Kinetics of [99mTc]heparin, venous scintigraphy with [99mTc] fibrinogen and β-thromboglobulin assay in the diagnosis of deep-vein thrombosis

In/. J. Nucl. Med. Viol. Printed m Great Britain

Vol. II, No.

314, pp. 235-241,

1984

0047-0740/84

$3.00 + 0.00 Ltd

Pergamon Press

Kinetics of [99”Tc]Heparin, Venous Scintigraphy with [99mTc] Fibrinogen and p-Thromboglobulin Assay in the Diagnosis of Deep-Vein Thrombosis B. KITSCHKE’, K. ALAOUI-BOUARRAQUI*, A. J. CIURANA’ A. L. WAGNER** and M. VEYRAC’ ‘Clinique Medicate A, Hopital Saint-Eloi, Montpellier and *Institut d’Hematologie, Radioelements, Montpellier, France

Laboratoire

des

(Received 19 Januar.r 1984) The information provided by three examinations: [99”Tc]heparin kinetics (KH), venous scintigraphy with [WmTc]fibrinogen (VSF) and plasmatic assay of /?-thromboglobulin (BTG), are studied on 23 patients separated into two groups on the basis of phlebographic results. VSF which could provide to a certain extent the same kind of information as phlebography, proves to have insufficient sensitivity. KH has high specificity but low sensitivity, while BTG performance is the opposite. The authors discuss the mechanisms responsible for these discordant results. They observe that in cases where the KH and BTG correspond, the diagnostic value is very satisfactory, so that a double positive can replace phlebography in cases where the latter is contra-indicated or difficult to perform.

Introduction A proper examination

and clinical experience are not sufficient to insure against diagnostic errors, either of exaggeration or underestimation, with regard to phlebothrombosis (PT) in lower limbs. An ideal com-

plementary technique to support clinical opinion remains to be found. It would have the advantages of sensitivity and specificity which permit the affirmation of normal flow or thrombosis in the venous trunks. In the case of thrombosis, it should allow one to specify the limits of the thrombus along the venous course. The diagnosis of PT is more difficult than it appears, and a dissociation between the level of symptoms and the level of anatomic lesions is also frequent. This further complicates matters, since the choice of a suitable treatment depends on knowing the extent of the thrombus. Phlebography generally has all these advantages because it can visualize a thrombus and its limits. The disadvantages of the technique are few, but categorical: the trophic condition of the lower limbs sometimes makes it unusable; iodine injection involves an allergic and nephrotoxic risk.

*All correspondence should be addressed to: Dr A. Wagner, Centre de Transfusion Sanguine, BP 1213, 34010 Montpellier, France.

A search for other techniques is thus necessary. In this article we present the results of a study on 23 patients, comparing the information provided by phlebography, which served as the reference examination, with those provided by kinetic measurements of technetium-99m labeled heparin (KH), venous scintigraphy with technetium-99m labeled fibrinogen (VSF) and plasmatic assay of fi-thromboglobulin (BTG).

Material and Methods Criteria for the inclusion of patients

Out of 42 patients admitted to the Medical Clinic A of Montpellier from January 1981 to June 1982 and treated for PT, 25 underwent phlebography. Phlebography was not performed in 17 cases because of the patient discomfort or according to the severity of the disease (9 cases), because PT had already been diagnosed and was being treated (5 cases) or because it was impossible (no venous catheterization in 2 cases, iodine allergy in one case). Out of the 25 patients who underwent phlebography, 13 were included in the study and constitute the first group. They had the KH in 13 cases, the PTG assay in 11 cases and the VSF in 6 cases. The 12 others were either already undergoing heparin treatment before hospitalization (5 cases) or else it 235

