Discussion
Main Findings
This review is, to our knowledge, the first to systematically identify
published studies attempting to provide risk scoring or prognostic
models for the prediction of PPH. Of eleven eligible studies, two have
presented externally validated risk tools but neither were developed
using recommended methods. Kim et al. (predicting blood transfusion
(≥8u) following CS for placenta praevia) and Rubio-Alvarez et al.
(predicting excessive postpartum blood loss in women with singleton
pregnancies who underwent vaginal delivery) did not define candidate
predictors and both demonstrated very low events per variable. Chi et
al., was the only study to have a tool applicable to the general
obstetric population but is of high risk of bias; until it has been
validated, it cannot be recommended for clinical use. The other eight
studies identified are not deemed suitable for use in clinical practice
due to a lack of clinical relevance of some study populations, high risk
of bias and lack of external validation. Six out of eleven studies
featured substantially less than 10 events per potential predictor being
tested such that sample size presents a major threat to reliability of
findings.
Strengths and Limitations
A strength of this review is the prospective publication of the protocol
in PROSPERO with strict adherence to this. The aim was to find a
clinically meaningful formula or tool which could be of use to a
clinician in daily practice. Numerous related studies have not published
a useable tool or logistic regression model with a formula which a
clinician could use in clinical practice – this may reflect poor (or
poorer than anticipated) performance of the model. This review benefits
from use of broad and general search criteria to maximise identification
of relevant studies. Additionally, the results yielded by the search
strategy were double-screened by two reviewers (CN and SN). The use of
the CHARMS checklist allowed for systematic data extraction and
assessment of risk of bias.
A limitation of this review is that three studies were unable to be
obtained which may have been appropriate for inclusion. One of these was
part of an unpublished PhD thesis and the other two were behind a
paywall.
This review highlights shortcomings regarding the risk of bias and
reporting of the included studies.
The review included only studies in English language such that this may
limit the value of the findings.
Interpretation
This review suggests that there are no published prediction tools for
PPH that are ready for clinical use. Future research to generate
prognostic models for use specifically in elective CS or in women aiming
for vaginal birth would facilitate advanced planning of personnel to
optimise care provided.
The clinical usefulness of models generated by some of the identified
studies is limited by the target population. Four studies focus on
vaginal births which is not clinically meaningful as this cannot be
guaranteed in advance. The circumstances during labour are subject to
change with a risk of CS present until the fetal head is delivered.
Therefore, despite one of these studies, Rubio-Alvarez et al., providing
an externally validated user-friendly risk prediction tool in Excel™,
its validity in practice is extremely limited as it is not possible to
know which women will give birth vaginally and thus for whom the model
is valid. Only one study produced a scoring tool aimed at use in the
general obstetric population but the study design was unclear and
attempts to contact the author were not successful. The study included
923 women in Beijing, China, of whom almost half had a PPH, and it did
not assess predictive performance via internal or external validation.
Therefore, despite the presentation of an equation to predict PPH with
AUC of 0.86, it’s lack of performance assessment means it cannot be
recommended for use in clinical practice.
Most studies were retrospective, meaning that some predictors may not
have been measured, but the vast majority of relevant risk factors for
PPH can be assessed retrospectively such that this is not considered to
be a major problem.
Some studies’ prediction models or tools are clinically unhelpful in
regard to the final predictors included due to some not being known at
time of birth. Both Biguzzi et al., and Rubio-Alvarez et al., included
neonatal birthweight as a predictor, which suggests that the intended
time for the nomogram and risk tool use is after weighing of the baby,
most likely once the highest risk of PPH has passed. These models are
therefore of limited value for preparation of resources prior to birth.
Estimated birthweight may be a more appropriate measure but this has not
been included as a predictor in any model.
Use of intrapartum factors can aid risk assessment in a dynamic
scenario. Two studies have included these: duration of the first and
second stage of delivery; non-use of uterotonics and cord traction.
Intrapartum risk scoring may be facilitated by use of electronic health
records, where the tool could be embedded within the system, but
otherwise may present logistical difficulties if it requires ongoing
computer access as per Rubio-Alvarez et al’s proposed risk tool.
Robust external validation was absent from all prediction models
identified, suggesting that this is poorly understood and undervalued.
Of the two models externally validated both utilised Hosmer-Lemesow
testing which is not recommended, and only one provided validation
results. Internal validation is a reasonable alternative as this
assesses how well the model performs in the underlying population from
which the model was developed, but only five studies did this and only
one is considered appropriate for prospective use (in placenta praevia
population) and thus this would benefit from future external validation.
The prediction models identified were at high risk of bias overall, with
lack of detail of candidate predictors, small sample size and suboptimal
statistical analysis being common, and missing data not reported in any
study. Without missing data information it is not possibly to fully
assess the related risk of bias.39
The need for adequately powered studies is clear. Half the included
studies have shown a low EPV (<10) with only one conducting
any shrinkage methods to overcome problems arising from overfitting of
the model (and risk of optimistic predictions) when there is a low
number of events. Despite this, several authors recommended use of
affected models without external validation.19,20,22As a result of heterogeneity and low EPV, it was not possible to conduct
a meta-analysis of the results. However, there is potential for
synthesis of findings for predicting PPH in a population of women with
placenta praevia, where individual participant data (IPD) meta-analysis
could be used.