- Email: [email protected]
Outcome of fetuses with lactic acidemia
FETUS, PLACENTA, AND NEWBORN
Outcome of fetuses with lactic acidemia Joseph S. Suidan, M.D., and Bruce K. Young, M.D. New York. New York Umbilical arterial lactate, pH, Po 2 , Pco 2 , and base deficit were measured at delivery of 468 live-born infants. The correlation between the 1-minute Apgar score and lactate (r = -0.34) was comparable to that between the t-minute Apgar scare and pH (r = 0.30). The multiple correlation coefficient between the 1-minute Apgar score and all five parameters combined (lactate, pH, Po 2 , Peo 2 , and base deficit) was 0.36. This was not significantly higher than the simple correlation coefficient for pH alone (0.30). Similar correlation coefficients were noted for the 5-minute Apgar score. The dividing point between normal and pathologic lactate levels was 3.70 mmol/L. A pH <7.20 and a lactate level 2:3.70 mmol/L had the same sensitivity, specificity, and positive and negative predictive values for low Apgar scores «7). While pH and lactate are equally correlated with fetal outcome, their combination with each other and with other blood gas parameters does not predict outcome better than either pH or lactate alone. (AM. J. OesTET. GVNECOL. 150:33, 1984.)
The lactic acid found in human fetal blood is thought to be primarily fetal in origin,I-5 and its level has been shown to increase with fetal asphyxia and neonatal depression.3 - 6 Until recently, however, the measurement of fetal lactate levels had little applicability in the assessment of fetal status during labor because of the volume of blood and the time required to perform the assay.7 However, the introduction of an enzymatic electrochemical sensor has made possible the measurement of lactate levels in capillary blood samples within minutes.s An immediate benefit that derives from this is the potential use of lactate measurement in fetal blood samples for the assessment of fetal stress during labor. In this study, we investigated the relationship between fetal lactacidemia at the time of delivery and fetal outcome as judged by the Apgar score. We also compared the fetal lactate level to other measures of fetal acid-base balance as predictors of fetal outcome.
Material and methods The study population consisted of 468 live-born infants in whom it was possible to obtain adequate umbilical blood samples. The incidences of depression and low birth weight in the study population and in the
From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, New York University Medical Center. Received for puhlication November 3,1983; revised February 1, 1984; accepted March 9, 1984. Reprint requests: Bruce K. Young, M .D., Department of Obstetrics and Gynecology, New York University Medical Center, 550 First Ave. , New York, NY 10016.
general obstetric population were similar. The incidence of low I-minute Apgar scores «7) was 10.9% in the study population and 10.2% in the general population . The incidence of low 5-rriinute Apgar scores «7) was 3.4% in the study population and 2.4% in the general population. The incidence of low birth weight «2500 gm) was 10.5% in the study group and 9.9% in the general population. At the time of delivery, umbilical arterial and venous blood samples were obtained immediately after clamping of the cord. The specimens were collected anaerobically into pre heparinized polyethylene syringes and analyzed immediately for pH, Po 2, PC02, and base deficit with a blood gas pH analyzer (Model 165, Corning Medical, Medfield, Massachusetts). The samples were also analyzed for whole-blood lactate by means of the enzymatic-electrochemical method incorporated in the lactate analyzer (Model 640, Kontron Analytical, Everett, Massachusetts), in accordance with the technique outlined by Soutter et al. 9 Briefly, this method relies on the specific and irreversible conversion of lactate to pyruvate by the nicotinamide adenine dinucleotide-independent lactate dehydrogenase (cytochrome b2) from yeast, with transfer of the [feed electrons to hexacyanoferrate and registration of the current necessary to reoxidize the hexacyanoferrate. The current measured is linearly related to the lactate concentration in the solution under study. The performance characteristics of this method are similar to those of the classical but more time-consuming enzymatic-spectrophotometric assays, and it is highly correlated with them .7. 9 It allows measurement of the lactate concen-
33
34
Suidan and Young
September I, 1984 Am. J. Obstet. Gynecol.
Table I. Correlation coefficients between the
Table III. Umbilical arterial pH and lactate in
I-minute Apgar score and umbilical artery parameters (n = 468)
vigorous and depressed newborn infants
Parameter
Lactate pH Po. Peo. Base deficit
Correlation coefficient (absolute value)
Significance*
r = 0.34 r =
0.30
r = 0.10 r =
0.20
r = 0.17
NS
P < 0.001 P < 0.05 p< 0.01
*Significance of the difference between the correlation coefficient of each parameter and that of lactate.
