1978, British Journal of Radiology, 51,106-110
The effect of osmotic diuresis on urinary iodine concentration using contrast media of differing osmolality By Judith A.W.Webb, B.Sc, M.R.C.P., F.R.C.R., I. Kelsey Fry, D.M., F.R.C.P., F.R.C.R., W. R. Cattell, M.D., F.R.C.P., Barbara Cummack, M.Sc, and Sandra E. Jewell, H.D.S.R. Departments of Radiology and Nephrology, St. Bartholomew's Hospital, London, E.C.1 {Received June, 1977 and in revised form July, 1977)
urine after the injection of doses containing an equal amount of iodine is higher with iocarmate and metrizamide than with iothalamate (Hilal, 1970; Benness and Glazer, 1973; Golman and Almen, 1973; 1976) and higher with metrizamide than with iocarmate (Golman and Almen, 1976). However, all these studies were carried out in animals which were concentrating their urine normally. In clinical practice urography with current contrast media is usually satisfactory in patients who are concentrating their urine. Difficulties occur when the kidneys are not concentrating well, especially when there is overt renal failure. In many such patients an endogenous osmotic diuresis contributes to the failure to concentrate urine. Since this is the type of patient who might benefit from the use of contrast media producing less osmotic diuresis, we decided to compare the excretion of sodium iocarmate and metrizamide with that of sodium iothalamate in animals with varying degrees of induced osmotic diuresis.
ABSTRACT
The urinary iodine concentrations of a monomer (sodium iothalamate), a dimer (sodium iocarmate) and a non-ionic compound (metrizamide) have been compared in dogs with varying levels of solute excretion. All the animals were undergoing maximal antidiuresis. In dogs with normal solute excretion, metrizamide and iocarmate produced higher urinary iodine concentrations than iothalamate. There was no significant difference between metrizamide and iocarmate. With increasing levels of solute excretion, the differences between the compounds were reduced. These findings suggest that contrast media of reduced osmolality are unlikely to have a special place in advanced renal failure.
The excretion of urographic contrast media is accompanied by an osmotic diuresis. This is an important factor limiting the concentration of iodine in the urine (Benness, 1965; Cattell et ah, 1967; Saxton, 1969). Recently, contrast media have been developed which have a lower osmolality in solution than the compounds currently in use. Two such compounds are sodium iocarmate, a dimer of sodium iothalamate and metrizamide, a non-ionic compound. Table I compares the osmolality per mg iodine in solution of these compounds with that of the monomer, sodium iothalamate. When the amount of solute required to give an equivalent dose of iodine is considered, it can be seen that the osmotic diuretic effect should be less with iocarmate than iothalamate, and less with metrizamide than iocarmate. Animal studies have confirmed that the concentration of iodine in the
MATERIALS AND METHODS
Studies were carried out in mongrel dogs of either sex, weighing 13.5-22.5 kg, fasted overnight but without fluid restriction. All experiments were carried out under sodium pentobarbitone anaesthesia (Nembutal 20-30 mg/kg i.v.) with the dogs
TABLE I OSMOLALITY COMPARISONS OF CONTRAST MEDIA
Contrast
Type of compound
Dose of contrast given
Osmolality of solution (/u.osmol/mg I in solution)
mg I/kg
solute/kg (/xosmol)
1. Sodium iothalamate
Monomer
5.18
500
2590
2. Sodium iocarmate
Dimer
3.40
500
1700 (65%)*
3. Metrizamide
Non-ionic
1.62
500
810 (31%)*
*Figures in brackets indicate solute load as a percentage of the solute load given with the monomer. 106
FEBRUARY 1978
The effect of osmotic diuresis on urinary iodine concentration breathing spontaneously through an endotracheal tube. Through a suprapubic laparotomy, both ureters were catheterized at pelvic brim level with soft polythene tubing (Portex umbilical cannula FG6, internal diameter 1.4 mm). Blood samples were taken through a catheter in the inferior vena cava. Blood pressure was monitored via a catheter in the femoral artery connected to a Statham transducer. In order to obtain comparable resting rates of urine flow, all animals were studied under conditions of maximal antidiuresis. This was achieved by giving 2 /xg 1-deamino-8-D-arginine vasopressin (DDAVP) intravenously. After 45 minutes, urine was collected for the measurement of flow rate and osmolality. Dogs were divided into three groups, given mannitol as follows: Group A. No mannitol. Group B. Loading dose of 1.5 mOsm/kg of 5% mannitol followed by a constant infusion to achieve a steady moderate, osmotic diuresis (0.5-0.8 ml/min). Group C. Loading dose of 3.0 mOsm/kg of 5% mannitol followed by a constant infusion to achieve a steady large osmotic diuresis (2.2-3.2 ml/min). After 45 minutes stabilization on the above infusions, three 15-minute control urine collections were made. Each group was then divided into subgroups 1, 2 and 3 which were given sodium iothalamate, sodium iocarmate and metrizamide respectively. The dose for all three contrast media was 500 mg iodine/kg injected over 30 seconds. Further urine collections were obtained at 0-3, 3-10, 10-20, 20-30, 30-45 and 45-60 minutes after the injection of contrast. In the analysis of the results the 0-3 minute urine collection has been excluded as this is affected by the washout of the renal tubules, collecting systems and ureteric catheters (Morales et ah, 1950;
Chinard, 1955). Heparinized blood samples were taken at the beginning and end of each urine collection. The concentration of contrast medium in plasma and urine was measured by a spectrophotometric method similar to that described by Purkiss et al. (1968). The osmolality of plasma and urine was measured on an advanced osmometer. Statistical analysis of the results was by Student's t test. RESULTS
Comparable basal rates of urine flow and solute excretion were obtained in 36 dogs. Twelve dogs were studied at each level of solute excretion. These groups of 12 dogs were divided into three subgroups of four dogs, and each subgroup received a different contrast medium (Table II).
