/mI for diatrizoate; and 0.32gm/ml for ioxaglate. In vitro viscosity wasdetermined by calculation of flow resistance of contrast media injected at a constant flow rate through a seriesof coronary angiographiccatheters of different sizes,as previously rep0rted.s Coronary arteriography. After routine diagnosticcoronary angiography was completed, 20 patients received three additional left coronary artery injections using the contrast

20” c (CPS)

37 Sb (370 mg/ml)

giography through small coronary catheters or potentially through the narrowed lumen of interventional systems.

three nonionic

Viscosity at

media. The order of contrast

me-

dia injection was randomized and the administration was double-blinded. The meanageof the patients (18 menand 2 women)was59years (range40 to 76years). Patient medicationsgiven during the study included diazepam(n = 1I), diphenhydramine (II. = 17), ergonovine (n = 3), and nitroglycerin (n = 7). The indications for cardiac catheteriza-

1986;159:561-3;

18.9

and

R&ton

W, et al. Invest

9.4

Radio1

1989;24:52-9.

tion included a history of recent chest pain consistentwith the diagnosisof coronary artery diseaseor evaluation of coronary statusfollowing cardic transplant (three patients, all with normal coronary arteriograms). Coronary artery disease(>70% narrowing in at least one major epicardial vessel)was present in 12 patients. Patients having diagnosticcoronary angiography revealing critical (>80% narrowing in s-2 vessels) coronary artery lesions,unstableangina, recent (~4 days) acute myocardial infarction, or any condition in which additional coronary contrast injections would not be clinically well tolerated were excluded from the study. Left ventriculography precededthe study injections and wasperformed in a right anterior oblique (30-degreeright anterior oblique) view using 42 ml contrast injected at 12 to 14 ml/set. All patients received 3000 units of heparin during the diagnostic study. For the study, syringes (20 ml) were filled with contrast medium by the circulating catheterization laboratory nurse in a sterile fashion. The syringes were code-labeledsuch that the operators and patients were blinded to their contents. The contents of the coded syringe were transferred to the ring-handled coronary control syringe (Namic Inc., GlensFalls, N.Y.) for duplication of the clinical technique of coronary artery injection. Contrast media were hand injected by the sameoperator accordingto the randomization schedule.All contrast media were warmed to room temperature (20’ C). The pressureof contrast medium injection by the operator wasmeasuredwith an in-line pressure transducer describedbelow. The arteria1pressure,electrocardiogram (lead II), and the injection pressureof the contrast syringe were recorded on an optical photographic oscilliscopic physiologic recorder (VR12, Electronics for Medicine, Pleasantville, N.Y.). The volume of contrast medium injected wasnoted by the angiographerafter each injection. The three study contrast injections wereperformed in a shallow20-degreeright anterior oblique projection through a 6F standard 4 cm left Judkins High-F10 catheter (Cordis Corp., Miami, Fla.). The image intensifier position was kept constant for the three injections, The patient wasinstructed to suspend respiration (i.e., not to inspire or expire) during injection. The catheter wasflushed with heparinized normal saline between injections. The time

162

Kern et al.

American

tii

I

January 1992 Heart Journal

iiiiiit

\

d

lniectlon

Pressure

Fig. 1. Syringe injection pressure tracings for the three contrast media during hand injection. The pressure curves are consistent and reproducible for the upstroke, duration, and peak pressure obtained when repeated for the same agent. Pressure x time areas were 130.1, 139.8, and 145.8 for ioversol, iohexol, and iopamidol, respectively.