Table II. Correlation coefficients between the 5-minute Apgar score and umbilical artery parameters (n = 468) Parameter
Lactate pH Po, Peo, Base deficit
Correlation coefficient (absolute value) r =
0.34
r = 0.31 r = 0.04 r = 0.23 r =
0.15
Significance*
NS
P < 0.001
NS
P < 0.01
*Significance of the difference between the correlation coefficient of each parameter and that of lactate. tration on as little as 50 JLl of blood within 5 minutes of sample withdrawal. Statistical analysis was done by means of simple and multiple linear regressions to calculate the simple and multiple correlation coefficients between the Apgar scores and various combinations of the umbilical artery parameters. The Fisher transformation was used to test the significance of differences between the correlation coefficients. The t test was used to compare means, and the X2 test was used to compare the incidences of low Apgar scores associated with different values of the umbilical artery parameters. Statistical significance was defined at the 0.05 level. Results
The simple correlation coefficients between the I-minute Apgar score and each of the parameters in the umbilical artery are shown in Table 1. The correlation of lactate with the I-minute Apgar score was not significantly stronger than that of pH. In comparison with the remaining acid-base parameters, lactate was significantly better correlated with the Apgar score. Interestingly, P0 2 , Pc0 2 , and base deficit were only minimally related to the Apgar score. Combining the pH and lactate and then all five parameters together (lactate, pH, P0 2 , Pc0 2 , and base deficit) in a multiple regression analysis gave a multiple correlation coefficient with the I-minute Apgar score of 0.36 in each case.
Parameter
VigoTOus*
Depressedt
(n = 417)
(n = 51)
Significance (t-test)
pH Lactate (mmoIlL)
7.26 ± 0.07 2.55 ± 1.23
7.18 ± 0.13 4.20 ± 2.80
p < 0.001 P < 0.001
Results are shown as mean ± SD. *One-minute Apgar score ;:::;7. tOne-minute Apgar score <7. This was not significantly higher than the simple correlation coefficient between the I-minute Apgar score and the pH (0.30). Also, the correlation of lactate with the 5-minute Apgar score was not significantly better than that of pH (Table II). When multiple regression analysis of the 5-minute Apgar score was performed against pH and lactate, then against all five parameters combined, the multiple correlation coefficients (r = 0.36 and r = 0.38, respectively) were not significantly higher than the simple correlation coefficient with pH alone (0.31). Table III shows the pH and lactate levels in vigorous (I-minute Apgar score 2::7) and in depressed (I-minute Apgar score <7) newborn infants. Depressed newborn infants had a significantly lower mean pH and a significantly higher mean lactate level than vigorous ones (p < 0.001). The cutoff point between normal and pathologic lactate levels was defined as the abscissa of the intersection of the normal distribution curve of lactate in vigorous babies with the normal distribution curve of lactate in depressed babies. Both curves were defined by the means and standard deviations shown in Table III. Mathematical computation of the abscissa of the intersection gave a value of 3.69 mmol/L as the upper normal limit of lactate in the umbilical artery. Applying the same analysis to the pH gave 7.20 as the lower normal limit of pH. This is not surprising since 7.20 is generally accepted as the dividing point between normal and acidotic fetal pH values.1O-14 In order to assess the clinical significance of these mathematically defined limits, we examined the incidence of low I-minute Apgar scores with gradually increasing lactate levels (Table IV). The incidence of depression associated with a lactate level in the range of 1. 70 to 3.69 mmol/L was not significantly higher than that associated with a lactate level below 1.70 mmol/L (p> 0.2). However, the group with a lactate level in the range of 3.70 to 5.69 mmol/L had a significantly higher incidence of depression than the group with a lactate level in the range of 1.70 to 3.69 mmol/L (p < 0.01). This confirms that a lactate level of 2::3.70 mmol/L is pathologic in that it is associated with a significant increase in the incidence of low I-minute Apgar scores.
Outcome of fetuses with lactic acidemia
Volume 150 Number I
Table IV. Incidence of low I-minute Apgar
Table V. Incidence of low I-minute Apgar
scores « 7) with the various levels of lactate in the umbilical artery blood
scores «7) with the various levels of pH in the umbilical artery blood
35
Incidence of low i-min Apgar scores «7)
Incidence of low i-min Apgar scores «7)
Lactate level (mmollL)
%
n
pH
%
n
<1.70 1.70-3.69 3.70-5.69 5.70-7.69 ?7.70
5.2 8.4 21 25 100
61116 22/262
?7.30 7.20-7.29 7.10-7.19 7.00-7.09 <7.00
6.0 7.5 22 32 75
81134 18/239
15173 3112 515
16172 6/19 3/4
Table VI. Comparison of a low pH «7.20) to a high lactate level (2:3.70 mmollL) in the umbilical artery blood as predictors of a low I-minute Apgar score «7) Lactate ?3.70 mmollL
pH <7.20 Parameter
%
Proportion of all fetuses with abnormal values Sensitivity Specificity Positive predictive value Negative predictive value
I
I
n
%
20
951468
19
90/468
49 83 26 93
25151 347/417 25195 347/373
45 84 26 93
23/51 350/417 23/90 350/378
n
Table VII. Comparison of a low pH «7.20) to a high lactate level (2:3.70 mmol/L) in the umbilical artery blood as predictors of a low 5-minute Apgar score «7) pH <7.20 Parameter
%
Sensitivity Specificity Positive predictive value Negative predictive value
