Meconium and fetal hypoxia: some experimental observations and clinical relevance
* Dr J. Westgate, The Liggins Institute, Faculty of Medical and Health Sciences, The University of Auckland, Private Bag 92019, Auckland, New Zealand.
In an experimental study, chemically sympathectomised near term fetal sheep and a control group were subjected to repeated episodes of acute hypoxia. Despite severe hypotension and metabolic acidosis, no animal in the control group had meconium-stained amniotic fluid, whereas every animal in the sympathectomised group had heavily meconium-stained amniotic fluid at the end of the experiments. These data and the available literature do not support a direct association between acute hypoxia and meconium-stained amniotic fluid but suggest that a reduction in sympathetic neural tone must be a component of meconium passage. Clinical and experimental data on the occurrence of meconium-stained amniotic fluid are reviewed.
The presence of meconium in the amniotic fluid is traditionally regarded as evidence of fetal compromise or ‘distress’. However, the association between meconium and fetal condition at birth is complex and puzzling. Meconium alone, in the absence of fetal heart rate (FHR) abnormalities, is not associated with poor fetal outcome but fetuses with both abnormal FHR patterns and meconium are more likely to be acidotic at birth and require resuscitation1. Increased catecholamine levels have been reported in the cord arterial blood of fetuses with meconium-stained amniotic fluid2 suggesting that increased sympathetic nervous system activity is associated with the passage of meconium. However, others have suggested that high catecholamine levels may inhibit meconium passage3. We report the incidence of meconium during experimental studies in fetal sheep in which the effect of chemical sympathectomy on the fetal response to repeated hypoxia was examined.
Sixteen Romney/Suffolk sheep were operated on under halothane anaesthesia (2%) using sterile techniques, as described previously4. Polyvinyl catheters were inserted into the right and left brachial arteries of the fetus and in the amniotic cavity and an inflatable occluder cuff (In Vivo Metric, Healdsburg, California) was placed around the umbilical cord. Two electrocardiogram (ECG) electrodes (Cooner wire AS633-3SSF) were placed subcutaneously, one over the apex of the heart and the other on the right shoulder. The maternal long saphenous vein was catheterised. After surgery, ewes were housed in a cage with free access to water and hay, supplemented with sheep nuts and alfalfa. They were kept in a temperature controlled room (16°C, humidity 50%) in a 12 hour light/dark cycle. Each received gentamicin (80 mg, intra-amniotically) daily throughout the experiments. After completion of the studies, animals were euthanised by overdose of intravenous pentobarbital. Studies were approved by the Animal Ethics Committee of the University of Auckland.
One day after surgery, chemical sympathectomy (n= 8) was performed with an infusion of 150 mg of 6-hydroxydopamine (6-OHDA) based on a dose of 50 mg/kg with estimated fetal weight being 3 kg5. Administration of 6-OHDA depletes noradrenaline from peripheral nerve endings and leads to the destruction of these nerve terminals, thus preventing any neural sympathetic activity. The humoral arm of the sympathetic nervous system, which acts through the release of adrenaline and noradrenaline from the adrenal glands, is not affected by 6-OHDA. In order to test the effectiveness of the chemical sympathectomy, a 2-μg bolus of tyramine was injected the following day. Tyramine acts to stimulate the release of noradrenaline from peripheral nerve endings and resulted in a rise in blood pressure and FHR in control fetuses but not in sympathectomised fetuses.
Experiments were started three to five days after surgery, at a gestational age of 126.3 ± 2.6 days (term is 147 days). Fetuses in both groups were subjected to repeated total umbilical cord occlusion for 2 minutes out of every 5 minutes by inflating the cuff around the umbilical cord with sterile saline, and then deflating it after 2 minutes. This procedure was repeated for up to 4 hours, or until the fetal mean arterial blood pressure (MAP) had fallen below 20 mmHg during two successive occlusions, or the fetal blood pressure failed to recover to baseline levels when the next occlusion was due. Fetal arterial blood gas analysis and measurements of glucose and lactate levels were performed immediately prior to the first occlusion and after every sixth occlusion and after the final occlusion. The presence and nature of meconium in the amniotic fluid was noted at postmortem 24 hours after the experiments.
Each umbilical cord occlusion resulted in a large variable FHR deceleration which lasted for the entire 2-minute occlusion. Fetal blood pressure fell progressively during occlusions from the third occlusion onwards to reach a mean (SD) of 17.6 (2.1) mmHg after 20.7 (4.8) occlusions in the control group and 16.4 (2.6) mmHg after 16.1 (5.8) occlusions in the sympathectomy group. There was a progressive fall in fetal pH and a rise in base deficit and lactate in both groups such that at the end of the occlusions the acid–base status in the control group was pH 6.99 (0.07), base deficit 17.0 (3.9), lactate 12.5 (2.0) mmol/L and in the sympathectomised group: pH of 6.95 (0.07), base deficit 17.5 (2.7) mmol/L and lactate 12.3 (2.1) mmol/L.
At postmortem, no fetus in the control group had meconium present in the amniotic fluid, whereas every fetus in the sympathectomised group had heavily meconium-stained amniotic fluid. There were no differences in fetal birthweight, gestational age or pre-occlusion acid–base status.
