Pregnant and Postpartum Admissions to the Intensive Care Unit:
A Ten-Year Review from Southwest Nigeria.
Ayodeji Oladiran,1 Owolabi Ayodele,2 Adeyemo Matthew,1
Obafemi Temiope,1 Gbala Michael.1
1Department of Obstetrics and Gynaecology, University of Medical Sciences Teaching Hospital, Ondo
2Department of Anaesthesia and Intensive Care University of Medical Sciences Teaching Hospital, Ondo.
ABSTRACT
Correspondence:
Oladiran Ayodeji.
Department of Obstetrics and
Gynaecology,
University of Medical Sciences
Teaching Hospital, Ondo.
Email: ladiran54@gmail.com;
ORCID ID: 0000-000-006-5812.
Background: Obstetric admissions into the Intensive Care Unit (ICU) reflect the
scope of severe maternal morbidities and maternal near-miss events prevalent in
the local obstetric population. Previous studies found no difference in the profile
of ICU admissions between developing and developed countries. Therefore,
differences in ICU management outcomes may be associated more with the
contextual nature of obstetric delays. Understanding the characteristics of patients
needing ICU care may facilitate the early identification of obstetric risk, prompt
risk assessment and timely transfer to the ICU. We studied the profiles and
outcomes of such patients over 10 years. Objectives: To analyze the
characteristics, profiles, and clinical outcomes of obstetric admissions into the ICU
at a tertiary hospital setting. Methods: A retrospective review of 49 obstetric
admissions into the University of Medical Sciences ICU in Ondo, southwest
Nigeria was conducted. Relevant data was extracted into pro forma forms designed
for this purpose. The ICU admission outcome was compared with the
characteristics of these patients using binomial regression statistics. Results: The
leading diagnoses associated with obstetric ICU admissions over the study period
were hypertensive diseases of pregnancy (34.7%), sepsis (20.4%),
anaesthetic/cardio-pulmonary complications (18.2%) and postpartum
haemorrhage (12.2%). Over 70% of admissions had Glasgow coma scale scores
of 8 or less. Clinical outcome was significantly associated with a Glasgow Coma
Scale score of ≤ 8 on admission. Maternal mortality was 61%, the majority of
which occurred within 24 hours of admission. Conclusions: The poor ICU
admission outcome reflects the severity of the patient’s clinical status at
presentation. The delay in seeking expert care might have worsened the already
high severe maternal morbidity incidence in the community, contributing to the
high mortality observed in this study. Efforts to mitigate late presentation at the
tertiary centres require strong healthcare reforms. Efforts to establish high-
dependency units should be explored to reduce the burden of obstetric admissions
in the ICU.
Keywords: Obstetric Admissions, Intensive Care Unit, Severe maternal
morbidities, High Dependency units, Healthcare Outcomes.
Oladiran et al. Pregnant and Postpartum Admissions to the Intensive Care Unit
Tropical Journal of Obstetrics and Gynaecology (TJOG) Vol. 43 No. 4 (2025)/Published by Journalgurus
300
INTRODUCTION
Pregnant and postpartum admissions into the Intensive
Care Unit (ICU) represent an escalation of treatment
options available to the Obstetrician. Such admission is
a tacit acknowledgement of the deteriorating clinical
and metabolic state of the obstetric patient, and a
collective effort to mitigate possible maternal mortality
through intensive monitoring and timely interventions.
Obstetric admission into the Intensive Care Unit (ICU)
is therefore indicative of severe morbidity.1 The unique
nature of obstetric medicine, with its altered physiology
of pregnancy and the simultaneous management of two
lives with different physiologies, presents a challenge.2
The potential for catastrophic complications is constant
and may develop rapidly.3
The ICU often provides for closer monitoring of
the obstetric patient and the prompt interventions often
required to avert poor obstetric outcome.2 It should be
noted that all medical conditions that can complicate
pregnancy and the puerperium are treated in the ICU.
There is a need for periodic audits of obstetric data
resulting from these critical interventions to evaluate
established management protocols and provide
guidance for crafting and designing fresh management
templates and policies. Obstetric admissions into the
ICU would therefore reflect the scope of severe
maternal morbidities and the maternal near-miss events
prevalent in the local obstetric population.