50 81 8.4 98
I
When the same concept was applied to the pH (Table V), there was no significant difference in the incidence of depression associated with a pH 2:7.30 compared to that associated with a pH in the range of 7.20 to 7.29 (p > 0.5). However, babies with a pH in the range of 7.10 to 7.19 had a significantly higher incidence of low I-minute Apgar scores than those with a pH in the range of 7.20 to 7.29 (p < 0.001), which confirms the clinical value of using 7.20 as a cutoff point between normal and acidotic pH values. Figs. 1 to 4 show the subdivisions of newborn infants with respect to Apgar scores, pH, and lactate with the dividing points of 7 for Apgar score, 7.20 for pH, and 3.70 mmol/L for lactate. The incidence oflow I-minute Apgar scores «7) was the same in acidotic babies (pH <7.20) and in hyperlactacidemic babies (26% in either case). Also the incidence of good I-minute Apgar scores (2:7) was the same in babies with a pH 2:7.20 and in babies with a lactate level <3.70 mmol/L (93% in
Lactate ?3.70 mmoliL n
%
8116 365/452 8/95 365/373
50 82 8.9 98
I
n 8116 370/452 8/90 370/378
each case). Good 5-minute Apgar scores were as frequent in babies with a pH 2:7.20 as in babies with a lactate level <3.70 mmollL (98% in each case), and the incidence of low 5-minute Apgar scores in acidotic babies (8.4%) was similar to that in hyperlactacidemic babies (8.9%). Tables VI and VII outline the characteristics of a pH <7.20 and a lactate level 2:3.70 mmollL as tests for the prediction oflow Apgar scores «7). The sensitivity 14, 16 of the test is the proportion of babies with a low Apgar score who have an abnormal test value (pH <7.20, lactate level 2:3.70 mmoIlL). The specificity l4, 16 is the proportion of babies with a good Apgar score who have a normal test value (pH 2:7.20, lactate level <3.70 mmoIlL). The positive predictive value l4 , 16 is the proportion of babies with an abnormal test value who have a low Apgar score. The negative predictive value '4 . 16 is the proportion of babies with a normal test value who have a good Apgar score. Tables VI and VII show that
36
Suidan and Young
September 1, 1984 Am. J. Obstet. Gynecol.
I.minut. Apgar scor. <7
5.minut. Apgar scor.
~7
<7
Z
a. >-
.. •.. a
-
Z
~
7.20
26
a. >-
..
--
347
-
70
pH, (The numbers refer to the number of newborn infants in each category.) .
I.minut. Apgar scor. <7
a
-
28
350
23
67
•.. ~
~3.70 mM/L
:i < 7.20 E
8
87
::::t
Fig. 1. Relationship'between the I-minute Apgar score and
.! < 3.70mM/L >-
365
~
25
::::t
--•"
8
a a
a
: i < 7.20 E
~7.20
•..
.~
.. -a .." :i
~7
E
:)
Fig. 2. Relationship between the I-minute Apgar score and lactate.
both a pH <7.20 and a lactate level 2::3.70 mmol/L in the umbilical artery had almost exactly the same characteristics for the prediction of a low I-minute Apgar score as well as for the prediction of a low 5-minute Apgar score. Moreover, the incidence of abnormal lactate values (19%) was almost the same as that of abnormal pH values (20%).
Comment Fetal blood sampling for pH and blood gases during labor is widely accepted in the management of the fetus at risk, and its value has been demonstrated in many studies. IO - 14 The recent introduction of rapid lactate assays on microsamples provides the obstetrician with a new means of assessing the biochemical condition of the fetus during labor. Actually, fetal lactate levels have been thought to provide better information about the
Fig. 3. Relationship between the 5-minute Apgar score and
pH.
fetal status than other acid-base data,S especially since lactic acid is the major specific end product of anaerobic metabolism and its accumulation at high concentrations in brain tissue causes edema and necrosis. IS In the present study, we have used the 1- and 5-minute Apgar scores for the evaluation of fetal outcome. Although the relationship between the Apgar scores and subsequent neonatal and neurological development is far from perfect, Apgar scores remain widely used for clinical assessment of the newborn infant. 16 We examined the relationship between fetal lactacidemia at the time of delivery and fetal outcome and compared this relationship to those existing between other acid-base parameters and fetal outcome. We found that the umbilical arterial lactate level correlates as well as the pH with Apgar scores, while both pH and lactate have better correlations with Apgar scores than the other acid-base parameters. We also found that combining the information from pH and lactate or from all five parameters (pH, lactate, Po 2 , Pco 2 , and base deficit) does not improve the ability to predict Apgar scores. The pathologic range of lactate was found to be 2::3.70 mmol/L, which is in general agreement with values of 3.30 mmol/L5 and 3.90 mmol/U reported in the literature. Lactate values 2::3.70 mmol/ L have about the same correlation as pH values <7.20 with low Apgar scores. Also lactate values <3.70 mmoll L have almost the same correlation as pH values 2::7.20 with good Apgar scores. The base deficit, which is an indirect reflection of metabolic acidosis, did not correlate as well as pH or lactate with fetal outcome. Nevertheless, it had significant correlations with both the I-minute Apgar score (r = -0.17, P < 0.001) and the 5-minute Apgar score (r = -0.15, P < 0.002). The relatively limited correlation between Apgar scores and both pH and lactate may be explained by
Outcome of fetuses with lactic acidemia
Volume 150 Number I
5.minute Apgar score
..•
!