Group A. DDAVP only (Table III) In all urine collections the iodine concentrations with iocarmate and metrizamide were significantly higher than those with iothalamate (p^0.05). There was no significant difference between the iodine concentrations with metrizamide and iocarmate at any time. The mean urine flow rates with metrizamide were less than those with iocarmate, which were in turn less than those with iothalamate. However, a significant difference was only consistently seen between metrizamide and iothalamate. Group B. DDA VP/moderate mannitol diuresis (Table IV) The urinary iodine concentrations with iocarmate were significantly higher than those with iothalamate up to 20 minutes. The iodine concentrations with metrizamide were significantly higher than those
TABLE II MEAN CONTROL URINE FLOW AND SOLUTE EXCRETION RATES GROUPS (12 DOGS IN EACH)
B. DDAVP/moderate mannitol diuresis*
A. DDAVP* Subgroups given different contrast media (four dogs in each)
C. DDAVP/large mannitol diuresis*
Urine flow rate (ml/min)
Solute excretion (^osm/min)
Urine flow rate (ml/min)
Solute excretion (jtiosm/min)
Urine flow rate (ml/min)
Solute excretion (/iosm/min)
1. Sodium iothalamate
0.14
351
0.72
657
2.63
1520
2. Sodium iocarmate
0.16
327
0.74
761
2.84
1604
3. Metrizamide
0.16
340
0.70
755
2.52
1636
*There is no significant difference (p^0.05) in urine flow rate or solute excretion between subgroups 1, 2 and 3 within Groups A, B and C. 107
VOL.
51, No. 602 Judith A. W. Webb, I. Kelsey Fry, W. R. Cattell, Barbara Cummack and Sandra E. Jewell TABLE III
TABLE IV
GROUP A. DDAVP ONLY (a) MEAN URINE CONTRAST CONCENTRATION (mg I/ml)
GROUP B. DDAVP/MODERATE MANNITOL DIURESIS (a) MEAN URINE CONTRAST CONCENTRATION (mg I/ml)
Collection period (min.) 1. lothalamate
Collection Period (min.)
3-10 10-20 20-30 30-45 45-60
3-10 10-20 20-30 30-45 45-60
63
79
91
105
111
1. Iothalamate
66
58
49
44
37
2. locarmate
128
136
147
145
139
2. Iocarmate
96
75
58
51
42
3. Metrizamide
155
145
159
160
146
3. Metrizamide
123
82
65
53
45
1.v.2. "1
0.001 0.001 0.01
0.01
0.02
1.v.2. ~)
0-01
0-05
NS
NS
NS
1.v.3. UsS
0.01
0.01
1.V.3.
