TO THE EDITOR:
We read with interest the recent article by Robins et al. (1) in which they reported biokinetic data of leukocytes labeled with stabilized 99mTc-HMPAO (exametazime) by a method that had earlier been described by their group in this journal (2) and elsewhere (3), with minor differences in methodology and results.
Labeling white cells with technetium rather than indium has obvious advantages in terms of availability, image quality, and radiation absorbed dose. It became a clinical reality with the introduction of HMPAO, first described for white cell labeling by Peters et al. (4) in 1986, and has since proved useful in imaging acute inflammation and inflammatory bowel disease. Hung et al. (2) have aimed to take this further by using HMPAO stabilized with methylene blue. This, however, requires a change in cell isolation procedure, because the dark blue color makes it impossible to distinguish the leukocyte sediment from the supernatant. Hung et al. have suggested using a double-dilution technique. Their protocol requires the addition of a total of 17 mL of a 12.6% acid citrate dextrose (ACD) solution in normal saline.
The biokinetics described by Robins et al. (1) can be summarized as follows. Recovery at 30 min and 1 h was 18% and 16%, respectively. The ratio of activity in liver to that in spleen changed from 1.3 at 1 h to 1.6 at 24 h after injection. Residence time in the lung was 0.55 h, and the mean half-clearance time was 16 min.
We are concerned that these biokinetics may represent a mild-to-moderate form of white cell activation during labeling. Robins et al. (1) already pointed out that the half-clearance time in the lung in their study was longer than the previously reported times of 7.7 min for Becker et al. (5) and 9.8 min for Brown et al. (6). We do not agree with the observation of Robins et al. that there was no visual evidence of prolonged lung retention. In Figure 2 in their paper, considerable lung uptake can still be seen 8 h after injection, which one would expect to observe only in a 1-h image. A slight progressive sequestration in the liver (rather than physiologic pooling in the spleen) and the surprisingly low recovery (which is normally in the 30%–40% range, again with no differences between technetium- and indium-labeled leukocytes), are also indicative of cell activation.
Indeed, higher than normal lung uptake with a prolonged intrapulmonary transit time, followed by a progressive accumulation in the liver rather than the spleen, are clear signs of white cell activation during labeling (5). There is still no better method for evaluating the viability of radiolabeled white cells than studying their kinetics in vivo.
Apart from general manipulation during the cell isolation and labeling process, there are two possible causes for leukocyte activation in this study: the labeling in saline rather than plasma and the methylene blue buffer solution.
The former has been extensively investigated in connection with the question of whether white cell labeling with indium should be performed with oxine (in saline) or tropolone (in plasma). There is a general consensus that labeling leukocytes in plasma rather than saline is advantageous because the cells are maintained in their normal physiologic environment. Labeling white cells in saline has long been recognized as a possible cause of cell activation, although clinical studies on the preference of oxine or tropolone gave conflicting results.
The influence of the latter factor, methylene blue, on leukocyte activation is not known. It can, however, indirectly be determined by changing the cell isolation protocol such that the ACD–saline solution is replaced with plasma, which can easily be achieved by moderately increasing the amount of blood drawn. However, labeling in plasma is known to lead to a decrease in labeling efficiency (7). Because stabilized HMPAO provides a remarkably high labeling efficiency in saline (1), it might be expected to give an acceptably high labeling efficiency in plasma as well. If the biokinetics of the white cells labeled with stabilized HMPAO in plasma do not correspond to what has previously been described, the methylene blue buffer solution is likely to have a detrimental effect. If it does, the use of stabilized HMPAO will be a welcome contribution to the daily work of any nuclear medicine department performing white cell labeling in vitro.
