Endogenous formation of carbon monoxide in newborn infants. I. Non-icteric and icteric infants without blood group incompatibility

Carboxyhemoglobin - the forgotten parameter of neonatal hyperbilirubinemia

BACKGROUND

Neonatal hyperbilirubinemia is influenced by a wide variety of factors, one of which is hemolysis. Serious hyperbilirubinemia may lead to a kernicterus with detrimental neurologic sequelae. Patients suffering from hemolytic disease have a higher risk of developing kernicterus. Carbon monoxide (CO), a byproduct of hemolysis or heme degradation, was described by Sjöstrand in the 1960s. It is transported as carboxyhemoglobin (COHb) and exhaled through the lungs. We were interested in a potential correlation between COHb and total serum bilirubin (TSB) and the time course of both parameters.

MATERIALS AND METHODS

We used a point of care (POC) blood gas analyzer and did a retrospective analysis of bilirubin and COHb data collected over a 60-day period.

RESULTS

An arbitrary cut-off point set at 2% COHb identified four patients with hemolytic disease of different origins who required phototherapy. In one patient with atypical hemolytic uremic syndrome (aHUS), COHb preceded the rise in bilirubin by about 2 days. Despite this displacement, there was a moderately good correlation of COHb with TSB levels <15 mg/dL (257 μmol/L) (r2: 0.80) and direct bilirubin (r2: 0.78) in the first patient. For all the four patients and all time points the correlation was slightly lower (r2: 0.59).

CONCLUSIONS

COHb might be useful as a marker for high hemoglobin turnover to allow an earlier identification of newborns at risk to a rapid rise in bilirubin.

Does vitamin C cause hemolysis in premature newborn infants? Results of a multicenter double-blind, randomized, controlled trial

OBJECTIVE

We tested the hypothesis that vitamin C supplementation of premature neonates is associated with hemolysis.

STUDY DESIGN

A double-blind, randomized, controlled trial of vitamin C supplementation (50 mg/day) was undertaken in premature neonates (birth weight, 1000 to 1500 gm). Infants were randomly assigned to receive vitamin C (Ce-Vi-Sol) (n = 32) or placebo (n = 24) for 14 days. Twenty-three subjects per group were required to detect a difference of 1 SD in corrected carboxyhemoglobin values (alpha = 0.05, beta = 0.10).

RESULTS

Day 14 vitamin C levels were lower in control subjects than in supplemented neonates (62 +/- 24 vs 125 +/- 62 micromol/L, p = 0.005). There was no difference in corrected blood carboxyhemoglogin concentrations (0.72 +/- 0.44 vs 0.72 +/- 0.23%; p = 0.95), other parameters of hemolysis, weight gain, blood sampled, presumed septic episodes, necrotizing enterocolitis, feeding intolerance, or transfusion. On day 14, bilirubin values were higher in control subjects than in the supplemented group (77 +/- 37 vs 55 +/- 33 micromol/L; p = 0.04). When a distant outlier in the nonsupplemented group was excluded (163 micromol/L), statistical significance was lost (73 +/- 32 vs 55 +/- 33 micromol/L; p = 0.09).

CONCLUSION

Oral supplementation of premature infants with vitamin C is not associated with evidence of increased erythrocyte destruction, hyperbilirubinemia, or other morbidity.

Severe neonatal jaundice associated with glucose-6-phosphate dehydrogenase deficiency: pathogenesis and global epidemiology

The association of glucose-6-phosphate dehydrogenase deficiency (G6PD def) with severe neonatal jaundice (NJ) and Kernicterus was described just over 30 years ago in reports from Sardinia (1-4) Singapore (5,6) and Greece (7,8). In this review we will examine the progress made since that time in our understanding of the pathophysiology of severe NJ associated with G6PD def, its global epidemiology, and its role in the balanced polymorphism of the G6PD def gene (Gd). Including this review in the Festschrift to Spyros Doxiadis should highlight his contribution to the field. In Greece the search for a new cause of severe NJ and Kernicterus was triggered by his realization that the recognized at the time, causes of severe NJ were absent in a large proportion of the neonates treated at the "Alexandra" Maternity Hospital in Athens (9).

The influence of gestational age on bilirubin conjugation in newborns

Unconjugated, mono- and diconjugated bilirubin levels were determined in serum soon after birth, and followed up for several days. Fourteen preterm neonates were studied with a gestational age below 33 weeks (n = 7) or between 34 and 37 weeks (n = 7), respectively, as well as 19 full-term newborns either untreated (n = 9) or treated by phototherapy (n = 10). Bilirubin and its derivatives were analysed by alkaline methanolysis and spectrometry after separation by thin-layer chromatography. In normal full-term neonates total and unconjugated bilirubin reached peak levels at days 2-4. Thereafter, a decline of 11% per day was detectable. Monoconjugates in serum amounted to 3.1 +/- 1.1% of total pigment and remained at that level. The relative amount of diconjugates increased from 0.55 +/- 0.25% (2-4th postnatal day) to 1.62 +/- 0.99% (9-13th day of life). The rapid decline of unconjugated bilirubin paralleled by an increase of diconjugates are an expression of the maturation process for bilirubin conjugation. The premature neonates with less than 33 weeks gestation exhibited an increase of unconjugated serum bilirubin up to the 4-5th postnatal day, the decline thereafter amounted 2% per day. The fraction of 2.3 +/- 1.1% monoconjugates was small and exhibited only a moderate increase in the follow up. In contrast diconjugates were undetectable or very low and remained at this level. These results suggest the presence of a more severe immaturity as well as a slower maturation process of bilirubin conjugation in preterm newborns.(ABSTRACT TRUNCATED AT 250 WORDS)