0-01
001
001
NS
NS
NS
NS
NS
NS
NS
NS
NS
2.v.3. J
0.001 0.001 0.05 NS
NS
U
^
2.v.3. J
Benness, 1970; Benness and Glazer, 1973; Golman and Almen, 1976). They also confirm the findings of Golman and Almen (1973; 1976) in rabbits that the non-ionic compound metrizamide produces higher urinary iodine concentrations and less increase in DISCUSSION Our results in dogs with no mannitol infusion con- urine flow rate than iothalamate. The findings differ firm reports that in dehydrated animals the dimer from those'of Evill and Benness (1974) who found no iocarmate produces higher urinary iodine concentra- difference in dogs between metrizamide and iothalations than the monomer iothalamate (Hilal, 1970; mate. However, they were using a smaller dose of 108
FEBRUARY
1978
The effect of osmotic diuresis on urinary iodine concentration contrast medium (125 mg 1/kg) which would produce a smaller diuresis. Thus, Golman and Almen (1973; 1976) have shown that the differences between contrast media of different osmolality are less with lower doses. We did not show any significant difference in urinary iodine concentration between iocarmate and metrizamide although it had been anticipated that the different solute load given with the compounds (Table I) would be reflected in the final urine. Our finding differs from that of Golman and Almen (1976) who obtained higher concentrations with metrizamide than iocarmate in rabbits. This discrepancy raises an important point. To explain it we must consider what is happening not only in the proximal tubule but also in the distal tubule, where the level of antidiuretic hormone (ADH) activity modifies the degree of osmotic diuresis (Zak et aL, 1954; Orloff et al, 1957). A nonreabsorbable solute produces an osmotic diuresis by reducing the amount of water reabsorbed in the proximal tubule. More of the filtered water is thus delivered to the distal tubule and so to the urine. The amount of water entering the distal tubule depends on the load of non-reabsorbed solute but the proportion reaching the urine also depends on the amount reabsorbed in the distal tubule under the influence of ADH. If the subject is dehydrated and reabsorbing water maximally the distal tubule may be able to reabsorb much of the additional water entering from the proximal tubule. Thus reabsorption of water distally may partially or completely obscure the effect of varying degrees of proximal osmotic diuresis. The level of ADH activity is therefore an important factor to be taken into account when comparing contrast media of different osmolality (Golman and Almen, 1976). It seems likely therefore that the difference between our results and those of Golman and Almen (1976) can be accounted for by the differences in the two experimental models. Their rabbits were not fluid restricted and had preliminary urinary osmolalities of "over 600 mOsm/kg". In our dogs there was a much greater degree of initial antidiuresis—the mean preliminary urine osmolality being 2050 mOsm/kg. For the same reasons, the differences we observed between compounds of differing osmolality may be more evident in man in whom urinary concentrating capacity is less than in dogs. Thus the maximum urinary osmolality in man is approximately 1200 mOsm/kg (Pitts, 1974) and in patients coming to urography the urinary osmolality is considerably lower than this. The studies with mannitol infusion were designed
to simulate the endogenous solute diuresis of renal failure. In renal failure it is believed that the nephron population in the kidney is reduced and that the normal daily load of water and solute is all carried by the few remaining nephrons (Bricker, 1969). These nephrons are thus hyperperfused or subject to a constant endogenous solute diuresis. This contributes to the impaired urine concentrating ability of patients with renal failure. In the dogs given a mannitol infusion sufficient to produce a moderate osmotic diuresis (Group B) solute excretion was increased to rather more than twice normal. This is equivalent to halving the nephron population so that each remaining nephron must carry double the solute load. In the dogs with a large osmotic diuresis (Group C) solute excretion was increased to about five times normal—equivalent to an 80% reduction of nephron population. Under conditions of moderate osmotic diuresis (Group B) there were significant differences in urinary iodine concentrations between iocarmate and iothalamate and between metrizamide and iothalamate. These differences were proportionally less than in dogs with normal solute excretion. As in the dogs with normal solute excretion, differences in iodine concentration between iocarmate and metrizamide were not significant. When the baseline solute excretion was greatly increased (Group C) the differences between the compounds were almost completely lost. These results show that any benefit obtained from using the compounds of reduced osmolality is decreased as the resting level of solute excretion increases. This must raise doubts about the value of dimers and non-ionic compounds in patients with renal failure in whom the endogenous osmotic diuresis is likely to be nearer that in Group C than in Group B. In conclusion this study has confirmed the advantages to be expected from contrast media with reduced osmotic properties under conditions of normal solute excretion. It has also highlighted the importance of considering resting solute excretion and the degree of ADH activity when comparing the effects of contrast media of differing osmolality. The results indicate that contrast media of reduced osmolality are only likely to be of value in patients with impaired renal function when endogenous solute excretion is not excessive. ACKNOWLEDGMENTS
We wish to thank May and Baker Ltd. and Nyegaard and Co. A/S for supplying the contrast media and for financial assistance, and Ferring Pharmaceuticals for supplying DDAVP.
109
VOL.
51, No. 602 Judith A. W. Webb, I. Kelsey Fry, W. R. Cattell, Barbara Cummack and Sandra E. Jewell
We also wish to thank Mr. Andrew J. Nunn of the M.R.C., T.B. and Chest Diseases Unit, Brompton Hospital, London, S.W.6, for the statistical analysis of the results.
urine following intravenous injection of an ionic and a non-ionic contrast medium in rabbits. Acta radiologica, Supplementum 335, 312-322. 1976. Metrizamide in experimental urography. VI— Effects of renal contrast media on urinary solutes. Acta
REFERENCES BENNESS, G. T., 1965. Double dose urography. Journal of the
Pharmacologica et Toxicologica,
38,120-136.
HILAL, S. K., 1970. Trends in preparation of new angiographic contrast media with special emphasis on polymeric derivatives. Investigative Radiology, 5,458-468.
College of Radiologists of Australasia, 9, 78-82.