Table

II. Mean

peak radiographic

5F Catheter Phantom Phantom Phantom 6F Catheter Phantom Phantom Phantom

density

in static arterial

phantoms

Diatrizoate

Iohexol

Ioversol

Ioxaglate

size 2.5 mm size 1.5 mm size 1.0 mm

10.5 6.5 1.2

7.6 5.3 1.0

10.0 6.3 2.4

9.0 5.9 2.0

size 2.5 mm size 1.5 mm size 1.0 mm

9.0 7.4 2.2

7.1 6.2 1.3

9.5 7.6 2.4

7.3 7.0 1.7

6.13

4.75

6.37

5.48

Mean index

between injections ranged from 1 to 3 minutes. The majority of injections (>50%) were more than 1 minute apart. Serial, repeated injections with the same contrast agent during routine studies demonstrated reproducible injection volumes and pressures to within t 10%. Syringe injection pressure measurements. A pressure transducer (Intelliflator, Merit Medical Systems, Salt Lake City, Utah), calibrated with a mercury manometer, was placed in-line between the coronary catheter and the coronary manifold (Namic Inc.) using a four-way, high-flow stopcock. The stopcock was open to both the catheter and transducer. The analog pressure transducer signal was interfaced directly to the photographic oscilliscope recorder. Custom software, provided by Merit Medical Systems, also recorded and stored the continuous output of pressure measured by the transducer (in atmospheres) in a digital format. Both the analog signal and digital printout demonstrated identical values for maximal pressure during validation testing with a standard power contrast injector (Mark II, Medrad Inc., Pittsburgh, Pa.).8 The calibration of the syringe pressure signals was set such that 1 cm deflection was equal to 2 atm. The reproducibility of values was f 0.2 atm for this system. The analog signal of syringe

pressure injections was digitized and planimetered to quantitate the area (pressure x time) difference between injections. The hand injection pressure curves of contrast media injection through a 6F catheter using this system are illustrated in Fig. 1. The pressure required to inject 7 cm3 of contrast at 3 cm3/sec using the power injector set at 150 psi through coronary catheters of varying size has been previously reported8 and was repeated for ioversol (Fig. 2). Ctuantitative angiographic densitometric analysis. Quantitative angiographic densitometric analysis was performed as previously described.Q In brief, following identification of the appropriate coronary injection sequences from the cineangiogram, end-diastolic frames were selected as those frames occurring just prior to the dispersal of contrast in the sinus of Valsalva caused by the onset of systole. In the first cardiac cycle of complete opacification, epicardial vessels were outlined on the tine projector (Vanguard XR35 projector, Vanguard Instrument Corp., Melville, N.Y.) screen with a wax pencil to provide an alignment reference for subsequent cardiac cyles and contrast injections. End-diastolic frames from three consecutive cardiac cycles representing early, mid, and late opacification were

volume

123

Number

1

Viscosity of radiocontrast

media

163

- Room Temp + Body Temp

Omnipaque

Optiray * +

room temp body temp

ATM

e

Room

+

Body Temp

Temp

6

2. Flow resistance(viscosity) of (A) Isovue (iopamidol), (B) Optiray (ioversol), and (C) Omnipaque (iohexol) when injected at room and body temperature (37’ C) through 12 cathetersat a constant flow rate. Shi, Shiley; 8F, 6 French; JR4, Judkins right 4 cm curve; G, guiding catheter; JL4, Judkins left 4 cm curve; AL2, Amplatz left 2 cm curve; AR2, Amplatz right 2 cm curve; ACX 2.5, ACX angioplasty catheter 2.5 mm balloon; ATM, atmospheres.Note the scale0 to 12 ATM (1 ATM = 760 mm Hg = 14.6 psi). Fig.

identified for each contrast injection. Appropriate frames were digitally acquired via a high resolution video camera optically coupled to the Vanguard projector. Each frame was optically magnified (2.4:1) and digitized into a 512 X 512 X 8 bit matrix (ADAC 4100 Plus, ADAC Laboratories, Milpitas, Calif.). Brightness parameters were manually adjusted to optimize the full range of gray scale valuesprior to acquisition of the first frame and werethen fixed at those parametersto assureuniform intensities for eachacquisition. Usingan area of approximately 60 square pixel dimensions, intensity values were determined for portions of two major epicardial vesselsand an adjoining portion of background for eachof the three cardiac cycles.g The qualitative assessment of diagnosticimagequality was madeby two independent expert angiographersblinded to the contrast agents, grading each study injection 1 to 3 (1 = not diagnostic, 2 = diagnostic, 3 = excellent). Data analysis. The variables analyzed included injection volume, grams of iodine administered, peak syringe injection pressure,time-pressurearea of syringe injection,

and quantitative densitometric analysis of the coronary cineangiogram.For both injection, volume and grams of iodine, a comparisonof the contrast media wascarried out in the “within’‘-patient ranks using the Friedman test.l” Other variables were analyzed by analysisof variance using PROC GLM in SAS (SAS Institute, Cary, N.C.).ll Preliminary analysesdid not detect statistically significant carryover or period effects, soin keeping with the original design,the model included patient and drug effects only. Statistical significancewas indicated by a probability (p) value (<0.05) associatedwith the null hypothesis. Data are presented asmean +- standard deviation unlessotherwise indicated. RESULTS In vitro radiographic contrast density. Table II shows the in vitro data for the mean peak radiographic density of diatrizoate, iohexol, ioversol, and ioxaglate. Iopamidol was not evaluated in this phantom.

