To the Editor:—
With great interest, we read the article by Bauer et al. 1on sublingual microvascular perfusion in 47 patients who underwent hypothermic cardiopulmonary bypass surgery. The authors report a 10% decrease in functional capillary density during cardiopulmonary bypass without changes in microvascular diameter or erythrocyte velocity. These results are partly in contrast with our recent findings in a comparable setting.2
Although both studies evaluated the sublingual microcirculation by a two-dimensional imaging technique, several differences do exist, of which the imaging technique itself and the subsequent way of analysis are most important. To understand the different results more comprehensively, we would like to comment on several aspects of the study by Bauer et al.
First, we used side-stream dark-field imaging, a novel technology based on the orthogonal polarization spectral imaging technique used by the authors. Side-stream dark-field imaging differs from orthogonal polarization spectral technology in terms of magnification and capillary contrast. These differences between the two-dimensional imaging techniques hinder comparison of results obtained by orthogonal polarization spectral versus side-stream dark-field imaging and may explain the differential results. Second, the authors used software to measure microvascular diameter, erythrocyte velocity, and functional capillary density. In our study, a semiquantitative analysis technique was used. Although software can be helpful in decreasing the burden of a time-consuming semiquantitative analysis, we have to look critically at the numbers produced by the software. For example, we would like to learn from the authors whether it was possible to measure erythrocyte velocity in each investigated capillary and venule. Using Microscan Analysis Software (MicroVisionMedical, Inc., Amsterdam, The Netherlands), we experienced that it was impossible to measure high erythrocyte velocities that do exist in a substantial number of capillaries. This problem is probably due to a limited video frame rate: 25 frames/s for phase alternating line standard. Finally, several issues remain unclear after reading the authors’ article. The inclusion criteria used by the authors are not exactly mentioned. Did the authors investigate consecutive, low-risk patients? What was the estimated risk of surgery for the patient population (logistic European System for Cardiac Operative Risk Evaluation [EuroSCORE])? What were the incidences of postoperative morbidity and mortality? We think it might be interesting to investigate a possible relation between intraoperative hypoperfusion of the microcirculation and postoperative outcome. This might be studied in a subgroup of patients with impaired functional capillary density during cardiopulmonary bypass. In addition to this, we wonder why the authors did not separate venules from capillaries, using a cutoff of 20 μm.
To conclude, it is of interest to note that both studies reported moderate changes in the sublingual microcirculation that probably reflect a complex pathophysiology during cardiopulmonary bypass. It is expected that novel bedside imaging technology will simplify further microcirculation research in patients based on studies that were performed previously in laboratory animals.3,4We should focus on the questions of which individual stimuli are responsible for the reported changes and whether these changes are of clinical significance. Larger studies, perhaps in high-risk patients, would be helpful to draw stronger conclusions.
*Erasmus Medical Center, Rotterdam, The Netherlands. c.denuil@erasmusmc.nl