1970. Urography with the iothalamate dimer and trimer. Australasian Radiology, 14, 416-418. BENNESS, G. T., and GLAZER, M., 1973. Urographic con-
MORALES, P. A., CROWDER, C. H., FISHMAN, A. P., MAXWELL, M. H., and GOMEZ, D. G., 1950. Measurement and
trast agents. Comparison of sodium and methylglucamine salts of iothalamate monomer and dimer. Clinical Radiology, 24, 445-448.
significance of urinary appearance time in the dog. American Journal of Physiology, 163, 454—460. ORLOFF, J., WAGNER, H. N., and DAVIDSON, D. G., 1957.
BRICKER, N. S., 1969. On the meaning of the intact
nephron hypothesis. American Journal of Medicine, 46, 1-11.
The effect of variations in solute excretion and vasopressin dosage on the excretion of water in the dog. Journal of Clinical Investigation, 37, 458-464. PITTS, R. F., 1974. In Physiology of the Kidney and Body Fluids, p.114 (Year Book Medical Publishers)."
CATTELL, W. R., KELSEY FRY, I., SPENCER, A. G., and PUR-
KISS, P., 1967. Excretion urography. I—Factors determining the excretion of Hypaque. British Journal of Radiology, 40, 561-580. CHINARD, F. P., 1955. Renal dead space as a function of urine flow in the anaesthetized dog. American Journal of Physiology, 180, 620-622.
PURKISS, P., LANE, R. D., CATTELL, W. R., KELSEY FRY, I.,
and SPENCER, A. G., 1968. Estimation of sodium diatrizoate by absorption spectrophotometry. Investigative Radiology, 3,271-274. SAXTON, H. M., 1969. Review article: Urography. British Journal of Radiology, 42, 321-346.
EVILL, C. A., and BENNESS, G. T., 1974. Metrizamide as a
non-ionic urographic agent. A comparison with sodium iothalamate and its dimer. Investigative Radiology, 9, 434-437.
ZAK,
G. A., BRUN, C , and SMITH, H. W., 1954. The
mechanism of formation of osmotically concentrated urine during the antidiuretic state. Journal of Clinical Investigation, 55,1064-1074.
GOLMAN, K., and ALMEN, T., 1973. Metrizamide in experi-
mental urography. I—Iodine concentration andflowof
Book review Radiographic Contrast Agents. Edited by Roscoe E. Miller and Jovitas Skucas, pp. xv + 515, illus., 1977 (BaltimoreLondon-Tokyo, University Park Press), £21 -25. The two editors have written the opening six chapters— everything about barium sulphate. This first third of their book is superb, and I would like to quote extensively from their summary: "All so-called 'colloidal' barium sulfate preparations for radiology contain additives or contaminants in small amounts that, through surface action, affect surface tension, hydration, particle charge, and pH. These physical properties markedly influence particle size, flocculation, suspension, and viscosity. These latter properties, in turn, largely determine the suspension's ability to settle, flow, bubble, form blood clots, and make lesion-like artifacts; to coat and stick to the mucosa, form various film thicknesses, crack, and achieve and maintain various degrees of density; and to be recorded on the radiograph. "Most barium sulfate additives are not listed on the container, are almost invariably present, and form a coat that is adsorbed on the aggregate particles' surface. This covering may consist of many substances in a variety of patterns with multiple degrees of ionic charge and hydration. "The radiologist, knowingly or unknowingly, purposefully or haphazardly, changes the viscosity, density, surface tension, and film-forming qualities of his suspensions by selecting different brands of barium sulfate and by choosing the concentration and mixing methods in their preparation . . . "By understanding the basic physical, chemical, and
colloidal properties of barium sulfate suspensions and by taking advantage of additives already present, the clinical radiologist can control the performance of his barium sulfate suspensions through intelligent modification. He can do this quite easily by altering rationally his preparation methods and by judicious combination of different compatible barium sulfate compounds in various proportions. With their multiple additives, the radiologist can then use these combined compounds under local conditions in different concentrations for numerous types and methods of examination with superior results . . . "These seemingly simple, but fundamental procedures, let alone the theory behind them, have not been suggested previously in the radiological, medical, patent, or advertising literature." This work goes a long way toward correcting our ignorance about the worst labelled substance in clinical medicine. The rest of the book contains expert contributions by 17 authors on hepatobiliary, pancreatic, genitourinary, angiographic, bronchographic, myelographic and miscellaneous contrast agents. There is afinalsection on paediatrics. These chapters are all clearly presented, and I thought the biliary section particularly good. Inevitably authors often have to go over ground already covered in other publications; the editors have set their contributors an unfairly hard task in matching the knowledgeable enthusiasm and originality of the opening barium chapters. The book is worth buying for these alone, and also as a compact practical reference work on all contrast media.
