Introduction

Fetal blood sampling during pregnancy and during labor are both rather complicated procedures, requiring specialized skills on the part of the doctors. They are also unpleasant for the mother, and involve risks for the fetus. Cord blood sampling at birth, on the other hand, is safe and easy. 

When the pH analyzer is located near the labor ward, very little time, effort, and supplies are required for this measurement. There is, of course, a discrepancy between the actual costs of measurement and what is charged for it. 

American studies report that patients are charged between USD 22 and 79 for cord blood measurement at birth [1].  

These charges often play a prominent role in discussions of whether cord blood pH should be routinely tested at birth. A survey on the use of cord blood pH measurement at birth showed that only 27 % of university obstetric units performed the measurement routinely [2].

The American College of Obstetricians and Gynecologists has addressed the issue of cord blood measurement at birth in several guidelines. In recent guidelines, it is advised that a segment of the cord be clamped immediately after birth. 

This segment can later be discarded if the five-minute Apgar score is satisfactory. If, however, serious problems with the neonate's condition persist beyond the first five minutes, the College advises that blood be drawn from the cord segment and sent for blood gas analysis. 

In Germany and the United Kingdom, the Societies of Obstetricians and Gynecologists advise to measure arterial cord blood pH at each birth. In the Netherlands, it is advised to measure cord blood pH at each hospital birth.

In this paper, we present the arguments for and against routinely measuring cord blood pH at birth. For clinicians, however, the main issue regarding cord blood pH measurement at birth, whether it is measured routinely or only in selected cases, remains the correct interpretation of the results. 

This relates not only to the clinical significance of normal and abnormal values, but also to the reliability of the measurement, the factors that influence it and the chosen cutoff points between normal and abnormal pH at birth.

Sampling and measurement

Arterial and venous samples from the cord vessels can be taken immediately after birth, even before clamping and cutting off the cord. It is recommended, though, to clamp the cord first and to use two extra clamps to isolate and remove a 10-20 cm long segment from which samples can be taken later. 

When clamping is started from the newborn end, the umbilical arteries probably to some extent empty towards the placenta before the next clamp is fixed. It is therefore better to place the first clamp at the placental end.

Isolating a section of the cord between clamps offers several advantages. Cord blood can be taken at ease by another person. Measurements can be repeated when there is doubt about the samples; for example, when arterial and venous measurements provide nearly identical pH values. 

Most importantly, the sampling is less likely to intervene with the emotional and medical aspects of childbirth, either in the delivery room or at the operating table. A clamped cord segment can remain at room temperature for 15-30 minutes, without sampling and measurement being affected [3]. 

When the measurement is to be postponed for longer than 30 minutes, samples should be taken in heparinized syringes and kept on ice [4]. The (acid) heparin concentration in these syringes should not exceed 50 IU/mL of blood [5].

It is important that blood samples are taken from both artery and vein. This will provide convincing evidence that the arterial pH was actually derived from an artery. If only one vessel is sampled, it can readily be argued that the pH reflects the venous pH, because sampling the vein is usually easier [6]. 

Obviously, correct sampling also requires appropriate measurement and reporting of values. Instructions of the manufacturer for calibration and quality control of the analyzer should be followed. Presently, most analyzers indicate whether a certain measurement is reliable. There should be adequate safeguards to ensure that cords and pH results are not erroneously assigned to other patients.

Validation and interpretation

If the two pH values diverge sufficiently, one may conclude that both arterial and venous pH have been obtained. How large the difference between the samples should be to ensure this is somewhat arbitrary. 

It is reasonably well-accepted, though, that venous pH should exceed arterial pH by at least 0.03 units to give credence to the results [6]. Data that do not meet this criterion are suspect of representing only venous pH.

Differences between repeated arterial blood pH measurements obtained from the same cord have been reported in a number of studies. These differences result from imprecision of the pH analyzer, biological variation among samples and variations in obtaining and handling the samples. 

Standard deviations (SD) of the differences between two samples were from 0.024 to 0.033 in studies that compared two measurements immediately after birth, and from 0.024 to 0.045 in studies comparing immediate measurements with delayed ones [3,7,8]. 

The 95 % confidence limit is therefore roughly the measured value +/– 0.06 for an individual arterial cord blood pH measurement.

For clinical interpretation of umbilical artery pH some consideration is necessary of what is normal or not. Fetal and neonatal blood pH usually remain fairly constant. During pregnancy, arterial cord blood pH has a mean value of 7.37, with 95 % of measurements being within the range of 7.43 and 7.31 [9]. 

Immediately after birth, mean arterial cord blood pH is lower, with a much larger spread around its mean, attesting both to the challenge of birth and its variation from one person to another. Blood of a deceased fetus or baby is usually around 6.8 and always below 7. 

Such low pH values in a live fetus or baby are indicative of asphyxia.

In a group of women with normal pregnancy and labor, excluding, for example, those with abnormal fetal heart rate tracings, mean arterial cord blood pH varied between 7.24 and 7.31, while lower limits of statistical normality (mean – 2 SD) varied from 7.07 to 7.22 [10]. 

In women with normal pregnancies having cesarean section before the onset of labor, mean arterial cord pH varied between 7.26 and 7.29 with lower limits of statistical normality between 7.16 and 7.21 [10].