.!" < 3.70mM/L >-
..•...
<7
~7
8
370
D
.-"
:i E
~3.70mM/L
that lactate is the major metabolic component in fetal acidosis. However, the combination of pH and lactate and of multiple fetal biochemical parameters adds nothing to our capability to predict neonatal depression available by measurement of either pH or lactate alone. The Po 2, PC02, and base deficit are poor predictors of fetal outcome as described by the Apgar score.
REFERENCES
D
-
37
8
82
:::»
Fig. 4. Relationship between the 5-minute Apgar score and lactate. many factors. While pH and lactate correlate better than the other acid-base parameters with Apgar scores, there are numerous metabolic and pathophysiologic changes that they do not reflect. Thus alterations in the fetal neurological status, drugs administered to the mother, anesthesia, and the effect of delivery are not necessarily reflected in pH and lactate measurements. Furthermore, the Apgar score is a gross general physical examination which reflects in a general way neonatal condition. Factors such as resuscitative procedures, prematurity, endotracheal intubation for suction of meconium, and other factors such as stimulation and temperature may affect the Apgar score. Thus it is not surprising that both pH and lactate correlate only modestly with Apgar scores. Finally, fetal acidosis may be secondary to sustained maternal metabolic acidosis and not indicative of fetal compromise. 12 If the baby is delivered during this time, the Apgar scores may be normal in the presence of some degree of fetal acidosis or preacidosis. However, if such acidosis is persistent, it may have a deleterious effect on fetal metabolism, so that the timing of delivery is relevant to the relationship between this secondary fetal acidosis and Apgar scores. Assuming that the capillary scalp blood values would not be more strongly correlated than the umbilical values with the Apgar scores, we conclude that lactate measurement in fetal blood is as good a means for monitoring the fetus in labor as the pH. Its close relationship to pH measurements probably reflects the fact
1. Hendricks CH. Studies on lactic acid metabolism in pregnancy and labor. AM j OBSTET GYNECOL 1957;73:492. 2. Otey E, Stenger V, Eitzman D, Andersen T, Gessner I, Prystowsky H. Movements of lactate and pyruvate in the pregnant uterus of the human. AM j OBSTET GYNECOL 1964;90:747. 3. Low jA, Pancham SR, Worthington D, Boston RW. The acid-base and biochemical characteristics of intrapartum fetal asphyxia. AM j OBSTET GYNECOL 1975;121:446. 4. Daniel SS, Adamsons K, james LS. Lactate and pyruvate as an index of prenatal oxygen deprivation. Pediatrics 1966;37:942. 5. Hsiung R, Piquard F, Schaefer A. Machu j, Haberey P, Dellenbach P. Lactaplasmie maternelle et foe tale au cours de I'accouchement. II. Accouchements pathologiques. Rev Fr Gynecol Obstet 1981;76:853. 6. Lin C-C, Moawad AH, Rosenow Pj, River P. Acid-base characteristics of fetuses with intrauterine growth retardation during labor and delivery. AM j OBSTET GYNECOL 1980;137:553. 7. Kragenings J. Methodology oflactate assay. In: Lactate in acute conditions. International Symposium. Basel: S. Karger AG. 1979:20. 8. Racine P, Engelhardt R. Higelin jC, Mindt W. An instrument for the rapid determination of L-lactate in biological fluids. Med Instrum 1975;9: 11. 9. Soutter WP, Sharp F, Clark DM. Bedside estimation of whole blood lactate. Br j Anaesth 1978:50:445. 10. Bretscher j, Saling E. pH values in the human fetus during labor. AM j OBSTET GYNECOL 1967;97:906. 11. Hon EH, Khazin AF, Paul RH. Biochemical studies of the fetus. II. Fetal pH and Apgar scores. Obstet Gynecol 1969;33:237. 12. Bowe ET, Beard RW, Finster M, Poppers Pj, Adamsons K, james LS. Reliability of fetal blood sampling. Maternal-fetal relationships. AM j OBSTET GYNECOL 1970; 107:279. 13. Miller FC. Prediction of acid-base values from intrapartum fetal heart rate data and their correlation with scalp and funic values. Clin Perinatol 1982;9:353. 14. Banta HD. Thacker SB. Assessing the costs and benefits of electronic fetal monitoring. Obstet Gynecol Surv 1979; 34:627. 15. Myers RE. Wagner KR, Courten GD. Lactic acid accumulation in tissue as cause of brain injury and death in cardiogenic shock from asphyxia. In: Lauersen NH, Hochberg HM, eds. Clinical perinatal biochemical monitoring. Baltimore: Williams & Wilkins, 1981: 11. 16. Niswander KR. Asphyxia in the fetus and cerebral palsy. In: Pitkin RM, Zlatnik Fj, eds. The year book of obstetrics and gynecology. Chicago: Year Book Medical Publishers, Inc., 1983: 107.