110
THOMAS SHERWOOD.
The effect of osmotic diuresis on urinary iodine concentration using contrast media of differing osmolality By Judith A.W.Webb, B.Sc, M.R.C.P., F.R.C.R., I. Kelsey Fry, D.M., F.R.C.P., F.R.C.R., W. R. Cattell, M.D., F.R.C.P., Barbara Cummack, M.Sc, and Sandra E. Jewell, H.D.S.R. Departments of Radiology and Nephrology, St. Bartholomew's Hospital, London, E.C.1 {Received June, 1977 and in revised form July, 1977)
urine after the injection of doses containing an equal amount of iodine is higher with iocarmate and metrizamide than with iothalamate (Hilal, 1970; Benness and Glazer, 1973; Golman and Almen, 1973; 1976) and higher with metrizamide than with iocarmate (Golman and Almen, 1976). However, all these studies were carried out in animals which were concentrating their urine normally. In clinical practice urography with current contrast media is usually satisfactory in patients who are concentrating their urine. Difficulties occur when the kidneys are not concentrating well, especially when there is overt renal failure. In many such patients an endogenous osmotic diuresis contributes to the failure to concentrate urine. Since this is the type of patient who might benefit from the use of contrast media producing less osmotic diuresis, we decided to compare the excretion of sodium iocarmate and metrizamide with that of sodium iothalamate in animals with varying degrees of induced osmotic diuresis.
ABSTRACT
The urinary iodine concentrations of a monomer (sodium iothalamate), a dimer (sodium iocarmate) and a non-ionic compound (metrizamide) have been compared in dogs with varying levels of solute excretion. All the animals were undergoing maximal antidiuresis. In dogs with normal solute excretion, metrizamide and iocarmate produced higher urinary iodine concentrations than iothalamate. There was no significant difference between metrizamide and iocarmate. With increasing levels of solute excretion, the differences between the compounds were reduced. These findings suggest that contrast media of reduced osmolality are unlikely to have a special place in advanced renal failure.
The excretion of urographic contrast media is accompanied by an osmotic diuresis. This is an important factor limiting the concentration of iodine in the urine (Benness, 1965; Cattell et ah, 1967; Saxton, 1969). Recently, contrast media have been developed which have a lower osmolality in solution than the compounds currently in use. Two such compounds are sodium iocarmate, a dimer of sodium iothalamate and metrizamide, a non-ionic compound. Table I compares the osmolality per mg iodine in solution of these compounds with that of the monomer, sodium iothalamate. When the amount of solute required to give an equivalent dose of iodine is considered, it can be seen that the osmotic diuretic effect should be less with iocarmate than iothalamate, and less with metrizamide than iocarmate. Animal studies have confirmed that the concentration of iodine in the
MATERIALS AND METHODS
Studies were carried out in mongrel dogs of either sex, weighing 13.5-22.5 kg, fasted overnight but without fluid restriction. All experiments were carried out under sodium pentobarbitone anaesthesia (Nembutal 20-30 mg/kg i.v.) with the dogs
TABLE I OSMOLALITY COMPARISONS OF CONTRAST MEDIA
Contrast
Type of compound
Dose of contrast given
Osmolality of solution (/u.osmol/mg I in solution)
mg I/kg
solute/kg (/xosmol)
1. Sodium iothalamate
Monomer
5.18
500
2590
2. Sodium iocarmate
Dimer
3.40
500
1700 (65%)*
3. Metrizamide
Non-ionic
1.62
500
810 (31%)*
*Figures in brackets indicate solute load as a percentage of the solute load given with the monomer. 106
FEBRUARY 1978
The effect of osmotic diuresis on urinary iodine concentration breathing spontaneously through an endotracheal tube. Through a suprapubic laparotomy, both ureters were catheterized at pelvic brim level with soft polythene tubing (Portex umbilical cannula FG6, internal diameter 1.4 mm). Blood samples were taken through a catheter in the inferior vena cava. Blood pressure was monitored via a catheter in the femoral artery connected to a Statham transducer. In order to obtain comparable resting rates of urine flow, all animals were studied under conditions of maximal antidiuresis. This was achieved by giving 2 /xg 1-deamino-8-D-arginine vasopressin (DDAVP) intravenously. After 45 minutes, urine was collected for the measurement of flow rate and osmolality. Dogs were divided into three groups, given mannitol as follows: Group A. No mannitol. Group B. Loading dose of 1.5 mOsm/kg of 5% mannitol followed by a constant infusion to achieve a steady moderate, osmotic diuresis (0.5-0.8 ml/min). Group C. Loading dose of 3.0 mOsm/kg of 5% mannitol followed by a constant infusion to achieve a steady large osmotic diuresis (2.2-3.2 ml/min). After 45 minutes stabilization on the above infusions, three 15-minute control urine collections were made. Each group was then divided into subgroups 1, 2 and 3 which were given sodium iothalamate, sodium iocarmate and metrizamide respectively. The dose for all three contrast media was 500 mg iodine/kg injected over 30 seconds. Further urine collections were obtained at 0-3, 3-10, 10-20, 20-30, 30-45 and 45-60 minutes after the injection of contrast. In the analysis of the results the 0-3 minute urine collection has been excluded as this is affected by the washout of the renal tubules, collecting systems and ureteric catheters (Morales et ah, 1950;
Chinard, 1955). Heparinized blood samples were taken at the beginning and end of each urine collection. The concentration of contrast medium in plasma and urine was measured by a spectrophotometric method similar to that described by Purkiss et al. (1968). The osmolality of plasma and urine was measured on an advanced osmometer. Statistical analysis of the results was by Student's t test. RESULTS
Comparable basal rates of urine flow and solute excretion were obtained in 36 dogs. Twelve dogs were studied at each level of solute excretion. These groups of 12 dogs were divided into three subgroups of four dogs, and each subgroup received a different contrast medium (Table II).