With a maternal near-miss incidence ratio of 70
per 1000 live births and severe maternal outcome of 85
per 1000 live births, our facility is a ready source of
eligible clientele for possible ICU admissions.4 Despite
the availability and perceived advantages of ICU care,
financial and other socio-cultural constraints may
hinder prompt ICU admissions. It had been previously
established that there is no difference in the profile of
ICU admissions in developing compared to developed
countries, with hypertensive disorders of pregnancy
being the most prevalent indication for ICU admission.3
These indications for ICU admission are often the same
causes of maternal mortality in the population.4 Nigeria
with a maternal mortality rate of 576 per 100,000 live
births, is responsible for 10% of global deaths for
pregnant women. 5
While the causes of maternal mortality are
generally known and are similar worldwide, delays in
accessing specialist care are generally responsible for
differences in mortality rates between high-resource
and low-resource countries. The recognition that
clinical and community services are inextricably
linked, and the contextual nature of obstetric delays,
should inform the study of the factors responsible for
such delays in the local environment. Even where ICU
facility is available, admissions to these institutions are
often hampered by the same prevalent delays in the
local environment. Such delays tend to compound what
may already be a dire clinical situation. There exists a
knowledge gap in our region of the delays that hinder
obstetric admissions to our intensive care units, and
their impact on clinical outcome. Understanding the
characteristics and profiles of patients needing such
care may be the necessary first step for early
identification, quick risk assessment and timely transfer
to the ICU. Our aim is to analyse the patients'
characteristics, profiles and outcomes of pregnant and
postpartum admissions to ICU in a tertiary facility in
southwest Nigeria, with a focus on identifying factors
associated with poor outcome.
MATERIALS AND METHODS
This study is a retrospective review of all pregnant and
postpartum admissions into the Intensive Care Unit
(ICU) at the University of Medical Sciences Teaching
Hospital in Ondo, South-West Nigeria. The study
period spans from January 1, 2014, to December 31,
2023. The UNIMEDTH started in late 2012 as the
Mother and Child Hospital, offering free obstetric care.
Fee-paying commenced in late 2016 till 2019, with
criteria-based subsidized insurance services operating
from 2020 till date. The Intensive Care Unit is a 5-bed
closed unit that started providing services from
inception, sourcing its clientele mainly from the
Surgical and Trauma Units within which it is located.
The ICU has always been a fee-for-service unit.
Pregnant and postpartum admissions into the ICU are
routed through the Emergency unit of the Obstetrics
and Gynaecology Department for referred cases. Direct
admissions of booked patients do occur from our wards
and postoperatively from our obstetric theatre.
Data Collection
Data were collected from the patient’s records, ICU
admission and discharge register, and the hospital's
electronic health records. The list of all women
admitted into the ICU was first compiled from the
source population of all admissions in the ICU
admission register. All pregnant and postpartum
women were identified and subsequently extracted
from this list. The case files of these women were
subsequently retrieved from our Health Information
unit for study and data extraction. The following
information was extracted into a structured pro forma
for each patient: age, parity, gestational age at
admission, year of admission, reason for admission,
length of stay in the ICU, interventions performed in
the ICU, and patient outcomes. This initial data
collection was done by one of the authors, an intensivist
working in the ICU. Subsequently, obstetric case file
reviews were performed by the leading author.
Ethical clearance for the study was obtained
from the University of Medical Sciences Review
Oladiran et al. Pregnant and Postpartum Admissions to the Intensive Care Unit
Tropical Journal of Obstetrics and Gynaecology (TJOG) Vol. 43 No. 4 (2025)/Published by Journalgurus
301
Board. The University Teaching Hospital Management
permitted the use of anonymised data because of the
study's retrospective nature. Informed consent from
patients was not necessary because only anonymised
data from the medical records was used.
Inclusion Criteria
The study population included all pregnant women and
those within six weeks postpartum admitted into the
ICU during the study period. Of 64 possible obstetric
admissions identified from the ICU admission register,
only 49 women had casefiles available for further
study.
Exclusion Criteria
Non-obstetric patients and all those obstetric patients
with unavailable records were excluded from the study.