236

B. KITSCHKEet ul.

was impossible to postpone treatment because of the urgency of the condition (5 cases), or because of the time needed to do the other examinations (2 cases). Exogenous heparin can potentially falsify the KH and the /?TG; hence we excluded this source of disturbance. A second group of 10 patients served as controls for assessing the specificity of the tests. These patients without PT underwent phlebography to evaluate chronic venous insufficiency (4 cases) or to refute a proposed PT diagnosis (chronic venous insufficiency, 2 cases; infection of a lower limb, 1 case; popliteal cyst, 3 cases). In this group we were able to perform the KH in 10 cases, the /ITG assay in 10 cases and the VSF in 7 cases. Kinetics of q9”Tc-labeled heparin (KH)

In PT, heparin is incorporated into the clot during formation and its plasmatic decrease is thus accelerated. It occurs according to a bi-exponential model, comprised of a fast component and a much slower component.“) The kinetics of heparin can be studied by the injection of 99mTc-labeled heparin. This was prepared with a single-dose vial of lyophilized reagent (Vectoscint-Solabco kit) to which a solution of P’“Tc]pertechnetate was added, with an activity of 0.5-l mCi (18.5-37 MBq). Blood samples were taken at 45, 60, 75 and 90 min post-injection. Plotting the results on semi-logarithmic coordinates allowed us to linearize the slow exponential and thus calculate the biological period (t,,J of [99mTc]heparin. In a series of normal subjects (N = lo), the mean t,,z was situated in the interval 142.5 f 4.9 min with 95% confidence limits. This result differs from the values proposed by the kitmanufacturer (180 _t 30 min) and is characterized by a low dispersion of individual values. We considered any result below 129 min (2 SD) as pathological. fl-Thromboglobulin (/3TG)

In 1975, Moore et al.(2) isolated and characterized the platelet specific protein /I-thromboglobulin. It is released from LXgranules during platelet aggregation. Its level in the blood has thus been proposed as an index of platelet release, as occurs in PT.‘3’ A radioimmunological technique (/l -thromboglobulin-Amersham kit) was used for our assays. Particular care was taken with samples to avoid any accidental in vitro platelet activation: rapid puncture without tourniquet, use of collecting tubes supplied by the manufacturer, immediate refrigeration, centrifugation at 1500-2000g and 4°C for 30 min, assay with 0.5 mL supernatant (plasma with low platelet content). In a series of normal subjects (N = 30), the mean plasmatic /ITG was situated in the interval 40 + 5 ng/mL with 95% confidence limits. We considered any result above 72 ng/mL (2 SD) as pathological.

Venous scintigraphy

with ‘““Tc-labeled

jihrinogen

( VW

Fibrinogen labeled with ‘251is already routinely used to detect thromboses by external measurements post-operatively, but this method does not permit visualization of the thrombus. Labeling with y9”Tc, which has little effect on the molecule,‘4~5’allows us to overcome this drawback. P’“Tc]pertechnetate with an activity of 7710 mCi (259-370MBq) was put into a single-dose vial containing a lyophilized mixture of human fibrinogen (5 mg) and stannous chloride (TCK-19 FibrinocisC.I.S. kit). Intravenous injection was administered within 2 h following preparation and, after urinary miction, scintigraphy was performed within 223 h post-injection, with a double detector heads whole body imager (CLEON 760). Typical results obtained with a normal subject are shown in Fig. 1. The criteria for scintigraphic diagnosis of thrombosis have been described by Jonckheer et al.:‘@ amputation of a large venous trunk (recent thrombosis), limited hyperfixation zone (formed thrombosis), detection of veins normally not visible (collateral circulation), visualization of an irregular venous course (recanalization). The fourth criterion is impossible to identify due to the limited detector definition. The third is unusable because of its occurence in chronic venous insufficiency. Only the first two criteria were investigated and are discussed. Phlebography

The classic technique of phlebography involving bilateral catheterization of a pedal vein was employed in this study. Tourniquet suppression allowed us to obtain iliacae and cava images. A femoral puncture was occasionally necessary.