Group A. DDAVP only (Table III) In all urine collections the iodine concentrations with iocarmate and metrizamide were significantly higher than those with iothalamate (p^0.05). There was no significant difference between the iodine concentrations with metrizamide and iocarmate at any time. The mean urine flow rates with metrizamide were less than those with iocarmate, which were in turn less than those with iothalamate. However, a significant difference was only consistently seen between metrizamide and iothalamate. Group B. DDA VP/moderate mannitol diuresis (Table IV) The urinary iodine concentrations with iocarmate were significantly higher than those with iothalamate up to 20 minutes. The iodine concentrations with metrizamide were significantly higher than those
TABLE II MEAN CONTROL URINE FLOW AND SOLUTE EXCRETION RATES GROUPS (12 DOGS IN EACH)
B. DDAVP/moderate mannitol diuresis*
A. DDAVP* Subgroups given different contrast media (four dogs in each)
C. DDAVP/large mannitol diuresis*
Urine flow rate (ml/min)
Solute excretion (^osm/min)
Urine flow rate (ml/min)
Solute excretion (jtiosm/min)
Urine flow rate (ml/min)
Solute excretion (/iosm/min)
1. Sodium iothalamate
0.14
351
0.72
657
2.63
1520
2. Sodium iocarmate
0.16
327
0.74
761
2.84
1604
3. Metrizamide
0.16
340
0.70
755
2.52
1636
*There is no significant difference (p^0.05) in urine flow rate or solute excretion between subgroups 1, 2 and 3 within Groups A, B and C. 107
VOL.
51, No. 602 Judith A. W. Webb, I. Kelsey Fry, W. R. Cattell, Barbara Cummack and Sandra E. Jewell TABLE III
TABLE IV
GROUP A. DDAVP ONLY (a) MEAN URINE CONTRAST CONCENTRATION (mg I/ml)
GROUP B. DDAVP/MODERATE MANNITOL DIURESIS (a) MEAN URINE CONTRAST CONCENTRATION (mg I/ml)
Collection period (min.) 1. lothalamate
Collection Period (min.)
3-10 10-20 20-30 30-45 45-60
3-10 10-20 20-30 30-45 45-60
63
79
91
105
111
1. Iothalamate
66
58
49
44
37
2. locarmate
128
136
147
145
139
2. Iocarmate
96
75
58
51
42
3. Metrizamide
155
145
159
160
146
3. Metrizamide
123
82
65
53
45
1.v.2. "1
0.001 0.001 0.01
0.01
0.02
1.v.2. ~)
0-01
0-05
NS
NS
NS
1.v.3. UsS
0.01
0.01
1.V.3.