Data Analysis
The data collected were verified from multiple sources,
primarily from the patient’s case-note, then from
registers in the Intensive Care unit and the Labour
ward. Missing data was identified and recorded on the
appropriate tables. They were subsequently excluded
from specific analysis.
Data were inputted into and analysed using
SPSS version 20. Descriptive statistics were computed
for all variables. The commonest causes of admission
were determined as well as the outcomes of these
admissions. A p-value of less than 0.05 was considered
statistically significant.
RESULTS
Baseline Characteristics
Over the ten years 2014 to 2023, the case files of 49
patients admitted into the ICU were available for
review. There were a total number of 18054 deliveries
during this period. (Table 2) The average rate of
obstetric admission into the ICU was 2.7 per 1000
deliveries. The highest number of admissions was 12 in
2019 while there was no recorded admission for 2017.
(Table 2; Figure 1) The mean age of study participants
was 32 years (±7.24). A majority (65.3%) of the
participants were multiparous. (Table 1). Based on
available data, 55.1% had caesarean delivery in the
indexed pregnancy, while 20.4% delivered vaginally.
Postpartum admission into the ICU was indicated in
95.9% of participants while 4.1% had antepartum
admission. An equal number (17/34 or 50%) of patients
were admitted from our facility as were referred from
outside for ICU care. The ICU admission duration was
up to 24 hours or less in 24 (48.9%) of 49 participants.
Only 6 participants were on admission for more than
one-week duration. (Table 1)
Indications for Admission
The various primary indications for ICU admission are
shown in Table 3. Obstetric indications were
responsible for the majority (51%) of admissions. The
highest
Table 1. Baseline characteristics of patients admitted
into the ICU
Characteristic
feature
frequency
frequ
ency
percen
t
Age in years
< 20
1
2.0
20 – 24
8
16.3
25 -39
35
71.4
≥ 40
4
8.2
Missing
1
2.0
Parity
0
1
2.0
1- 4
32
65.3
≥ 5
2
4.1
Missing
14
28.6
Mode of delivery
SVD
10
20.4
C/S
27
55.1
Missing
12
24.5
Time of admission
Ante-
partum
2
4.1
Post-
partum
47
95.9
Initial place of
treatment prior to
ICU admission
Outside
UNIMEDTH
17
34.7
UNIMEDTH
17
34.7
Missing
15
30.6
Duration of ICU
admission
0 – 1 day
24
49.0
2- 3 days
8
16.3
4 – 7 days
6
12.2
> 7 days
6
12.2
Missing
5
10.2
Table 2. Record of year of admission to the ICU a
Year
No of
admission
Annual
delivery
Rate per 1000
delivery
2014
6
2894
2
2015
3
3340
0.8
2016
4
2210
1.8
2017
0
1294
0
2018
6
1146
5
2019
12
1943
6
2020
5
1594
3
2021
1
828
1
2022
4
1329
3
2023
7
1476
4
Total
48
18054
Oladiran et al. Pregnant and Postpartum Admissions to the Intensive Care Unit
Tropical Journal of Obstetrics and Gynaecology (TJOG) Vol. 43 No. 4 (2025)/Published by Journalgurus
302
a 1 patient did not have year of admission reported
obstetric indication for admission was due to
Hypertensive diseases of pregnancy, contributing to 17
cases or 34.7%, while obstetric haemorrhage was the
reason in 6 patients (12.2%). Non-obstetric indications
were responsible for ICU admission in 49% of patients.
Sepsis contributed the highest reason for admission in
10 patients or 20.4% of cases. This is followed by
anaesthetic complications (9 cases or 18.2%).
In this study, Hypertensive diseases, Sepsis,
anaesthetic complications and haemorrhage
contributed 85.7% of indications for ICU admissions.
Additionally, estimates of effect size may be unstable
and vulnerable to sampling variability.
In view of the above, further studies with larger
cohorts may be necessary to further interrogate our
findings and bolster the robustness of the conclusions
derived.