164

Kern et al.

Table

III. A summary

American

of contrast

media injection

January 1992 Heart Journal

data Range

Parameter

Contrast agent (drug)

Standard deviation

Mean

Low

High

NO.

Volume (ml) Iopamidol Iohexol Ioversol

5.4 5.7b 5.6b

Iopamidol Iohexol Ioversol

7.5a 7.2a 6.6

Pressure area (pressure X time) Iopamidol Iohexol Ioversol

126.68 130.2a

4.0 4.0 4.0

7.0 7.0 7.0

20 20

1.1 0.8

4.8 5.0 5.1

8.4 10.4 7.7

20 20 20

30.2 22.6 24.6

85.4 102.4 67.2

205.7 188.8 159.4

20 20 20

33.1 34.0

135.4 148.5 126.5

255.0 255.0 255.0

18* 18 18

42.1 56.3 26.8

161.5 152.5 134.0

18* 18

1.0

1.0 1.0

20

Pressure 0.9

105.0

Arterial density 194.1

Iopamidol Iohexol Ioversol

193.0 202.2

Density minus background Iopamidol Iohexol Ioversol “p < 0.05 “we”* Ioversol. bp < 0.05 versus Iopamidol. *Number of patients with all three injections

36.9

96.6

34.8 27.9 24.8

101.4

96.3

suitable

for quantitative

densitometric

Radiographic density (opacification) was highest for ioversol and lowest for iohexol. Static image opacification was directly proportional to iodine concentration and inversely proportional to viscosity. The in vitro injection pressures (i.e., flow resistance, Fig. 2) were lowest for ioversol (Optiray) compared with iohexol (Omnipaque) or iopamidol (Isovue). Contrast volume and injection pressure (Table III). The average volume of contrast media administered to patients was 5.6 t- 1.0 ml, with a range from 4 to 7 ml for each injection of iopamidol, iohexol, and ioversol. The volume of injection of iopamidol was slightly, but significantly less (5.4 -+ 1.0 ml) than that of ioversol (5.6 + 1.0 ml) and iohexol (5.7 + 1.0 ml, both p < 0.05). The calculated absolute iodine administered in grams of iodine was lower for ioversol (1.79 + 0.32 gm) than for iopamidol(l.98 + 0.39 gm) and iohexol (1.98 + 0.35 gm) (both p < 0.05), with the overall test for drug effect highly significant (p = 0.0019). There were also significant differences among the contrast media with respect to pressure of injection. Ioversol had a significantly lower injection pressure (6.6 +- 0.8 atm and 105 + 25 units [pressure x time = area units]) than iopamidol (7.5 k 0.9 atm

18

analysis.

and 127 + 30 units) or iohexol (7.2 k 1.1 atm and 130 f 23 units) (both p < 0.05). Angiographic data (Tables III and IV). Quantitative densitometric analysis failed to demonstrate significant differences among the contrast media with respect to the image density parameters for any individual agent. The quality of the coronary angiograms was determined as diagnostic (quality score 22) for all three contrast agents. Side effects of contrast administration were not present with any of the nonionic agents. No patient had an adverse effect as a result of the additional three coronary study injections. DISCUSSION

This study demonstrates that although differences in iodine concentration result in different in vitro image opacification, operator compensation with more rapid contrast delivery (i.e., slightly higher volume) produces identical image quality when images are examined both subjectively and by quantitative densitometric analysis. The slightly higher contrast injection volume was clinically insignificant and undetectable, since lower viscdsity facilitated easier injection. In addition, the equivalent density was