Carboxyhemoglobin measurements in the diagnosis of ABO hemolytic disease

To assess hemolysis in hyperbilirubinemic infants with ABO isoimmunization, we measured the carboxyhemoglobin (HbCO) concentrations in full-term infants with ABO isoimmunization during the first week of life. Two groups of infants, ABO compatible (n = 34) and ABO incompatible (n = 30), were further divided into two groups with and without hyperbilirubinemia. All the infants from ABO incompatible pregnancies showed a positive indirect Coombs' test. The groups with hyperbilirubinemia included infants with maximum serum total bilirubin levels about 15 mg/dL and infants treated by phototherapy or exchange transfusion. In the ABO incompatible infants, the HbCO levels in those with hyperbilirubinemia were significantly higher than in infants without hyperbilirubinemia during the first week after birth (p less than 0.01 at 24, 72, 120 hours after birth, respectively). The levels of HbCO in the ABO incompatible infants with hyperbilirubinemia were significantly higher compared with the levels in the ABO compatible infants with hyperbilirubinemia (p less than 0.05 at 24 hours, p less than 0.01 at 72 and 120 hours after birth, respectively). HbCO measurement may facilitate the early diagnosis of hemolytic disease and the prediction of jaundice caused by ABO isoimmunization.

Reduction of neonatal blood eliminates the error induced by oxygen saturation in the spectrophotometric measurement of carboxyhaemoglobin

Blood carboxyhaemoglobin (COHb) is a sensitive index of haemolysis and has been used in assessing the cause of different types of neonatal jaundice. Although the introduction of automated spectrophotometry provides rapid and accurate measurement in adult blood, in neonates oxygenated foetal haemoglobin (HbF) is thought to interfere with COHb measurement. In an attempt to eliminate this problem, the haemoglobin in neonatal blood was reduced with sodium dithionite. Cord blood from 50 infants was measured before and after reduction using an IL-282 co-oximeter; COHb levels fell after reduction. A significant positive correlation was found between apparent COHb% and oxygenation of cord blood. In contrast, no significant correlation was found between these parameters in adult blood where COHb values remained the same or rose slightly after reduction. In 20 healthy non-icteric neonates the mean reduced blood COHb value was not significantly different from the mean COHb value of 23 healthy non-smoking adults. We suggest that COHb in neonatal blood can be simply and accurately measured by the IL-282 co-oximeter provided that the blood is fully reduced.

The role of haemolysis in neonatal hyperbilirubinaemia as reflected in carboxyhaemoglobin levels

Carboxyhaemoglobin levels were measured in 250 consecutive, normal, term newborns and in a group of 75 severely jaundiced infants in an effort to assess the role of haemolysis in non-specific neonatal hyperbilirubinaemia--"Physiologic jaundice"-as well as in severe jaundice of various etiologies. In normal newborns a significant correlation (r=0.3293, p less than 0.001) was found between third-day carboxyhaemoglobin and serum bilirubin levels. Significant correlations were also found between the cord and third day values for carboxyhaemoglobin (infants of non-smoking mothers) for reticulocyte count, and for serum bilirubin. Third day haemoglobin showed no correlation with any of the other parameters including cord haemoglobin. Of the infants with third day carboxyhaemoglobin up to 0.75%, 1.7% had third day serum bilirubin values above 16.0 mg/100 ml. The corresponding percentage for the infants with carboxyhaemoglobin above 0.75% was 6.1%. It is concluded that increased rates of haemolysis due to as yet unspecified caused play an important role in the non-specific hyperbilirubinaemia of normal term newborns. As expected, high levels of carboxyhaemoglobin were found in infants with severe jaundice due to Rhesus and ABO haemolytic disease and glucose-6-phosphate dehydrogenase deficiency but also in jaundiced prematures and in Greek infants with severe jaundice of unknown cause.

Smoking during pregnancy and its effects on the fetus

The smoking habits of a group of pregnant women have been recorded and related to the level of carboxyhaemoglobin (COHb) in the circulating blood. Simultaneous maternal and fetal COHb levels at delivery were estimated and on average the fetal levels were 1.8 times greater than the related maternal level. Samples of fetal blood were exposed to carbon monoxide in vitro and an appreciable shift to the left of the haemoglobin dissociation curve was demonstrated. Cigarette smoking during pregnancy diminishes the oxygen carrying capacity of both fetal and maternal blood, affects maternal oxygenation by increased pulmonary venous admixture and diminishes the oxygen available to the fetus at the tissue level by its effect on fetal oxyhaemoglobin dissociation.

Endogenous production of carbon monoxide in normal and erythroblastotic newborn infants

The endogenous production of carbon monoxide ( V(CO)) in newborn infants was measured by serial determinations of blood carboxyhemoglobin during rebreathing in a closed system. Mean V(CO) in nine full-term infants was 13.7 +/-3.6 mul CO/kg per hr (SD), and in four erythroblastotic infants V(CO) ranged from 37 to 154 mul CO/kg per hr preceding exchange transfusion. Mean red cell life-span (MLS) and total bilirubin production were calculated from V(CO). MLS in normal newborns was 88 +/-15 days (SD), and bilirubin production was 8.5 +/-2.3 mg/kg per 24 hr. This is more than twice the amount of bilirubin normally produced in the adult per kilogram of body weight. Normal infants achieved a net excretion of bilirubin of at least 5.6 +/-2.3 mg/kg per 24 hr (SD) as calculated from the bilirubin production and the measured rise in serum bilirubin concentration.The measurement of V(CO) should prove valuable in the study of red blood cell survival and bilirubin metabolism in the newborn infant.