Table 3. Diagnosis of patients at the time of admission
to the ICU
Reasons for
Admission
Freq
%
Obstetric
Hypertensive
diseases of
pregnancy
17
34.7
Haemorrhage
6
12.2
IUFD
2
4.1
Non-Obstetric
Sepsis
10
20.4
Anaesthetic
complication
9
18.4
Cardiomyopat
hy
1
2.0
Haemoglobino
pathy
1
2.0
Pulmonary
Oedema
(? Amniotic
fluid
embolism)
1
2.0
Respiratory
1
2.0
Missing
1
2.0
Total
49
100
Table 4. GCS scores of patients admitted to the ICU
(≤8 = 25)
Variable
Frequency
Percent
GCS score
3
15
30.6
4
4
8.2
6
1
2.0
7
2
4.1
8
3
6.1
10
4
8.2
14
1
2.0
15
5
10.2
Missing
14
28.6
Total
49
100.0
Initial Clinical Assessment in ICU
The assessment of clinical severity using the Glasgow
Coma Scale score, was recorded for 35 of 49 ICU
admissions. Of these, 25 individuals (71.4%) had a
GCS score of 8 or less. The remaining 10 individuals
were assigned scores equal to 10 or more. (Table 4).
The mean GCS score of survivors was 9.91±4.5571
while the mean GCS score of those who died was
5.21±3.647. (Figure 3). The difference was statistically
significant. (Mann-Whitney U test p= 0.006)
Table 5. Complications in patients admitted to the ICU*
Complications
Subtypes of
complications
Freq.
%
Pulmonary
Pulmonary oedema,
ARDS, Respiratory
insufficiency
21
42.9
Cardiovascular
Cardiac arrest,
Cardiopulmonary arrest,
Acute Left Ventricular
Failure, Hypovolaemic
shock, Haemorrhagic
shock.
16
32.7
Renal
AKI
14
28.6
Others
Hypoxic brain injury,
HELLP syndrome,
Sepsis, Intra-abdominal
sepsis, Septic shock,
Enterocutaneous fistula,
DIC, Puerperal sepsis.
8
16.3
* Multiple responses included. AKI- Acute kidney injury,
ARDS- acute respiratory distress syndrome,
HELLP- Haemolysis elevated liver enzymes low platelets.
Table 6. Interventions administered in the ICU for
admitted patients
Interventions
Frequency
Valid
Percent
Mechanical ventilation
33
67.3
Inotropes
22
44.9
Radiological Interventions
10
20.4
Central Catheter
5
10.2
Haemodialysis
4
8.2
ECG/ECHO
1
2.0
Anticoagulants
1
2.0
Oladiran et al. Pregnant and Postpartum Admissions to the Intensive Care Unit
Tropical Journal of Obstetrics and Gynaecology (TJOG) Vol. 43 No. 4 (2025)/Published by Journalgurus
303
Figure 2. Clinical Outcome of Patients admitted to the ICU
Table 7. Admission diagnosis vs clinical outcome
Complications Encountered and Interventions
Given
Often, complications had occurred by the time of ICU
admissions. The different complications encountered in
all 49 individuals admitted to the ICU are detailed in
Table 5. These complications are grouped collectively
into pulmonary, cardiac, renal, and other conditions.
Pulmonary complications were diagnosed in 21
individuals, 16 individuals had cardiac challenges and
14 had complications in the kidneys. (Table 5) Multiple
organ disease was present in 5 patients. Individuals
Table 8. Results of analytical models
Figure 3: Mean Glasgow Coma Scale scores vs patients’ clinical
outcome (Died/Survived). The mean GCS score for the 30 mortality
cases was relatively lower at 5.21±3.647 compared with that for the
19 patients who survived which is 9.91 (±4.571).
with a diagnosis of eclampsia and sepsis suffered
multiple complications that included acute kidney
injuries, HELLP syndrome, and multiple organ
diseases. Therefore, multiple complications may be
present in the same individual. The pulmonary
complications were mainly hypoxic events that
necessitated intubation and mechanical ventilation.
Thirty-three of 49 (67.3%) patients required
mechanical ventilation. (Table 6) Cardiac cases
mandated the use of inotropic drugs while some
anaesthetic complications were due to inadvertent
“high spinal” administered preparatory to surgical
deliveries. The renal complications mostly emanated
from women with initial diagnosis of hypertensive
disorders of pregnancy. These acute kidney injuries
(AKI) were sometimes treated with dialysis. Patients
with multiple organ disease, hypoxic brain injuries,
pulmonary oedema, and disseminated intravascular
coagulopathies, were represented and grouped as
Oladiran et al. Pregnant and Postpartum Admissions to the Intensive Care Unit
Tropical Journal of Obstetrics and Gynaecology (TJOG) Vol. 43 No. 4 (2025)/Published by Journalgurus
304
“Others”. The interventions necessitated by these
complications are tabulated in Table 6. Noteworthy is
that 33 individuals (67.3%) required mechanical
ventilation.