Results [99mTc]heparin kinetics

The results are compiled in Tables l(a) (Group 1: patients with PT) and l(b) (Group 2: patients with out PT). They are expressed in terms of true and false positives (TP, FP), and true and false negatives (TN, FN) with reference to phlebographic results. Seven TP vs 6 FN were observed in Group 1, and a single FP vs 9 TN in Group 2. The sensitivity is thus only 54% for the diagnosis of PT (7 out of 13) but the specificity is 90% (9 out of 10). The prognostic value of a positive result is 87% (7 out of 8) but only 60% for a negative result (9 out of 15). The diagnostic effectiveness is 69% (16 out of 23). The poor results in Group 1 prompted us to investigate potentially more satisfactory subgroups. We separated the apparently older PT (7 cases) from the more recent ones (less than 48 h: 6 cases) and the sural PT (7 cases) from the more extensive cases (6 cases). This brought no improvement [Table l(c)]. Among the four massive PT formed during our observations, we noted 2 FN and 2 TP.

Kinetics of KH, VSF and BTG in diagnosis of PT b-Thromboglobulin

assay

The results are expressed as previously [Tables 2(a) and (b)]. Out of the 11 assays in Group 1, we obtained 9 TP and 2 FN; out of the 10 assays in Group 2, 6TN and 4 FP. The sensitivity for PT diagnosis is thus 82% (9 out of 11) but the specificity is only 60% (6 out of 10). The prognostic value from a positive result represents 69% (9 out of 13) and from a negative result 75% (6 out of 8). The diagnostic effectiveness is 71% (15 out of 21). The sensitivity seems to increase in the case of massive PT [Table 2(c)]. Venous scintigraphy

with [99mTc]jbrinogen

The results are compiled in Tables 3(a) and (b). In Group 1 we observed 2 TP and 4 FN. Moreover, the information provided by the two positive results was quite inferior to that of phlebography, which allowed us to visualize a femoral thrombus in case 13 and an ilio-femoral thrombus in case 14. Hence, scintigraphic examination on its own would not have allowed an adequate therapeutic choice. In Group 2 the one FP is a “double” because it combines a right sural hyperfixation-the seat of an erysipelas-with a left femoral amputation, the vein having been freed for phlebography. We must finally note the moderate diffuse fixation with respect to the popliteal cysts, a sign of stasis. Hence, we obtain a sensitivity of 33% (2 out of 6) for this examination and a specificity of 86% (6 out of 7). The prognostic value of positive results represents 66”/, (2 out of 3), and from negative results 60% (6 out of 10). The diagnostic effectiveness is 61% (8 out of 13).

Discussion Venous scintigraphy

with [99mTcJjibrinogen

This examination is the only one which can theoretically provide the same information as phlebography: diagnosis and localization. Even though we only performed it 13 times, its lack of sensitivity is obvious. We excluded the signs of simple stasis encountered in almost all the scintigrams, which would have caused us to lose all specificity in the examination. Hence, only 2 PT were detected by a hyperfixation zone, but this zone was much more limited than the one delineated by phlebography, which makes the examination useless for a therapeutic choice. In spite of this lack of sensitivity, there was a FP on a lymphangitis. The popliteal cysts provoked diffuse hyperfixation which was easily distinguishable from a thrombotic zone. The fixation of fibrinogen to an experimental thrombus has been proved.(7,8’The principal study of human pathology is that of Jonckheer:(@ out of 13 patients, 8 with PT, he obtained 7 TP, 5 FP and 1 FN. The method seems sensitive, even too sensitive (5 FP) in comparison to what we have observed. The differences between the results can probably be attrib-