Outcome of fetuses with lactic acidemia Joseph S. Suidan, M.D., and Bruce K. Young, M.D. New York. New York Umbilical arterial lactate, pH, Po 2 , Pco 2 , and base deficit were measured at delivery of 468 live-born infants. The correlation between the 1-minute Apgar score and lactate (r = -0.34) was comparable to that between the t-minute Apgar scare and pH (r = 0.30). The multiple correlation coefficient between the 1-minute Apgar score and all five parameters combined (lactate, pH, Po 2 , Peo 2 , and base deficit) was 0.36. This was not significantly higher than the simple correlation coefficient for pH alone (0.30). Similar correlation coefficients were noted for the 5-minute Apgar score. The dividing point between normal and pathologic lactate levels was 3.70 mmol/L. A pH <7.20 and a lactate level 2:3.70 mmol/L had the same sensitivity, specificity, and positive and negative predictive values for low Apgar scores «7). While pH and lactate are equally correlated with fetal outcome, their combination with each other and with other blood gas parameters does not predict outcome better than either pH or lactate alone. (AM. J. OesTET. GVNECOL. 150:33, 1984.)
The lactic acid found in human fetal blood is thought to be primarily fetal in origin,I-5 and its level has been shown to increase with fetal asphyxia and neonatal depression.3 - 6 Until recently, however, the measurement of fetal lactate levels had little applicability in the assessment of fetal status during labor because of the volume of blood and the time required to perform the assay.7 However, the introduction of an enzymatic electrochemical sensor has made possible the measurement of lactate levels in capillary blood samples within minutes.s An immediate benefit that derives from this is the potential use of lactate measurement in fetal blood samples for the assessment of fetal stress during labor. In this study, we investigated the relationship between fetal lactacidemia at the time of delivery and fetal outcome as judged by the Apgar score. We also compared the fetal lactate level to other measures of fetal acid-base balance as predictors of fetal outcome.
Material and methods The study population consisted of 468 live-born infants in whom it was possible to obtain adequate umbilical blood samples. The incidences of depression and low birth weight in the study population and in the
From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, New York University Medical Center. Received for puhlication November 3,1983; revised February 1, 1984; accepted March 9, 1984. Reprint requests: Bruce K. Young, M .D., Department of Obstetrics and Gynecology, New York University Medical Center, 550 First Ave. , New York, NY 10016.
general obstetric population were similar. The incidence of low I-minute Apgar scores «7) was 10.9% in the study population and 10.2% in the general population . The incidence of low 5-rriinute Apgar scores «7) was 3.4% in the study population and 2.4% in the general population. The incidence of low birth weight «2500 gm) was 10.5% in the study group and 9.9% in the general population. At the time of delivery, umbilical arterial and venous blood samples were obtained immediately after clamping of the cord. The specimens were collected anaerobically into pre heparinized polyethylene syringes and analyzed immediately for pH, Po 2, PC02, and base deficit with a blood gas pH analyzer (Model 165, Corning Medical, Medfield, Massachusetts). The samples were also analyzed for whole-blood lactate by means of the enzymatic-electrochemical method incorporated in the lactate analyzer (Model 640, Kontron Analytical, Everett, Massachusetts), in accordance with the technique outlined by Soutter et al. 9 Briefly, this method relies on the specific and irreversible conversion of lactate to pyruvate by the nicotinamide adenine dinucleotide-independent lactate dehydrogenase (cytochrome b2) from yeast, with transfer of the [feed electrons to hexacyanoferrate and registration of the current necessary to reoxidize the hexacyanoferrate. The current measured is linearly related to the lactate concentration in the solution under study. The performance characteristics of this method are similar to those of the classical but more time-consuming enzymatic-spectrophotometric assays, and it is highly correlated with them .7. 9 It allows measurement of the lactate concen-
33
34
Suidan and Young
September I, 1984 Am. J. Obstet. Gynecol.
Table I. Correlation coefficients between the
Table III. Umbilical arterial pH and lactate in
I-minute Apgar score and umbilical artery parameters (n = 468)
vigorous and depressed newborn infants
Parameter
Lactate pH Po. Peo. Base deficit
Correlation coefficient (absolute value)
Significance*
r = 0.34 r =
0.30
r = 0.10 r =
0.20
r = 0.17
NS
P < 0.001 P < 0.05 p< 0.01
*Significance of the difference between the correlation coefficient of each parameter and that of lactate.