Clinical Outcome of Patients Admitted to ICU
A total of 30 (61.2%) patients died following ICU
admission while 19 (38.8%) survived. (Figure 2, Table
7). Seven of 18 (38.8%) individuals with a primary
admission diagnosis of hypertension survived. Three of
6 (50%) individuals with a primary admission
diagnosis of haemorrhage survived. Only two of 10
(20%) individuals with a primary admission diagnosis
of sepsis also survived. Of the total 30 deaths in this
study, 19 deaths occurred within 24 hours of ICU
admission. Overall, 11 of 17 individuals with an
admission diagnosis of hypertension died, while 8 of 10
individuals with sepsis died. The admission diagnosis
of sepsis had the highest case fatality rate of 80%. The
top four reasons that accounted for 85.7% of ICU
admissions were responsible for 26 (86.6%) of the total
30 deaths.
Statistical Analysis
The binomial regression model comparing the clinical
outcome with the Glasgow coma assessment score
reached statistical significance with a p-level of 0.004.
(Figure 3) The same binomial model comparing clinical
outcomes with the duration of ICU admission did not
reach statistical significance. (p-value of 0.079)
Whether or not a patient had assisted ventilation
(intubation plus mechanical ventilation) did not
significantly influence clinical outcome (p-value of
0.665).
The distribution of the GCS score at the time of
admission was assessed for normality using the
Shapiro-Wilk test (W = 0.775, p < 0.001), with the
result confirming a non-normal distribution. A Mann-
Whitney U test was thus considered to compare the
admitting GCS between the group of patients that
survived and those who died. There was a significant
difference in the mean GCS score between the two
groups; z = -2.730, p = 0.006.
Power Analysis
Despite a moderate rank-biserial correlation (≈ 0.56),
the estimated power was only 8.4%, which is well
below the conventional threshold of 80%. This suggests
a high probability of predisposition to Type II error.
Limitations of the Analysis/Study
The small sample size (n = 35 for “GCS”) severely
limits the statistical power to detect significant
associations between cohorts of survivors and
mortalities. Low power increases the risk of false-
negative findings (Type II error).
DISCUSSION
This study represents the first audit of the hospital data
on pregnant and postpartum admissions to the ICU unit
of our facility. Obstetric admissions into the ICU were
64 (10.3%) of a total of 620 admissions (both obstetric
and non-obstetric). The mortality rate of 61% among
obstetric ICU admissions in this study is higher than the
figures of Onyekwulu et al. [] (39.7%) from Nigeria and
Prin and colleagues (49%) from Malawi. 7
Our figure reflects the severity of morbidities of
the patients in this study at presentation, a pointer to
significant delay in accessing expert obstetric care. Our
study showed that 71.4% of patients in this study had
Glasgow Coma Scale scores of 8 or less on ICU
admission, further confirming the obstetric delay
prevalent in our area. Their poor clinical outcome was
significantly associated with the Glasgow Coma Scale
score of 8 or lower. The mean Glasgow Coma Score of
those who survived was significantly higher compared
with those who died. We agree with the findings of
Adeniran et al that the clinical state of the critically ill
obstetric patients at admission remains a major
determinant of the outcome of ICU care.
The development of severe complications in
these individuals within 24 hours of admission and the
need for mechanical ventilation in 67.3% of these
patients were pointers to the progression and severity
of the disease process. The need for mechanical
ventilation has been shown to portend increased
mortality risks following ICU admission.8,9 Adeniran
and co-workers found that mechanical ventilation is a
statistically significant predictor of obstetric ICU
deaths.8 The study by Anane-Fenin and colleagues
from Ghana found those patients intubated to be at least
six times more at risk of death when compared with
those who were not.9 Others found that as many as 90%
of ICU patients in developing countries would require
mechanical ventilation.2 Our study did not find a
significant association between the need for mechanical
ventilation and clinical outcome. This was probably
due to the moribund clinical state of our patients at
presentation, blunting the treatment impact of
mechanical ventilation.
