Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Editor—The paper of Zhang et al 1 on the effects of dopamine relies on the assumption that left ventricular output (LVO) is a measure of systemic blood flow (SBF). This is not true in the presence of any shunt through the ductus arteriosus where LVO becomes the sum of SBF and volume of blood shunting back into the lungs through the duct, and so overestimates SBF. While Zhang et al 1 define the ducts in their study babies as insignificant, we would question the validity of the criteria used to make this definition. We have never seen shunting only in diastole, chamber enlargement is an inconsistent sign of ductal significance which develops after day one,2 and shunt velocity has little relation to shunt size during week one.3 In fact, the left to right velocity often increases as the duct constricts and the shunt diminishes in size.3 Bidirectional shunting is also usually predominantly left to right.
There is much in the data presented by Zhang et al 1 to suggest that these ducts were highly significant. Firstly, they selected a population at high risk for a significant duct,4 and secondly they measured a mean diameter of 2.9 mm. We showed that this measure is the most accurate predictor of early haemodynamic significance and that preterm ducts over 2 mm in diameter, with very few exceptions, have a haemodynamic impact.3 ,5 In this haemodynamic milieu, LVO is actually a measure of pulmonary blood flow, and right ventricular output becomes the better measure of SBF,5 but even this can be confounded by the common finding of left to right atrial shunting.3 ,6 That this haemodynamic impact is often present from the very early postnatal period is emphasised in fig 1which is of a 5 hour old infant of 26 weeks gestation, in whom we measured an LVO of 350 ml/min per kg but a right ventricular output (RVO) of 90 ml/min per kg. This baby had an unrestricted duct (2.6 mm) with a low velocity (0.7 m/s) but high volume left to right shunt. In other words LVO was overestimating SBF by a factor of over 300%. This is not an unusual finding.3
This means that the changes in LVO in response to dopamine documented by Zhang et al 1 may have occurred in the systemic circulation but, equally, could have occurred in the pulmonary circulation. The same uncertainty applies to changes in calculated vascular resistance. In other words, LVO may have increased or decreased solely because dopamine has changed the volume of the ductal shunt back into the lungs.
Early preterm ductal shunting should not be assumed to be inconsequential, and interpretation of early postnatal measures of either right or left ventricular output in preterm infants should be approached with caution.
Dr Zhang et al respond: Dr Evans and his colleagues have raised two main issues about our paper.1-1 The first relates to the haemodynamic significance of shunting through the ductus in our subjects. We readily acknowledge that shunting was present and that the measured left ventricular output was therefore an approximation of systemic tissue blood flow. This was unavoidable, as most babies were studied on the first day after birth, when the ductus is proceeding to close.1-2 However, on the basis of clinical assessment as well as other data not presented—for example, average left ventricular shortening fraction was 29%—we considered that myocardial dysfunction rather than ductal shunting was the most likely cause of the systemic hypotension shown.
Our study was confined to a select subgroup of babies with hypotension who failed to respond to volume loading and were thus started on inotropes. In contrast, the studies cited by Evans and colleagues were performed in groups of infants who may or may not have received volume loading and/or inotropic treatment1-3-1-7 and in whom the level of myocardial function ranged from depressed to normal.1-6 Furthermore, most of our subjects were studied within the initial 24 hours after birth, whereas the studies cited1-3-1-7 were performed over the course of the first postnatal week. Given the substantial physiological changes in the cardiovascular system,1-8 1-9 findings obtained in the middle or end of the initial week may not necessarily be applicable to the first day after birth.
Evans and colleagues state that “left ventricular output may have increased or decreased solely because dopamine changed the volume of the ductal shunt back to the lungs”. The physiological changes we observed suggest that this was unlikely and that dopamine exerted a vasoactive effect within the systemic circulation. Thus the premise that the increase in left ventricular output was related only to greater left to right ductal shunting is not in accord with the concomitant rise in systemic blood pressure induced by dopamine. The contention that falls in left ventricular output were merely due to a reduction in ductal shunting is inconsistent with the observed fall in superior mesenteric artery velocity and rise in superior mesenteric vascular resistance.
A possibility deserving of consideration is that dopamine changed the degree of ductal shunting and that this comprised one component of the alteration in left ventricular output. We are not aware of any published data in newborn infants that support direct relaxation or constriction of the ductus by dopamine or of a differential effect of dopamine on the systemic and pulmonary vascular beds. Indeed, in studies performed in newborn lambs, we observed that, at infusion rates of up to 15 μg/min per kg, the systemic and pulmonary vascular effects of dopamine were proportionally similar in magnitude (J J Smolich, H Park, and D J Penny, unpublished observations). In our view therefore it was reasonable to assume that, while ductal shunting was present in our babies, the degree was not altered to a substantial extent by dopamine. However, the notion that inotropic treatment could alter ductal shunting in preterm infants has potentially important clinical ramifications and so is worth investigating.
Finally, although we agree that the illustrative example of the degree of dissociation which may occur between the level of left and right ventricular outputs is striking, examination of group data from the same laboratory suggests that, on average, left ventricular output is only 7–15% higher than right ventricular output in ventilated preterm infants.1-6 Moreover, while we did not include the data in our paper, the pulmonary trunk velocity-time integral in our infants was not significantly different from the ascending aortic velocity-time integral (p > 0.9), suggesting that the marked degree of dissociation between ventricular outputs evident in the illustrative example was not a feature of the babies in our study.