A cutoff point for abnormality should be fairly sensitive (80-90 %) while having a positive predictive value of at least 5-10 % in detecting infants who contracted asphyxial damage during (rather than before or after) labor. 

Several studies indicate that cutoff values of 7.00 or 7.05 comply with these expectations [6,11,12]. However, as the 95 % confidence interval of an individual measurement is relatively broad (+0.06 to –0.06), the use of a single cutoff value is unrealistic. 

Therefore, the following classification of arterial cord blood pH values in three categories seems more realistic: normal (when > 7.11), borderline (when ≤ 7.11 and ≥ 6.99) and abnormal (when < 6.99). 

There is statistical and physiological evidence in support of this classification. However, many authors set cutoff points between normal and abnormal arterial cord blood pH at birth at unrealistically high levels, varying between 7.10 and 7.20.

Clinical significance of normal and abnormal values

Cord blood pH values at birth do not provide a highly sensitive and specific indication of permanent damage. There are several obvious reasons for this. Fetuses and neonates may recover more or less quickly from an asphyxial episode, and such an episode may or may not result in permanent organ damage. Thus, the majority of infants born with low cord blood pH enjoy normal psychomotor development [11,12]. 

On the other hand, unrecorded asphyxial episodes in utero or in early infancy may engender permanent damage. This explains why many brain-damaged infants were born with normal cord blood pH. In addition, permanent damage may follow genetic or infectious disease without any history of disturbed fetal homeostasis.

If umbilical pH cannot predict long-term outcome, what, if any, is its significance? A normal cord blood pH indicates that the fetus was in good condition at birth. 

When it corresponds with the clinical appearance, the obstetrician can conclude that the baby does not need immediate specialized care. Umbilical pH also provides an important element in the evaluation of intrapartum care, both in cases with good and with worrying outcome. 

Normal cord blood pH at birth provides evidence that the management of birth did not harm the fetus. It may also indicate in retrospect that emergency obstetric intervention was either effective or superfluous.

 A normal cord blood pH at birth will, with reasonable certainty, exclude oxygen deficiency during labor. This is also true for neonates born with normal umbilical pH, but clinically depressed. The cause for these low Apgar scores should then be sought in factors such as maternal medication, birth trauma, congenital anomalies, etc.

An abnormal cord blood pH indicates that the fetus was in a state of biochemical decompensation at the time of birth. If this is accompanied by a low Apgar score, resuscitation procedures will have been initiated before the abnormal pH result was available. 

Adequate ventilation will usually suffice to overcome the acidemia. However, documentation of perinatal asphyxia indicates that the neonate should be cared for in an intensive care unit, because of the increased risk of hypoxic ischemic encephalopathy. 

When an infant is born with abnormal cord blood pH but normal Apgar scores, the outlook is more favorable, but full examination is necessary, including observation at least until the blood pH has returned to normal. 

Abnormal cord blood pH values should aid obstetricians and perinatologists to critically evaluate their care during pregnancy and labor, for instance in considering whether and when fetal blood sampling should have been performed or what signs of impending problems were inadequately acted upon.

Finally, parents may expect a prognosis on the future development of their baby. Such prognosis cannot be given on the basis of cord blood pH or Apgar score. Careful observation of the neonate during the first days of life will be necessary. 

If signs of moderate or serious encephalopathy (Sarnat II or III) develop, the risk of death or permanent disability is considerable, and a causal relationship with perinatal asphyxia will be likely [13]. Absence of these signs is associated with good prognosis. 

Residual neurological abnormalities as a result of perinatal asphyxia are then unlikely [11,12,14,15].

Advantages of routine cord blood pH measurement

An arterial cord blood pH, correctly sampled, measured, validated and interpreted, provides the most objective and sensitive index for asphyxia during labor. Its measurement is non-invasive and also inexpensive, provided that the delivery room is equipped, as it should be, with a pH analyzer for testing fetal blood samples taken during labor. 

The major advantage of routine measurement is that caregivers receive immediate feedback on their care during childbirth, and can learn from that information. Many neonates with grave acidemia will be detected earlier, leading to a more timely reaction and, in some instances, prevention of further complications. 

In addition, routine measurement offers new opportunities for audit. Regular discussion of cases with abnormal cord blood pH at birth will add to the quality of hospital policies and their application. 

With regard to later claims or complaints, a normal cord pH practically excludes asphyxia during labor. Finally, routine collection of cord blood pH offers reliable outcome measures for retrospective studies.

Disadvantages of routine cord blood pH measurement

About 98 % of routinely measured arterial cord blood pH will turn out to be normal or borderline, while about 90 % of infants with abnormal results will not suffer permanent consequences. In many cases, perinatal asphyxia can be ascertained without routine pH measurement. 

Routine measurement means extra costs and extra workload. Despite this extra effort, cord blood pH will remain unreported in at least 10 % of births. Moreover, if only validated measurements are accepted, around 25 % of cord blood values will remain unreported or dubious in terms of whether they reflect arterial or venous pH [10]. 

The lack of universally agreed cutoff points between normal and abnormal pH leaves room for conflicting interpretations. Moreover, in case of litigation, adequate documentation of unexpected acidemia at birth may well have negative consequences for obstetricians who seek to document and audit their practice.