The plethora of complications ranging from
HELLP (haemolysis, elevated liver enzymes, low
platelet count) AKI, acute respiratory distress, hypoxia
and others, very early in the admission, may be
responsible for this finding. It is important therefore to
consider this and other factors when comparing the
clinical outcomes of different studies. Lower ICU
maternal mortality rate may reflect a lower threshold
for ICU admission. 12 Obstetric admissions from the
study by Zhao and colleagues in China, had a mean
Oladiran et al. Pregnant and Postpartum Admissions to the Intensive Care Unit
Tropical Journal of Obstetrics and Gynaecology (TJOG) Vol. 43 No. 4 (2025)/Published by Journalgurus
305
Sequential Organ Failure Assessment score of 2, a
pointer to the low admission threshold in their cohort.
Cases of inadvertent ‘high spinal’ early in the life of our
institution, added to our poor clinical outcome.
Hypertension in pregnancy, postpartum
haemorrhage, sepsis and cardio-pulmonary
complications were the indications for ICU admission
in the majority of the patients in this study. This finding
is consistent with other studies from low-resource
nations. 6,8-11 Despite this similarity with other studies,
63.3% (19 out of 30) died within 24 hours of ICU
admissions, due mostly to diagnoses of hypertensive
diseases of pregnancy, postpartum haemorrhage and
sepsis. More than half (53.9%) of ICU obstetric deaths
occurred within 24 hours of admission in the series
from Anane-Fenin and her colleagues from Ghana. The
mortality in the study by Onyekwulu et al however
occurred within 3 days. The delay of access to expert
obstetric care appears to be a common factor operating
in low-resource countries. It is possible that the poorer
outcome in our study stemmed from a longer delay of
access to expert care.
The duration of ICU admission has no
significant association with clinical outcomes in this
study. When deaths from cardio-pulmonary causes are
excluded, 22 of 30 deaths in this study were due to
hypertensive disorders of pregnancy, sepsis, and
postpartum haemorrhage. These diseases are eminently
predictive and preventable. The strategies for
predicting and assessing these obstetric indications for
ICU admission should be our critical focus if we desire
to significantly reduce ICU admission mortality. Such
efforts would involve the early recognition of these
diseases, prompt morbidity risk assessment, and the
timely transfer to tertiary centres to receive expert care.
Therefore, heightened clinical vigilance by healthcare
providers, especially at lower-level health institutions,
is essential.
The delay of access to expert obstetric care had
been previously recognised by other investigators.12 It
remains the major constraint to the reduction of these
maternal mortalities. Efforts to mitigate late
presentation at tertiary centres would require strong
health sector reforms. These should include upgrading
the quality of care at the lower-level health centres,
through continuous training and re-training of the
healthcare personnel and the provision of relevant
equipment necessary for needed risk assessment.
Continuous and periodic quality of care assessment
should be carried out especially at the lower-level
health centre in accordance with WHO vision and
standards. 13 Tertiary care centres should maintain
continuous interaction with community care personnel
in their locality through periodic participatory lectures,
workshops and training in basic obstetrics, particularly
in areas related to the recognition and assessment of
maternal morbidities and other high-risk obstetrics
Our study found that 50% of obstetric ICU
admissions were referrals from outside facilities. We
found no evidence that patients referred from outside
had a worse clinical outcome than those admitted
directly from our facility.
Maternal ICU admission is recognised as an
indicator of severe maternal morbidity. 14-16 Despite our
high maternal near-miss incidence,4 the critical care
utilisation of 2.7 per 1000 deliveries in this study, is
low. Our figure is lower than the 5.87% reported by
Onyekwulu and co-workers from Enugu, southeast
Nigeria.6 Anane-Fenin and others reported 14.7 ICU
obstetric admissions per 1000 deliveries. Individual
hospital and administrative protocols are probably
responsible for these differences. Among other factors
contributing to delay in accessing ICU admissions,
financial considerations, probably made significant
impact, especially in relation to the relatively low
patronage of our facility. Our ICU demands the
payment of a mandatory premium financial deposit
before admission. This appears contradictory in a
setting where emergency obstetric services were free to
patients for several years during the ten-year study
period.