0-01
001
001
NS
NS
NS
NS
NS
NS
NS
NS
NS
2.v.3. J
0.001 0.001 0.05 NS
NS
U
^
2.v.3. J
Benness, 1970; Benness and Glazer, 1973; Golman and Almen, 1976). They also confirm the findings of Golman and Almen (1973; 1976) in rabbits that the non-ionic compound metrizamide produces higher urinary iodine concentrations and less increase in DISCUSSION Our results in dogs with no mannitol infusion con- urine flow rate than iothalamate. The findings differ firm reports that in dehydrated animals the dimer from those'of Evill and Benness (1974) who found no iocarmate produces higher urinary iodine concentra- difference in dogs between metrizamide and iothalations than the monomer iothalamate (Hilal, 1970; mate. However, they were using a smaller dose of 108
FEBRUARY
1978
The effect of osmotic diuresis on urinary iodine concentration contrast medium (125 mg 1/kg) which would produce a smaller diuresis. Thus, Golman and Almen (1973; 1976) have shown that the differences between contrast media of different osmolality are less with lower doses. We did not show any significant difference in urinary iodine concentration between iocarmate and metrizamide although it had been anticipated that the different solute load given with the compounds (Table I) would be reflected in the final urine. Our finding differs from that of Golman and Almen (1976) who obtained higher concentrations with metrizamide than iocarmate in rabbits. This discrepancy raises an important point. To explain it we must consider what is happening not only in the proximal tubule but also in the distal tubule, where the level of antidiuretic hormone (ADH) activity modifies the degree of osmotic diuresis (Zak et aL, 1954; Orloff et al, 1957). A nonreabsorbable solute produces an osmotic diuresis by reducing the amount of water reabsorbed in the proximal tubule. More of the filtered water is thus delivered to the distal tubule and so to the urine. The amount of water entering the distal tubule depends on the load of non-reabsorbed solute but the proportion reaching the urine also depends on the amount reabsorbed in the distal tubule under the influence of ADH. If the subject is dehydrated and reabsorbing water maximally the distal tubule may be able to reabsorb much of the additional water entering from the proximal tubule. Thus reabsorption of water distally may partially or completely obscure the effect of varying degrees of proximal osmotic diuresis. The level of ADH activity is therefore an important factor to be taken into account when comparing contrast media of different osmolality (Golman and Almen, 1976). It seems likely therefore that the difference between our results and those of Golman and Almen (1976) can be accounted for by the differences in the two experimental models. Their rabbits were not fluid restricted and had preliminary urinary osmolalities of "over 600 mOsm/kg". In our dogs there was a much greater degree of initial antidiuresis—the mean preliminary urine osmolality being 2050 mOsm/kg. For the same reasons, the differences we observed between compounds of differing osmolality may be more evident in man in whom urinary concentrating capacity is less than in dogs. Thus the maximum urinary osmolality in man is approximately 1200 mOsm/kg (Pitts, 1974) and in patients coming to urography the urinary osmolality is considerably lower than this. The studies with mannitol infusion were designed
to simulate the endogenous solute diuresis of renal failure. In renal failure it is believed that the nephron population in the kidney is reduced and that the normal daily load of water and solute is all carried by the few remaining nephrons (Bricker, 1969). These nephrons are thus hyperperfused or subject to a constant endogenous solute diuresis. This contributes to the impaired urine concentrating ability of patients with renal failure. In the dogs given a mannitol infusion sufficient to produce a moderate osmotic diuresis (Group B) solute excretion was increased to rather more than twice normal. This is equivalent to halving the nephron population so that each remaining nephron must carry double the solute load. In the dogs with a large osmotic diuresis (Group C) solute excretion was increased to about five times normal—equivalent to an 80% reduction of nephron population. Under conditions of moderate osmotic diuresis (Group B) there were significant differences in urinary iodine concentrations between iocarmate and iothalamate and between metrizamide and iothalamate. These differences were proportionally less than in dogs with normal solute excretion. As in the dogs with normal solute excretion, differences in iodine concentration between iocarmate and metrizamide were not significant. When the baseline solute excretion was greatly increased (Group C) the differences between the compounds were almost completely lost. These results show that any benefit obtained from using the compounds of reduced osmolality is decreased as the resting level of solute excretion increases. This must raise doubts about the value of dimers and non-ionic compounds in patients with renal failure in whom the endogenous osmotic diuresis is likely to be nearer that in Group C than in Group B. In conclusion this study has confirmed the advantages to be expected from contrast media with reduced osmotic properties under conditions of normal solute excretion. It has also highlighted the importance of considering resting solute excretion and the degree of ADH activity when comparing the effects of contrast media of differing osmolality. The results indicate that contrast media of reduced osmolality are only likely to be of value in patients with impaired renal function when endogenous solute excretion is not excessive. ACKNOWLEDGMENTS
We wish to thank May and Baker Ltd. and Nyegaard and Co. A/S for supplying the contrast media and for financial assistance, and Ferring Pharmaceuticals for supplying DDAVP.
109
VOL.
51, No. 602 Judith A. W. Webb, I. Kelsey Fry, W. R. Cattell, Barbara Cummack and Sandra E. Jewell
We also wish to thank Mr. Andrew J. Nunn of the M.R.C., T.B. and Chest Diseases Unit, Brompton Hospital, London, S.W.6, for the statistical analysis of the results.
urine following intravenous injection of an ionic and a non-ionic contrast medium in rabbits. Acta radiologica, Supplementum 335, 312-322. 1976. Metrizamide in experimental urography. VI— Effects of renal contrast media on urinary solutes. Acta
REFERENCES BENNESS, G. T., 1965. Double dose urography. Journal of the
Pharmacologica et Toxicologica,
38,120-136.
HILAL, S. K., 1970. Trends in preparation of new angiographic contrast media with special emphasis on polymeric derivatives. Investigative Radiology, 5,458-468.
College of Radiologists of Australasia, 9, 78-82.