Table II. Correlation coefficients between the 5-minute Apgar score and umbilical artery parameters (n = 468) Parameter
Lactate pH Po, Peo, Base deficit
Correlation coefficient (absolute value) r =
0.34
r = 0.31 r = 0.04 r = 0.23 r =
0.15
Significance*
NS
P < 0.001
NS
P < 0.01
*Significance of the difference between the correlation coefficient of each parameter and that of lactate. tration on as little as 50 JLl of blood within 5 minutes of sample withdrawal. Statistical analysis was done by means of simple and multiple linear regressions to calculate the simple and multiple correlation coefficients between the Apgar scores and various combinations of the umbilical artery parameters. The Fisher transformation was used to test the significance of differences between the correlation coefficients. The t test was used to compare means, and the X2 test was used to compare the incidences of low Apgar scores associated with different values of the umbilical artery parameters. Statistical significance was defined at the 0.05 level. Results
The simple correlation coefficients between the I-minute Apgar score and each of the parameters in the umbilical artery are shown in Table 1. The correlation of lactate with the I-minute Apgar score was not significantly stronger than that of pH. In comparison with the remaining acid-base parameters, lactate was significantly better correlated with the Apgar score. Interestingly, P0 2 , Pc0 2 , and base deficit were only minimally related to the Apgar score. Combining the pH and lactate and then all five parameters together (lactate, pH, P0 2 , Pc0 2 , and base deficit) in a multiple regression analysis gave a multiple correlation coefficient with the I-minute Apgar score of 0.36 in each case.
Parameter
VigoTOus*
Depressedt
(n = 417)
(n = 51)
Significance (t-test)
pH Lactate (mmoIlL)
7.26 ± 0.07 2.55 ± 1.23
7.18 ± 0.13 4.20 ± 2.80
p < 0.001 P < 0.001
Results are shown as mean ± SD. *One-minute Apgar score ;:::;7. tOne-minute Apgar score <7. This was not significantly higher than the simple correlation coefficient between the I-minute Apgar score and the pH (0.30). Also, the correlation of lactate with the 5-minute Apgar score was not significantly better than that of pH (Table II). When multiple regression analysis of the 5-minute Apgar score was performed against pH and lactate, then against all five parameters combined, the multiple correlation coefficients (r = 0.36 and r = 0.38, respectively) were not significantly higher than the simple correlation coefficient with pH alone (0.31). Table III shows the pH and lactate levels in vigorous (I-minute Apgar score 2::7) and in depressed (I-minute Apgar score <7) newborn infants. Depressed newborn infants had a significantly lower mean pH and a significantly higher mean lactate level than vigorous ones (p < 0.001). The cutoff point between normal and pathologic lactate levels was defined as the abscissa of the intersection of the normal distribution curve of lactate in vigorous babies with the normal distribution curve of lactate in depressed babies. Both curves were defined by the means and standard deviations shown in Table III. Mathematical computation of the abscissa of the intersection gave a value of 3.69 mmol/L as the upper normal limit of lactate in the umbilical artery. Applying the same analysis to the pH gave 7.20 as the lower normal limit of pH. This is not surprising since 7.20 is generally accepted as the dividing point between normal and acidotic fetal pH values.1O-14 In order to assess the clinical significance of these mathematically defined limits, we examined the incidence of low I-minute Apgar scores with gradually increasing lactate levels (Table IV). The incidence of depression associated with a lactate level in the range of 1. 70 to 3.69 mmol/L was not significantly higher than that associated with a lactate level below 1.70 mmol/L (p> 0.2). However, the group with a lactate level in the range of 3.70 to 5.69 mmol/L had a significantly higher incidence of depression than the group with a lactate level in the range of 1.70 to 3.69 mmol/L (p < 0.01). This confirms that a lactate level of 2::3.70 mmol/L is pathologic in that it is associated with a significant increase in the incidence of low I-minute Apgar scores.
Outcome of fetuses with lactic acidemia
Volume 150 Number I
Table IV. Incidence of low I-minute Apgar
Table V. Incidence of low I-minute Apgar
scores « 7) with the various levels of lactate in the umbilical artery blood
scores «7) with the various levels of pH in the umbilical artery blood
35
Incidence of low i-min Apgar scores «7)
Incidence of low i-min Apgar scores «7)
Lactate level (mmollL)
%
n
pH
%
n
<1.70 1.70-3.69 3.70-5.69 5.70-7.69 ?7.70
5.2 8.4 21 25 100
61116 22/262
?7.30 7.20-7.29 7.10-7.19 7.00-7.09 <7.00
6.0 7.5 22 32 75
81134 18/239
15173 3112 515
16172 6/19 3/4
Table VI. Comparison of a low pH «7.20) to a high lactate level (2:3.70 mmollL) in the umbilical artery blood as predictors of a low I-minute Apgar score «7) Lactate ?3.70 mmollL
pH <7.20 Parameter
%
Proportion of all fetuses with abnormal values Sensitivity Specificity Positive predictive value Negative predictive value
I
I
n
%
20
951468
19
90/468
49 83 26 93
25151 347/417 25195 347/373
45 84 26 93
23/51 350/417 23/90 350/378
n
Table VII. Comparison of a low pH «7.20) to a high lactate level (2:3.70 mmol/L) in the umbilical artery blood as predictors of a low 5-minute Apgar score «7) pH <7.20 Parameter
%
Sensitivity Specificity Positive predictive value Negative predictive value