The charging of high fees contributed to an
avoidable delay that resulted in lower ICU patronage
by those in dire need of admission. The ease of access
to ICU admission is also affected by the paucity of
these facilities in our geographical area, noting that no
ICU facility is dedicated to the exclusive use by
obstetric patients. Our 5-bed ICU was designed
primarily to serve the needs of the surgical and trauma
clientele of our hospital and remains the only
government facility in the south senatorial zones of
Ondo state. The consequent high ICU burden from
obstetric admissions needs to be relieved. Our high
maternal mortality rate from this study supports the
need for urgent service delivery improvements. Hence
the need to rethink our ICU admission policies.
While ICU admission is primarily geared
towards addressing organ failures, the majority of
obstetric indications for ICU admission require
continuous surveillance of their clinical parameters.
Also, ICU admissions mean a rising need for more
medical resources and expenses. There should
therefore be a move away from obstetric ICU
admission to the continuous reduction of our high
maternal morbidity burden. This is particularly relevant
to our low-resource setting. The evidence from the
work of Beza and co-workers from Ethiopia showed
that the expansion of High Dependency Units (HDUs)
is key for mitigating the ICU burden from obstetric
admission. 11
Anane-Fenin and colleagues from Ghana
supported the same standards. HDU will fulfil this role
when severe maternal morbidities are referred early
before complications set in. The patients in our study
had already developed serious complications and organ
Oladiran et al. Pregnant and Postpartum Admissions to the Intensive Care Unit
Tropical Journal of Obstetrics and Gynaecology (TJOG) Vol. 43 No. 4 (2025)/Published by Journalgurus
306
damage within the first 24 hours of ICU admission.
This is evident from the number of complications
encountered by our patients, even within 24 hours of
admission. The study by Gu et al from China, affirmed
that the introduction of HDU was followed by a
reduction in ICU utilization in their obstetric
population. They found that the establishment of HDU
led to a 20% decrease in ICU admission. 17 In the face
of chronic financial and personnel shortages, the HDU
option should be explored. Our results and experiences
with HDU would then dictate policy reforms. While
HDU provides a level of care which lies in between a
general obstetric ward and an ICU, it is uncertain
whether they could improve patient outcomes. This is
an area for further investigation and future research.
Limitations of this Study
The retrospective nature of this study suffered from
limitations of data quality evidenced by missing data on
some important items like Glasgow Coma Scale scores.
The possibility of inherent selection bias cannot be
completely ruled out. Also, the generalisation of the
findings in this study is limited as they are derived from
a single institution with relatively small number of
eligible participants. The small sample size for GCS
scores limits the statistical power to detect significant
associations between cohorts of survivors and those
who died. Further studies with larger cohorts may be
necessary to bolster the robustness of the conclusions
so derived. A prospective multicentre study might have
provided a better and more reliable data set.
CONCLUSION
We found a high maternal mortality of 61% amongst
our cohort of ICU admissions. The majority of these
ICU admissions had significantly poor GCS scores on
admission. The high maternal mortality in this series is
associated with significantly lower Glasgow Coma
Scores. Those who survived had significantly higher
GCS scores than those that died. Our findings point to
significant delays in accessing expert obstetric care in
the context of an already high prevalence of severe
maternal morbidity. Training and re-training of
healthcare workers, particularly at primary and
secondary health facilities, would improve their
recognition of obstetric morbidities, their assessment
and the timely referral for expert obstetric care. Our
findings demand a policy change in healthcare
administration to address prevalent obstetric delays.
Based on available evidence from other health centres
in low-resource settings, we believe the establishment
of High Dependency units is an option worth exploring.
Recommendation
a. Identify the scope of severe maternal morbidities
and near-miss events in the community.
b. Reduce the burden of severe maternal morbidities
by mitigating the factors responsible for obstetric
delays in the population under consideration. Strong
health sector reforms are necessary for this to
happen.
c. Explore the option of establishing High
Dependency Units as a first step towards limiting
the burden of obstetric ICU admissions. This may
prove to be a more cost-effective solution in a low-
resource setting.