These data show, firstly, that repeated episodes of fetal asphyxia leading to severe metabolic acidosis and episodic hypotension in previously healthy fetuses were not associated with the presence of meconium in amniotic fluid of near term fetal sheep. Secondly, the observation that loss of neural sympathetic activity was associated with meconium passage indicates that neural sympathetic tone is responsible for the maintenance of anal sphincter tone. Because meconium passage was determined at the end of the study, it is not possible to be sure whether meconium passage occurred in response to sympathectomy alone or only with the combination of asphyxia and sympathectomy. Chemical sympathectomy results purely in a loss of neural sympathetic activity, leaving the adrenal response intact. Indeed, previous studies have shown that circulating catecholamine levels are increased after chemical sympathectomy, and increase further during acute asphyxia compared with intact controls5. Our data indicate, however, that this response is not sufficient to maintain anal sphincter tone. Thus, although the mechanisms of meconium passage are not necessarily identical in man and sheep, these data indicate that a reduction in neural sympathetic tone is required for fetal defecation to occur.
Very few experimental studies have specifically addressed the relationship between meconium and hypoxia or asphyxia. Acute inhalational hypoxaemia was not associated with the passage of meconium in fetal sheep6,7, but rather with reduced peristalsis in the fetal gut8. In a small study of three animals, meconium was observed after 30 minutes partial umbilical cord occlusion resulting in hypoxia and acidosis7. Massive meconium release prior to fetal demise was noted after injection of sodium pentobarbital, probably related to terminal hypotension and loss of tone in the fetus6.
If acute hypoxia or ‘fetal distress’ is not consistently associated with the presence of meconium-stained amniotic fluid, what is? We know that meconium is related to gestational age, being more common after term9. It is also related to maternal ingestion of bowel purgatives10,11. Experimentally, prolonged infusion of cholic acid into the fetus was associated with the presence of meconium due to its stimulatory effect on colonic motility12. An increase in meconium-stained amniotic fluid has been reported from randomised trials of vaginal misoprostol; this has been hypothesised to represent a direct stimulatory effect of the drug or metabolites on bowel motility13. Ethnic differences in the incidence of meconium have been noted, for example, meconium-stained amniotic fluid is more common in Pacific Island women14 and black African infants15. It has been suggested that these ethnic differences relate to differences in maturity of the gastrointestinal system15, however, the circumferential musculature of the colon is complete by 10 weeks of gestation and innervation is complete by the 12th week16. Development of the longitudinal musculature starts at the anal canal and progresses up towards the caecum, fully encircling the colon by the fourth month of pregnancy. Postmortem studies have shown a normal distribution of ganglion cells in the preterm fetus from at least 24 weeks of gestation. It may be that the ethnic differences in the incidence of meconium relate to differences in the intake of dietary substances, which stimulate colonic motility, or to differential rates of maturation of the gut innervation. More recently, it has been suggested that meconium may correlate better with chronic asphyxia rather than with acute asphyxia as meconium-stained amniotic fluid is associated with raised erythropoietin levels in mixed arteriovenous umbilical cord blood17.
Have we been somewhat naive in thinking that a fetus does not normally pass bowel contents until delivery? Infants born with an anorectal malformation usually have an abnormally dilated large bowel proximal to the dilatation, suggesting that the fetus would normally defecate in utero18. Experimental studies have demonstrated that fetal swallowing and defecation of X-ray contrast medium occurs in normoxic fetuses in utero8,19. The passage of radioactive technetium from the bowel to the amniotic fluid in fetal rabbits can be prevented by purse string closure of the anus20. Rather than being constipated in utero, it seems more likely that regular fetal defecation is a normal physiological process. Even more intriguing is the finding that fetal hypoxia results in the accumulation of radioactive technetium in the amniotic fluid, suggesting that the mechanisms which normally result in its clearance from the amniotic fluid have been impaired by hypoxia20. These data lead to the hypothesis that meconium is normally passed into the amniotic fluid but cleared rapidly, probably by fetal swallowing and membrane macrophages, so that the amniotic fluid is normally not obviously meconium stained. In the presence of hypoxia, these clearance mechanisms are likely to be impaired so meconium accumulates. This hypothesis is appealing and could explain the association between meconium and raised erythropoietin levels but does not explain the association between meconium and gestational age. Nor does it explain why the amniotic fluid draining from women with preterm ruptured membranes is usually clear, when presumably most amniotic fluid escapes before it can be swallowed or scavenged by macrophages.
The lack of experimental association between hypoxia and meconium does not alter the significance of the relationship between thick meconium-stained amniotic fluid and increased perinatal morbidity21–23. Thick meconium suggests low amniotic fluid volume and thus a reduced volume of dilution for the meconium. Oligohydramnios occurs secondary to chronic fetal hypoxia and placental insufficiency and these considerations are very likely to lead to a strong but indirect association between thick meconium and poor outcome. Other mechanisms such as increased likelihood of cord occlusion with low amniotic fluid volume and the vasoconstricting effect of meconium on umbilical and placental vessels24 may also contribute to the increased morbidity noted in these fetuses.
In conclusion, the present experimental data and the available literature do not support a direct association between acute hypoxia and meconium-stained amniotic fluid. The data do suggest that a reduction in neural tone must be a component of meconium passage. This may occur physiologically, and/or in response to gut irritation. If in utero fetal defecation is a normal physiological process, we do not have a clear understanding on how the green discolouration is normally removed from the amniotic fluid and why meconium staining is more common at advanced gestation. The answers to these questions await further developments in our understanding of amniotic fluid volume regulation.
The authors' work reported in this review has been supported by National Institutes of Health grant RO-1 HD32752, and by grants from the Health Research Council of New Zealand, Lottery Health Board of New Zealand and the Auckland Medical Research Foundation.