Thank you for your interest in our work and for the opportunity to
clarify a few points. The focus of this review was to determine an
association between umbilical cord pH and perinatal and long term outcomes
and not prediction of these outcomes hence the use of odds ratios as the
summary statistic (1,2). Table 1 reports all the studies included in the
review and all the index tests that the authors of these individual
studies reported however different index tests were analysed separately
i.e. arterial or venous values. Sub-group analysis according to threshold
effect was performed for arterial pH (table 4) and this showed that the
most substantial association was at a pH threshold of 7.00 but that an
association did exist across all thresholds. Furthermore all analyses were
performed using the bivariate approach which takes into account any
threshold effect (3).
In answer to the concerns regarding the association with cerebral
palsy we accept the concerns regarding the definitions of cerebral palsy
used within this subgroup however as with any meta-analysis we can only
report the reference standards as used by the authors of the primary
studies. The concerns regarding relation of cerebral palsy to an
intrapartum hypoxic-ischaemia are discussed further in the paper in which
we state a stronger association with hypoxic ischaemic encephalopathy than
cerebral palsy.
Our main conclusion is that low arterial cord pH showed a strong,
consistent, and temporal association with these outcomes. We acknowledged
that we did not explore the use of cord pH as a prognostic test and that
further research within this area is required.
Reference List
(1) Deeks J. Systematic reviews of evaulations of diagnostic and
screening tests. BMJ 2001;323:157-62
(2) Grennland S. Interpretation and choice of effect measures in
epidemiologic analyses. Am J Epidemiol 1987;125:761-8
(3) Reitsma JB, Glas AS, Rutjes AW, Scholten RJ, Bossuyt PM,
Zwinderman AH. Bivariate analysis of sensitivity and specificity produces
informative summary measures in diagnostic reviews. J Clin Epidemiol
2005; 58(10):982-990.
Rapid Response:
Response to Dr Alouini
Thank you for your interest in our work and for the opportunity to
clarify a few points. The focus of this review was to determine an
association between umbilical cord pH and perinatal and long term outcomes
and not prediction of these outcomes hence the use of odds ratios as the
summary statistic (1,2). Table 1 reports all the studies included in the
review and all the index tests that the authors of these individual
studies reported however different index tests were analysed separately
i.e. arterial or venous values. Sub-group analysis according to threshold
effect was performed for arterial pH (table 4) and this showed that the
most substantial association was at a pH threshold of 7.00 but that an
association did exist across all thresholds. Furthermore all analyses were
performed using the bivariate approach which takes into account any
threshold effect (3).
In answer to the concerns regarding the association with cerebral
palsy we accept the concerns regarding the definitions of cerebral palsy
used within this subgroup however as with any meta-analysis we can only
report the reference standards as used by the authors of the primary
studies. The concerns regarding relation of cerebral palsy to an
intrapartum hypoxic-ischaemia are discussed further in the paper in which
we state a stronger association with hypoxic ischaemic encephalopathy than
cerebral palsy.
Our main conclusion is that low arterial cord pH showed a strong,
consistent, and temporal association with these outcomes. We acknowledged
that we did not explore the use of cord pH as a prognostic test and that
further research within this area is required.
Reference List
(1) Deeks J. Systematic reviews of evaulations of diagnostic and
screening tests. BMJ 2001;323:157-62
(2) Grennland S. Interpretation and choice of effect measures in
epidemiologic analyses. Am J Epidemiol 1987;125:761-8
(3) Reitsma JB, Glas AS, Rutjes AW, Scholten RJ, Bossuyt PM,
Zwinderman AH. Bivariate analysis of sensitivity and specificity produces
informative summary measures in diagnostic reviews. J Clin Epidemiol
2005; 58(10):982-990.
Competing interests:
None declared
Competing interests: No competing interests