1970. Urography with the iothalamate dimer and trimer. Australasian Radiology, 14, 416-418. BENNESS, G. T., and GLAZER, M., 1973. Urographic con-
MORALES, P. A., CROWDER, C. H., FISHMAN, A. P., MAXWELL, M. H., and GOMEZ, D. G., 1950. Measurement and
trast agents. Comparison of sodium and methylglucamine salts of iothalamate monomer and dimer. Clinical Radiology, 24, 445-448.
significance of urinary appearance time in the dog. American Journal of Physiology, 163, 454—460. ORLOFF, J., WAGNER, H. N., and DAVIDSON, D. G., 1957.
BRICKER, N. S., 1969. On the meaning of the intact
nephron hypothesis. American Journal of Medicine, 46, 1-11.
The effect of variations in solute excretion and vasopressin dosage on the excretion of water in the dog. Journal of Clinical Investigation, 37, 458-464. PITTS, R. F., 1974. In Physiology of the Kidney and Body Fluids, p.114 (Year Book Medical Publishers)."
CATTELL, W. R., KELSEY FRY, I., SPENCER, A. G., and PUR-
KISS, P., 1967. Excretion urography. I—Factors determining the excretion of Hypaque. British Journal of Radiology, 40, 561-580. CHINARD, F. P., 1955. Renal dead space as a function of urine flow in the anaesthetized dog. American Journal of Physiology, 180, 620-622.
PURKISS, P., LANE, R. D., CATTELL, W. R., KELSEY FRY, I.,
and SPENCER, A. G., 1968. Estimation of sodium diatrizoate by absorption spectrophotometry. Investigative Radiology, 3,271-274. SAXTON, H. M., 1969. Review article: Urography. British Journal of Radiology, 42, 321-346.
EVILL, C. A., and BENNESS, G. T., 1974. Metrizamide as a
non-ionic urographic agent. A comparison with sodium iothalamate and its dimer. Investigative Radiology, 9, 434-437.
ZAK,
G. A., BRUN, C , and SMITH, H. W., 1954. The
mechanism of formation of osmotically concentrated urine during the antidiuretic state. Journal of Clinical Investigation, 55,1064-1074.
GOLMAN, K., and ALMEN, T., 1973. Metrizamide in experi-
mental urography. I—Iodine concentration andflowof
Book review Radiographic Contrast Agents. Edited by Roscoe E. Miller and Jovitas Skucas, pp. xv + 515, illus., 1977 (BaltimoreLondon-Tokyo, University Park Press), £21 -25. The two editors have written the opening six chapters— everything about barium sulphate. This first third of their book is superb, and I would like to quote extensively from their summary: "All so-called 'colloidal' barium sulfate preparations for radiology contain additives or contaminants in small amounts that, through surface action, affect surface tension, hydration, particle charge, and pH. These physical properties markedly influence particle size, flocculation, suspension, and viscosity. These latter properties, in turn, largely determine the suspension's ability to settle, flow, bubble, form blood clots, and make lesion-like artifacts; to coat and stick to the mucosa, form various film thicknesses, crack, and achieve and maintain various degrees of density; and to be recorded on the radiograph. "Most barium sulfate additives are not listed on the container, are almost invariably present, and form a coat that is adsorbed on the aggregate particles' surface. This covering may consist of many substances in a variety of patterns with multiple degrees of ionic charge and hydration. "The radiologist, knowingly or unknowingly, purposefully or haphazardly, changes the viscosity, density, surface tension, and film-forming qualities of his suspensions by selecting different brands of barium sulfate and by choosing the concentration and mixing methods in their preparation . . . "By understanding the basic physical, chemical, and
colloidal properties of barium sulfate suspensions and by taking advantage of additives already present, the clinical radiologist can control the performance of his barium sulfate suspensions through intelligent modification. He can do this quite easily by altering rationally his preparation methods and by judicious combination of different compatible barium sulfate compounds in various proportions. With their multiple additives, the radiologist can then use these combined compounds under local conditions in different concentrations for numerous types and methods of examination with superior results . . . "These seemingly simple, but fundamental procedures, let alone the theory behind them, have not been suggested previously in the radiological, medical, patent, or advertising literature." This work goes a long way toward correcting our ignorance about the worst labelled substance in clinical medicine. The rest of the book contains expert contributions by 17 authors on hepatobiliary, pancreatic, genitourinary, angiographic, bronchographic, myelographic and miscellaneous contrast agents. There is afinalsection on paediatrics. These chapters are all clearly presented, and I thought the biliary section particularly good. Inevitably authors often have to go over ground already covered in other publications; the editors have set their contributors an unfairly hard task in matching the knowledgeable enthusiasm and originality of the opening barium chapters. The book is worth buying for these alone, and also as a compact practical reference work on all contrast media.
110
THOMAS SHERWOOD.