50 81 8.4 98
I
When the same concept was applied to the pH (Table V), there was no significant difference in the incidence of depression associated with a pH 2:7.30 compared to that associated with a pH in the range of 7.20 to 7.29 (p > 0.5). However, babies with a pH in the range of 7.10 to 7.19 had a significantly higher incidence of low I-minute Apgar scores than those with a pH in the range of 7.20 to 7.29 (p < 0.001), which confirms the clinical value of using 7.20 as a cutoff point between normal and acidotic pH values. Figs. 1 to 4 show the subdivisions of newborn infants with respect to Apgar scores, pH, and lactate with the dividing points of 7 for Apgar score, 7.20 for pH, and 3.70 mmol/L for lactate. The incidence oflow I-minute Apgar scores «7) was the same in acidotic babies (pH <7.20) and in hyperlactacidemic babies (26% in either case). Also the incidence of good I-minute Apgar scores (2:7) was the same in babies with a pH 2:7.20 and in babies with a lactate level <3.70 mmol/L (93% in
Lactate ?3.70 mmoliL n
%
8116 365/452 8/95 365/373
50 82 8.9 98
I
n 8116 370/452 8/90 370/378
each case). Good 5-minute Apgar scores were as frequent in babies with a pH 2:7.20 as in babies with a lactate level <3.70 mmollL (98% in each case), and the incidence of low 5-minute Apgar scores in acidotic babies (8.4%) was similar to that in hyperlactacidemic babies (8.9%). Tables VI and VII outline the characteristics of a pH <7.20 and a lactate level 2:3.70 mmollL as tests for the prediction oflow Apgar scores «7). The sensitivity 14, 16 of the test is the proportion of babies with a low Apgar score who have an abnormal test value (pH <7.20, lactate level 2:3.70 mmoIlL). The specificity l4, 16 is the proportion of babies with a good Apgar score who have a normal test value (pH 2:7.20, lactate level <3.70 mmoIlL). The positive predictive value l4 , 16 is the proportion of babies with an abnormal test value who have a low Apgar score. The negative predictive value '4 . 16 is the proportion of babies with a normal test value who have a good Apgar score. Tables VI and VII show that
36
Suidan and Young
September 1, 1984 Am. J. Obstet. Gynecol.
I.minut. Apgar scor. <7
5.minut. Apgar scor.
~7
<7
Z
a. >-
.. •.. a
-
Z
~
7.20
26
a. >-
..
--
347
-
70
pH, (The numbers refer to the number of newborn infants in each category.) .
I.minut. Apgar scor. <7
a
-
28
350
23
67
•.. ~
~3.70 mM/L
:i < 7.20 E
8
87
::::t
Fig. 1. Relationship'between the I-minute Apgar score and
.! < 3.70mM/L >-
365
~
25
::::t
--•"
8
a a
a
: i < 7.20 E
~7.20
•..
.~
.. -a .." :i
~7
E
:)
Fig. 2. Relationship between the I-minute Apgar score and lactate.
both a pH <7.20 and a lactate level 2::3.70 mmol/L in the umbilical artery had almost exactly the same characteristics for the prediction of a low I-minute Apgar score as well as for the prediction of a low 5-minute Apgar score. Moreover, the incidence of abnormal lactate values (19%) was almost the same as that of abnormal pH values (20%).
Comment Fetal blood sampling for pH and blood gases during labor is widely accepted in the management of the fetus at risk, and its value has been demonstrated in many studies. IO - 14 The recent introduction of rapid lactate assays on microsamples provides the obstetrician with a new means of assessing the biochemical condition of the fetus during labor. Actually, fetal lactate levels have been thought to provide better information about the
Fig. 3. Relationship between the 5-minute Apgar score and
pH.
fetal status than other acid-base data,S especially since lactic acid is the major specific end product of anaerobic metabolism and its accumulation at high concentrations in brain tissue causes edema and necrosis. IS In the present study, we have used the 1- and 5-minute Apgar scores for the evaluation of fetal outcome. Although the relationship between the Apgar scores and subsequent neonatal and neurological development is far from perfect, Apgar scores remain widely used for clinical assessment of the newborn infant. 16 We examined the relationship between fetal lactacidemia at the time of delivery and fetal outcome and compared this relationship to those existing between other acid-base parameters and fetal outcome. We found that the umbilical arterial lactate level correlates as well as the pH with Apgar scores, while both pH and lactate have better correlations with Apgar scores than the other acid-base parameters. We also found that combining the information from pH and lactate or from all five parameters (pH, lactate, Po 2 , Pco 2 , and base deficit) does not improve the ability to predict Apgar scores. The pathologic range of lactate was found to be 2::3.70 mmol/L, which is in general agreement with values of 3.30 mmol/L5 and 3.90 mmol/U reported in the literature. Lactate values 2::3.70 mmol/ L have about the same correlation as pH values <7.20 with low Apgar scores. Also lactate values <3.70 mmoll L have almost the same correlation as pH values 2::7.20 with good Apgar scores. The base deficit, which is an indirect reflection of metabolic acidosis, did not correlate as well as pH or lactate with fetal outcome. Nevertheless, it had significant correlations with both the I-minute Apgar score (r = -0.17, P < 0.001) and the 5-minute Apgar score (r = -0.15, P < 0.002). The relatively limited correlation between Apgar scores and both pH and lactate may be explained by
Outcome of fetuses with lactic acidemia
Volume 150 Number I
5.minute Apgar score
..•
!