‘37

uted to problems of image quality and the choice of scintigraphic criteria for anomaly. The two series unfortunately correspond with respect to the weak diagnostic effectiveness: 54 and 61%. Heparin therapy, which diminishes the rate of fibrinogen-fixation to the thrombus, cannot be questioned here. The results should rather be attributed to periodic variations in the thrombus activity, and hence in its capacity to incorporate labeled fibrinogen. P’“Tc]heparin

kinetics

This examination appears to lack sensitivity (6 FN) but is very specific (1 FP). In case of diagnostic doubt concerning a patient for whom phlebography is difficult or inadvisable, a positive result with this test would be enough to confirm the diagnosis. Two studies have already been carried out with this examination. Esquerre et al.“’ studied 19 patients with PT, of which 17 were already on heparin treatment. The authors seem satisfied with their results. but the values they considered pathological are not precise, which makes it difficult to evaluate the effectiveness of this examination. Out of 10 patients with PT, Bouvier”’ observed two normal values, 6 doubtful values, and only two values that were clearly pathological. In the face of these conflicting results, only the high specificity of our series stands out. The normal value that we defined on the basis of 10 subjects without venous pathology (142.5 f 4.9) is lower than that of the other teams (usually about 180 min) and partially explains this specificity. Esquerre elsewhere demonstrated”’ that heparin clearance is greater when its plasmatic concentration is weaker. Hence, the absence of exogenous heparin in our patients increases the value of our few positive results. The FN must doubtless be attributed to the natural periodic development of the thrombus, with dormant phases during which heparin is not incorporated into the clot. p - Thromboglobulin

assay

In this case, sensitivity largely surpasses specificity. The norms are not in question here because all the authors agree to the upper limit of approximately 70 ng/mL. The results are, on the whole, satisfactory. though they do not achieve the perfection of the Ludlam et al. series”‘) in which 6 patients with PT had a /?TG of between 80 and 140 ng/mL, whereas the unconfirmed suspected PT and the controls all had levels lower than 60 ng/mL. The other series are less satisfactory and correspond to our results.“’ “I In spite of all the precautions we took, the FP are probably due to platelet activation in t’itro after sampling, which increased falsely the plasmatic concentration of /lTG. The assay of platelet factor 4 combined with that of /?TG’13) could make up the difference between activation in uivo (rise in plasmatic

B. KIXSCHKE et al.

238

Table 1. Kinetics of [99”Tc]heparin (a). Pafienfs wirh PT (Group 1) Case No. Heparin f,,r (min) Interpretation

1

2

4

5

6

I

8

10

12

13

14

17

19

108 TP

142 FN

102 TP

194 FN

109 TP

126 TP

140 FN

123 TP

186 FN

112 TP

134 FN

157 FN

118 TP

Table l(b). Patients without PT (Group 2) Case No. Heparin fm (min) Interuretation

3

9

11

15

16

18

145 TN

90 FP

183 TN

186 TN

170 TN

152 TN

20 21 22 ._~~~_~_-.~___~~ 280 TN

160 TN

23

._~

210 TN

130 TN

Table l(c). Subgroups in Group 1

Old PT (N = 7) Recent PT (N = 6) Sural PT (N = 7) Massive PT (N = 6)

TP

FN

4 3 5 2

3 3 2 4

TP = true positive, TN = true negative, FP = false positive, FN = false negative.

Table 2. fi-Thromboglobulin assay (a). Patients with PT (Group 1) Case No. /I-Thromboglobulin (ng/mL) Interpretation

2

4

5

6

7

8

10

12

13

17

19

106 TP

160 TP

192 TP

110 TP

26 FN

210 TP

68 FN

110 TP

210 TP

74 TP

89 TP

Table 2(b). Patients wirhout PT (Group 2) Case No. P-Thromboglobulin (ng/mL) Interpretation

3

9

11

15

16

18

20

21

22

23

200 FP

198 FP

56 TN

65 TN

68 TN

88 FP

60 TN

52 TN

74 FP

58 TN

Table 2(c). Subgroups in Group 1

Old PT Recent PT Sural PT Massive PT

(N=6) (N=5) (N=6) (N=5)

TP

FN

5 4 4 5

1 : 0

Table 3. Venous scintigraphy with [99mTc]fibrinogen (a). Patients with PT (Group 1) Case No.

