| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Department of Anesthesiology, Perioperative and Pain Medicine Brigham and Women’s Hospital Harvard Medical School Boston, MA
To the Editor:
I read the article by Langevin et al. (1) with interest. The authors have suggested a novel explanation for the clinical impression of nitroglycerin’s (TNG) utility, namely that TNG may not relax the uterus per se, but rather increases its compliance in response to passive stretch. The model used was unique and raises several questions.
First, the most direct evidence to support the compliance hypothesis was obtained in an in vitro rabbit model in which uterine strips were repeatedly contracted to exhaustion in calcium free buffer. TNG failed to alter the response to passive stretch, as measured by the area under the length-tension curve for 10 minutes. In the clinical setting, however, the effectiveness of TNG is gauged by an immediate relaxation to passive stretch (in seconds), and the uterus is then contracted firmly again with oxytocin. Did the strips vary in the time-dependence of the response to stretch? It would have been informative if the authors had reported compliance in the traditional fashion, i.e., the immediate or steady-state tension versus length. Finally, is it possible that the strips were unable to contract under any circumstances, after being calcium-depleted?
The authors cite our previous work (2) demonstrating relaxation of active rhythmic and tonic contractions of rat myometrium in the presence, but not absence of placental tissue but suggested that we did not demonstrate the mechanism to be release of nitric oxide (NO). In the full report however, the effect of TNG was eliminated by methylene blue, an inhibitor of guanylyl cyclase, as well as pyrogallol, a superoxide anion generator, which destroys NO. Moreover, placental tissue modestly increased the action of TNG in aortic tissue, but not of the NO-independent tocolytic drugs Mg+2 and terbutaline (3). Recent clinical data from randomized trials also confirm the ability of TNG to attenuate uterine contraction in the absence of passive stretch (4).
Finally, I do not believe the authors’ results and those of these other investigations (including ours) necessarily describe the same phenomenon. A decrease in active contraction tension would be explained by a change in compliance only if the baseline tone of the uterine strips decreased in direct proportion to the change in contractile force. We did not observe such a decrease, and furthermore, the algorithm used to integrate activity of the strips subtracted baseline tension so as to only record active contraction. Therefore, it would seem that the decrease in active tension seen in our preparation (3) is unlikely to be explained by a change in compliance alone.
References
Department of Anesthesiology University of Florida Gainesville, FL 32610-0254
In Response:
I am pleased that Dr. Segal and his colleagues continue to take interest in our work, and we greatly appreciate the opportunity to respond to each of the insightful questions posed in Dr. Segal’s letter. First, we appreciate Dr. Segal’s suggestion that our model was unique, and although we generally like to think that our work is particularly novel, the model used in our studies was actually not unique at all but used by many others studying similar phenomena. Indeed, calcium-depleted muscle preparations are used quite commonly in the study of passive compliance in many tissues that contain smooth muscle, such as the bladder (1). Second, we must point out that nitroglycerin (TNG) did alter the response of the uterus to passive stretch, nearly doubling it in our preparation. Third, we chose to integrate the area under the length-tension curve for 10 minutes, although more than 90% of the area under these curves is found in the first minute consistent with ability of TNG to "relax" (i.e., accommodate stretch) almost immediately when tension is applied (2). We considered reporting the instantaneous compliance that Dr. Segal would have preferred but could not determine at what point to report this because the tissue accommodates stretch very rapidly and in inverse proportion to its mass. Hence, we did report compliance in the usual fashion, namely a change in tension with respect to length but over a time interval rather than at a specific instant in time. We submitted examples of these curves with our original manuscript as Dr. Segal wishes we had, but they were removed on revision of the final manuscript. Finally, Dr. Segal’s suggestion that the tissue in our preparation was unable to contract under any circumstances really addresses the continuing controversy over the mechanism of action of TNG in uterine tissue at term. Does TNG alter tension by an effect on the force of contraction or an effect on passive compliance?
In actively contracting muscle, if the tension measured at the ends of the tissue is decreased after treatment with TNG there is no way to know if it results from a decline in the force of contraction or an increase in the compliance. This is especially true in tissues containing smooth muscle because stretching the tissue is a stimulus for contraction. Well, if TNG has an effect in tissue that is not contracting, then it MUST have an effect on compliance because there is no way to decrease the force of contraction below zero. Had we been able to perform our experiments in tissue that was completely noncontractile as Dr. Segal suggests, we would have done so. However, we felt compelled to maintain a small contractile response in our muscle strips (approximately 10% of baseline) to convince ourselves that the tissue remained viable after the calcium was depleted.
Regarding our citation of Segal et al.’s work (3), we recognize that there was data indicating that methylene blue and pyrogalol returned uterine contractions to baseline, at least at the 50% response dose of TNG. Our concern, however, is that the authors compared uterine contractile response to TNG with and without placenta. It is not clear from their manuscript what the response to adding these agents to the bath in the presence of the placenta without TNG would have been. Although they state that there was no effect of the placenta on uterine contraction, there is no statistical comparison between uterus with placenta and TNG versus uterus, placenta, and TNG in the presence of methylene blue or pyrogallol. Therefore, we do not know if the addition of methylene blue or pyrogallol eliminated the effect of TNG or the placenta on the uterine tissue. In addition, the authors point out that the data on the R50 dose of TNG was so variable that they were forced to compare the R25 doses of TNG with and without placenta. This suggests that the standard errors were quite large. Yet they apparently used that data in their analysis of the effect of methylene blue and pyrogallol. They only report percent of maximum contraction and not the absolute tensions, so the actual impact of the variability on their findings is difficult to assess. Finally, not measuring nitric oxide (NO) in the head gas always leaves some doubt as to whether the concentration of this intermediate was actually altered when the measurements were made, which admittedly is a shortcoming in our work as well.
We do not dispute the clinical efficacy of TNG in some settings (4) but we remain unconvinced that the effect of this drug is NO-mediated. Diamond and Marshall’s (5) and Diamond’s (6) work suggests that the mechanism of action of the nitrosovasodilators in nonvascular smooth muscle may not be identical to that in vascular smooth muscle. If we better understood the mechanism of action of these agents in nonvascular tissues, we would be in a position to understand why TNG frequently has no apparent benefit in reducing uterine tone. Furthermore, we might be able to determine what settings and at what dose the administration of TNG is appropriate, and we might be able to develop analogs to produce more potent and reliable responses without the risk of a commensurate decrease in blood pressure. Dr. Segal is correct in his final comment that their results and ours are not mutually exclusive and may describe different phenomena. However, it should be obvious to everyone that we must not extrapolate data from term pregnancy to the setting of preterm labor (7), which may be an entirely different issue. We need more in vivo and in vitro data to better characterize the mechanism at term, in the hypercontractile uterus and in preterm labor.
References
This article has been cited by other articles:
|
|
 |
|
  R. W. Hong, M. L. V. H. Greenfield, and L. S. Polley Nitroglycerin for uterine inversion in the absence of placental fragments. Anesth. Analg., August 1, 2006; 103(2): 511 - 512. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