Conflict of Interest: There is no conflict of interest.
REFERENCES
1. Baskett T, Sternadel J. Maternal intensive care and near-
miss mortality in Obstetrics. BJOG 1998; 105: 981- 982.
2. Heinonen S, Tyrvainen E, Saarikoski S, Ruokonen E.
Need for maternal critical care in Obstetrics: a
population-based analysis. Int J Obstet Anesth 2002; 11:
260-264.
3. Tang LC, Kwok AC, Wong AY et al. Critical care in
obstetrical patients: an eight-year review. Chin Med J
(Engl). 1997; 110: 936- 941.
4. Ayodeji O F, Adeyemo M, Gbala M, Akinkunmi B.
Facility Based audit of Maternal Near Miss, Maternal and
Neonatal Mortality in a Tertiary Hospital, South-West
Nigeria: One-Year Review. Womens Health Sci J. 8(1);
22024: 000226.
5. www.unicef.org/nigeria/situation-women-and-children-
nigeria. Last assessed 22/03/2023.
6. Onyekwulu FA, Okeke TC. Trends of critical care
management of obstetric patients in a tertiary hospital in
sub-Saharan Africa. Int J Res Med Sci 2019;7: 1420-4.
7. Prin M, Kadyaudzu C, Aagaard K, Charles A. Obstetric
admissions and outcomes in an intensive care unit in
Malawi. International Journal of Obstetric Anaesthesia
2019; 39: 99-104.
https://doi.org/10.1016/j.ijoa.2019.03.004
8. Adeniran AS, Bolaji BO, Fawole AA, Oyedepo OO.
Predictors of maternal mortality among critically ill
obstetric patients. Malawi Med J 2015; 27(1):16-19.
9. Anane-Fenin B, Agbeno EK Osarfo J, Opoku Anning
DA, Boateng AS, Ken-Amoah S et al: (2021) A ten-year
review of indications and outcomes of obstetric
admissions to an intensive care unit in a low-resource
country. PLoS ONE 16;(12): e0261974.
https://doi.org/10.137/journal.pone.0261974
10. Abiodun J. Critical care of the obstetric patients in the
Intensive Care Unit of a Nigerian Tertiary Hospital- a 4-
year Review. Glob Scholast Res J Multidiscip 2015; 1.
11. Beza Z, Tadese R, Teshome H, Tadele G, Siferih M.
Admission indications, initial diagnoses, Interventions
and patient outcomes within the sole obstetric high-
dependency unit in Ethiopia. BMC Women’s Health
2024; 24: 329. https://doi.org/10.1186/s12905-024-
03175-z
Oladiran et al. Pregnant and Postpartum Admissions to the Intensive Care Unit
Tropical Journal of Obstetrics and Gynaecology (TJOG) Vol. 43 No. 4 (2025)/Published by Journalgurus
307
12. Zhao Z, Han S, Yao G, Li S, Li W, Zhao Y et al.
Pregnancy-Related ICU Admissions From 2008 to 2016
in China: A First Multicentre Report. Critical Care
Medicine 2018; 46(10):
doi:10.10.1097/CCM.0000000000003355e1002
13. O Tuncalp, WM Were, C MacLennan, OT Oladapo, AM
Gulmezoglu, R Bahl, B Daelmans et al: Quality of care
for pregnant women and newborns- the WHO vision.
BJOG. 2015 May 1;122 (8):1045-1049.
doi:10.1111/1471-0528.13451
14. Zeeman GG. Obstetric critical care: A blueprint for
improved outcomes. Crit Care Med 2006; 34: S208-S214
15. Zwat JJ, Dupuis JR, Richters A, et al. Obstetric intensive
care unit admission: A 2-year nationwide population-
based cohort study. Intensive Care Med 2010; 36: 256-
263.
16. Pollock W, Rose L, Dennis CL. Pregnant and postpartum
admissions to the intensive care unit: A systematic review
Intensive Care Med 2010; 36:1465- 1474.
17. Ning Gu, Yaning Zheng, and Yimin Dai. Severe
maternal morbidity: admission shift from intensive care
unit to obstetric high dependency unit. BMC Pregnancy
and Childbirth 2022; 22:140.