Result of scintigraphy

10 12 13 14 17 19

Normal Normal Left calf hyperhxation Left sure-popliteal hyperhxation Normal Normal

Interpretation FN FN TP+ TP* FN FN

--

Table 3(b). Parienrs withour PT (Group 2) Case No.

Result of scintigraphy

Interpretation

11 15 16 18 21 22 23

Normal Normal Normal Normal Right sural hyperfixation; left iliac amputation Normal Normal

TN TN TN TN FP TN TN

Fig. 1. Venous scintigraphy with [w’Tc]fibrinogen (normal subject).

239

Kinetics of KH, VSF and fiTG in diagnosis of PT

content of both molecules) and that occurring after bad sampling (increase of /ITG only). The FN are more difficult to explain. Smith et aLCi4)only observed 12 increased /ITG in the first sampling of 24 patients with PT but 21 out of 24 had increased /ITG at least once in successive samplings over one week. This once again poses the problem of the anatomic activity of the clot, which is uneven. The half-life of /ITG is short (100 min) and if its secretion is only triggered by new periods of platelet activation, it is conceivable that normal levels could be observed between two secretory periods. Combination

of the results

Of the three techniques of examination used in this study, VSF is not only the least effective but also the most inconvenient because it implies the transfer of the patient to the Nuclear Medicine Department, which can be sometimes dangerous according to the risk of emboly. The other two methods can be performed more easily and without transfer. Assays and readings are quite rapid. PTG poses the usual problem of organization and cost of radioimmunoassays in case of urgency. Eleven patients had corresponding results from ,!ITG and KH: 6 double negatives and 5 double positives. The 6 double negatives were all true and, out of the 5 double positives, only one patient did not have active PT. The diagnostic effectiveness of this pair of results is 91%. Thus, for a patient whose clinical examination revealed little, a double positive is a good indication that phlebography should be

241

performed to assess the extent of the thrombus. If this examination is inadvisable (renal insufficiency for example), the double positive should be sufficient grounds for undertaking an anticoagulant treatment. A double negative in cases of obesity or chronic renal insufficiency with aching legs, would appear to preclude active PT.

References 1. Esquerre J. P., Boneu B. and Guiraud R. Inr. J. NW/. Med. Biol. 6, 215 (1979). 2. Moore S., Pepper D. S. and Cash J. D. Biocl~in~. Biophys. Acta 379, 360 (1975). 3. Zahavi J. and Kakkar V. V. Thromh. Haemost. 44, 23

(1980). 4 Harwig J. F., Harwig S. S. L., Wells L. D. and Welch M. J. Int. J. Appl. Radiat. Isot. 27, 5 (1976). 5. Jeghers O., Abramovici J., Jonckheer M. and Ermans A. M. Eur. J. Nucl. Med. 3, 95 (1978). 6 Jonckheer M., Abramovici J., Jeghers O., Dereume J. P. and Goldstein M. Eur. J. Nucl. Med. 3, 233 (1978). 7. Hale T. I. and Jucker A. Eur. J. Nucl. Med. 3, 267 (1978).

8. Harwig S. S. L., Harwig J. F., Coleman R. E. and Welch M. J. J. Nucl. Med. 17, 40 (1976). 9. Bouvier J. F. Bulletin de I’dcomen 6, 169 (1981). 10. Ludlam C. A., Bolton A. E., Moore S. and Cash J. D. Lancet 2, 259 (1975). 11. Conard J., Terrier E., Baillet M., Horellon

M. H., Jaulmes B., Samama M. and Baillet J. NOUI,. Press. Med. 10, 1327 (1981).

12.

Van Hulsteijn H., Briet E., Koch C., Hermans J. and

Bertina R. Acta Med. Sand. 211. 323 (1982). 13. Kaplan K. L. and Owen J. Blood, 57, i99 (i981). 14. Smith R. C., Duncanson J., Ruckley C. V., Webber R. G., Allan N. C., Dawes J., Bolton A. E., Hunter W. M.. Pepper D. S. and Cash J. D. Thromh. Haemost. 39, 338 (1978).