REFERENCES
REPLY:
We appreciate the comments of Drs. Grüning and Franke and would like to offer our response as follows. Drs. Grüning and Franke correctly point out that increased lung half-clearance time and decreased peak recovery values of stabilized 99mTc-exametazime–labeled leukocytes may indicate greater cell activation during labeling, as acknowledged in our article (1). However, our study indicated that lung residence times (0.55 h vs. 0.57 h) and dosimetry values (0.0056 mGy/MBq vs. 0.011 mGy/MBq) were smaller compared with those reported using nonstabilized 99mTc-exametazime–labeled leukocytes (1,2), and lung uptake did not appear increased by visual assessment on review by several experienced readers (1). To highlight the bowel activity for publication, we intentionally made the 8-h image quite dark in Figure 2 of our article (1). Geometric mean analysis of image data also showed less lung uptake with stabilized 99mTc-exametazime–labeled leukocytes compared with previous studies of unstabilized 99mTc-exametazime–labeled leukocytes at both 1 h and 24 h after injection (i.e., 7.6% ± 1.1% vs. 8.0% at 1 h, and 3.3% ± 0.5% vs. 5.3% at 24 h) (1,2).
As noted by Drs. Grüning and Franke, the ratio of liver uptake to spleen uptake of stabilized 99mTc-exametazime–labeled leukocytes increased between 1 and 24 h (1.31 and 1.63, respectively) (1). These values are very similar to the liver-to-spleen uptake ratios reported at 1 and 24 h for unstabilized 99mTc-exametazime–labeled leukocytes (1.26 and 1.50, respectively) (2). The kinetics in the liver are complex, and activity may be affected by transient sequestration, irreversible uptake, or physiologic pooling in the liver. It has been shown that both splenic activity and liver activity will decrease between 1 and 4 h and the splenic pool will continue to decrease, whereas liver activity then tends to rise slowly up to 24 h. Therefore, an increase in the liver-to-spleen ratio after 4 h is expected (3,4). Moreover, in our study, hepatic uptake showed a larger decrease between 1 and 24 h with stabilized 99mTc-exametazime–labeled leukocytes (from 24.3% ± 6.3% to 15.2% ± 2.9%) (1) compared with unstabilized 99mTc-exametazime–labeled leukocytes (from 17.4% to 15.6%) (2), suggesting a larger percentage return of stabilized 99mTc-exametazime–labeled leukocytes to the circulation. Hepatic uptake at typical imaging times (i.e., 4–24 h) of leukocytes labeled by this method is in reasonable agreement with previously reported values (3,4).
Peak recovery of leukocytes labeled with stabilized 99mTc-exametazime was approximately 18% (1), which is less than that reported for leukocytes labeled with other agents. As pointed out by Drs. Grüning and Franke, and acknowledged in our article (1), reduced recovery is of concern for cell damage during labeling. Drs. Grüning and Franke suggested that cell labeling in a saline medium could contribute to cell activation. It seems unlikely that the relative reduction in peak recovery observed in our study was related to the use of saline, because most other methods reporting higher recovery also involved the use of a saline medium for cell labeling. The addition of methylene blue during labeling seems a more plausible explanation for this difference. A recent unpublished study conducted in our laboratory showed that there was little difference in chemotactic behavior among three radiolabeled leukocyte preparations: (a) unstabilized 99mTc-exametazime (i.e., without methylene blue), (b) stabilized 99mTc-exametazime with 250 μg methylene blue, and (c) stabilized 99mTc-exametazime with 500 μg methylene blue. Chemotaxis is not synonymous with cell activation but indicates preserved viability and function, and the clinical relevance of transient sequestration in the lungs is unknown.
As has been recognized by others, Drs. Grüning and Franke remark that there is “still no better method for evaluating the viability of radiolabeled white cells than studying their kinetics in vivo.” By the same token, the best measure of the clinical utility of radiolabeled leukocytes is their performance in patient studies. In earlier studies of 111In-oxine–labeled leukocytes, Datz et al. (5) found no correlation between test sensitivity and uptake in the lungs, or between leukocytes labeled with 111In-oxine in saline vs. 111In-tropolone in plasma. It is uncertain whether somewhat greater cell activation, as evidenced by lower peak recovery and prolonged lung half-clearance time (1), would affect the clinical utility of leukocytes labeled with this method, and clinical trials using stabilized 99mTc-exametazime–labeled leukocytes seem to be warranted for further assessment.