Volume Number

123 1

achieved with ioversol with a significantly lower total iodine content when compared with iohexol and iopamidol. Assessment of the diagnostic quality of angiograms also was consistent with clinical practice, with no appreciable difference among the three contrast agents, as previously reported.3T 12,l3 In vitro studies indicating that the largest differences in contrast media viscosity are most apparent for catheter sizes smaller than 6F8 and the results of the current study suggest that the differences in contrast media viscosity in the catheterization laboratory translate into consistent reductions in hand injection pressure required to force contrast media through small catheters. As hypothesized, greater iodine concentration (volume x absolute iodine content) provided higher degrees of image opacification of the static arterial phantom under conditions of constant pressure injection. The image opacification could be accurately approximated by the iodine concentration divided by the viscosity. The clinical portion of this study also validated this hypothesis, in that a higher volume (with a lower injection pressure) likely delivered an equivalent iodine concentration during coronary angiography to achieve near identical image density. For most coronary angiographic procedures using large-size (7F or 8F) coronary catheters, the viscosity and iodine concentration issues are of little clinical significance. However, reducing the difficulty of contrast injection while maintaining satisfactory image quality appears to have relevance in a number of clinical settings, such as hyperdynamic flow states (e.g., aortic stenosis, hypertension, aortic insufficiency, high cardiac output), distal contrast injection during coronary angioplasty, and coronary visualization during use of interventional equipment with reduced contrast lumina. Images required during procedures using small diameter guiding catheters for coronary angioplasty5 may also be improved by using the lowest viscosity contrast medium available. In addition to viscosity and iodine concentration, a number of other factors have a significant influence on the quality and density of coronary angiography (i.e., images). Patient-related factors include body weight, lung water, positioning of adjacent and superimposed anatomic structures, assuming consistent and optimal functioning of x-ray generation and imaging recording systems. These additional factors were not evaluated, but were assumed to have remained constant for the study period. Clinical significance. Given equivalency of image opacification, hemodynamic and adverse effects, se-

Viscosity of radiocontrast

Table

media

165

IV. Angiographic quality Quality Nondiagnostic I

Iopamidol Iohexol Ioversol

score

Diagnostic 2

0 0 0

20 19 19

Excellent 3 0 1 1

lection of a low viscosity media theoretically provides an advantage during procedures using small diameter catheters or interventional procedures requiring contrast visualization through reduced catheter lumina. The authors Donna Sander

thank Steve Taylor of Merit for manuscript preparation.

Medical

Systems

and

REFERENCES

1. Gertz EW, Wisneski JA, Chiu D, Akin JR, Hu C. Clinical sunerioritv of a new nonionic contrast agent (iopamidol) for cardiac angiography. J Am Co11 Cardio11985;5:250-8. JW. Wieland J. Davis CA. Giles BD. Passione D. Rav 2. Hirshfeld MB, Ripley NS. Hemodynamic and electrocardiographic effects of ioversol during cardiac angiography: comparison with iopamidol and diatrizoate. Invest Radio1 1989;24:138-44. 3. Matthai WH, Hirshfeld JW Jr. Choice of contrast agents for cardiac angiography: review and recommendations based on clinically important distinctions. Cathet Cardiovasc Diagn 1991;22:278-89. 4. Fischer HW, Thompson KR. Contrast media in coronary arteriography: a review. Invest Radio1 1978;13:450-9. 5. Kern MJ, Cohen M, Tailey JD, Litvack F, Serota H, Aguirre F, Deligonul U, Bashore TM. Early ambulation after 5 French diagnostic cardiac catheterization: results of a multicenter trial. J Am Co11 Cardiol 1990:15:1475-583. 6. Kern MJ, Talley JD, Deligonul U, Serota H, Aguirre F, Gudipati C, Ring M, Joseph A, Yussman ZA, Kulick D, Salinger M. Preliminary experience with 5 French diagnostic catheters as guiding catheters for coronary angioplasty. Cathet Cardiovasc Diagn 1991;22:60-3. 7. Thomson KR, Evil1 CA, Fritzsche J, Benness GT. Comparison of iopamidol, ioxaglate and diatrizoate during coronary arteriography in dogs. Invest Radio1 1980;3:234-41. 8. Roth R, Akin M, Deligonul U, Kern MJ. Influence of radiographic contrast media viscosity to flow through coronary angiographic catheters. Cathet Cardiovasc Diagn 1991;22:290-4. 9. Mancini GBJ, Simon SB, McGillem MJ, LeFree MT, Freidman HZ, Vogel RA. Automated quantitative coronary arteriography: in vivo morphologic and physiologic validation of a rapid method utilizing digital angiography. Circulation 1987; 75~452-60. 10. Conover WJ. Practical nonparametric statistics. 2nd ed. New York: John Wiley & Sons, Inc, 1980. 11. SASISTAT User’s Guide. version 6.4th ed. vol. 2. Gary, NC: SAS Institute, Inc, 1989. 12. Piscion F, Foxaccio A, Santinelli V, DePaola M, Villari B, Spinazzi A, Condorelli M, Chiariello M. Are ioxaglate and iopamidol equally safe and well tolerated in cardiac angiography? A randomized, double-blind clinical study. AM HEART J

1990;120:1130-6.