.!" < 3.70mM/L >-
..•...
<7
~7
8
370
D
.-"
:i E
~3.70mM/L
that lactate is the major metabolic component in fetal acidosis. However, the combination of pH and lactate and of multiple fetal biochemical parameters adds nothing to our capability to predict neonatal depression available by measurement of either pH or lactate alone. The Po 2, PC02, and base deficit are poor predictors of fetal outcome as described by the Apgar score.
REFERENCES
D
-
37
8
82
:::»
Fig. 4. Relationship between the 5-minute Apgar score and lactate. many factors. While pH and lactate correlate better than the other acid-base parameters with Apgar scores, there are numerous metabolic and pathophysiologic changes that they do not reflect. Thus alterations in the fetal neurological status, drugs administered to the mother, anesthesia, and the effect of delivery are not necessarily reflected in pH and lactate measurements. Furthermore, the Apgar score is a gross general physical examination which reflects in a general way neonatal condition. Factors such as resuscitative procedures, prematurity, endotracheal intubation for suction of meconium, and other factors such as stimulation and temperature may affect the Apgar score. Thus it is not surprising that both pH and lactate correlate only modestly with Apgar scores. Finally, fetal acidosis may be secondary to sustained maternal metabolic acidosis and not indicative of fetal compromise. 12 If the baby is delivered during this time, the Apgar scores may be normal in the presence of some degree of fetal acidosis or preacidosis. However, if such acidosis is persistent, it may have a deleterious effect on fetal metabolism, so that the timing of delivery is relevant to the relationship between this secondary fetal acidosis and Apgar scores. Assuming that the capillary scalp blood values would not be more strongly correlated than the umbilical values with the Apgar scores, we conclude that lactate measurement in fetal blood is as good a means for monitoring the fetus in labor as the pH. Its close relationship to pH measurements probably reflects the fact
1. Hendricks CH. Studies on lactic acid metabolism in pregnancy and labor. AM j OBSTET GYNECOL 1957;73:492. 2. Otey E, Stenger V, Eitzman D, Andersen T, Gessner I, Prystowsky H. Movements of lactate and pyruvate in the pregnant uterus of the human. AM j OBSTET GYNECOL 1964;90:747. 3. Low jA, Pancham SR, Worthington D, Boston RW. The acid-base and biochemical characteristics of intrapartum fetal asphyxia. AM j OBSTET GYNECOL 1975;121:446. 4. Daniel SS, Adamsons K, james LS. Lactate and pyruvate as an index of prenatal oxygen deprivation. Pediatrics 1966;37:942. 5. Hsiung R, Piquard F, Schaefer A. Machu j, Haberey P, Dellenbach P. Lactaplasmie maternelle et foe tale au cours de I'accouchement. II. Accouchements pathologiques. Rev Fr Gynecol Obstet 1981;76:853. 6. Lin C-C, Moawad AH, Rosenow Pj, River P. Acid-base characteristics of fetuses with intrauterine growth retardation during labor and delivery. AM j OBSTET GYNECOL 1980;137:553. 7. Kragenings J. Methodology oflactate assay. In: Lactate in acute conditions. International Symposium. Basel: S. Karger AG. 1979:20. 8. Racine P, Engelhardt R. Higelin jC, Mindt W. An instrument for the rapid determination of L-lactate in biological fluids. Med Instrum 1975;9: 11. 9. Soutter WP, Sharp F, Clark DM. Bedside estimation of whole blood lactate. Br j Anaesth 1978:50:445. 10. Bretscher j, Saling E. pH values in the human fetus during labor. AM j OBSTET GYNECOL 1967;97:906. 11. Hon EH, Khazin AF, Paul RH. Biochemical studies of the fetus. II. Fetal pH and Apgar scores. Obstet Gynecol 1969;33:237. 12. Bowe ET, Beard RW, Finster M, Poppers Pj, Adamsons K, james LS. Reliability of fetal blood sampling. Maternal-fetal relationships. AM j OBSTET GYNECOL 1970; 107:279. 13. Miller FC. Prediction of acid-base values from intrapartum fetal heart rate data and their correlation with scalp and funic values. Clin Perinatol 1982;9:353. 14. Banta HD. Thacker SB. Assessing the costs and benefits of electronic fetal monitoring. Obstet Gynecol Surv 1979; 34:627. 15. Myers RE. Wagner KR, Courten GD. Lactic acid accumulation in tissue as cause of brain injury and death in cardiogenic shock from asphyxia. In: Lauersen NH, Hochberg HM, eds. Clinical perinatal biochemical monitoring. Baltimore: Williams & Wilkins, 1981: 11. 16. Niswander KR. Asphyxia in the fetus and cerebral palsy. In: Pitkin RM, Zlatnik Fj, eds. The year book of obstetrics and gynecology. Chicago: Year Book Medical Publishers, Inc., 1983: 107.