MATERNAL MORBIDITY AND USING IT TO IMPROVE MATERNAL OUTCOME
A Dissertation Submitted to
THE TAMILNADU DR. M.G.R MEDICAL UNIVERSITY CHENNAI
In Partial fulfillments of the Regulations for the Award of the Degree of
M.S. (OBSTETRICS & GYNAECOLOGY) BRANCH – II
GOVERNMENT STANLEY MEDICAL COLLEGE CHENNAI
MAY 2018
CERTIFICATE BY THE INSTITUTION
This is to certify that dissertation entitled “IMPLEMENTATION OF MEOWS IN PREDICTING MATERNAL MORBIDITY AND USING IT TO IMPROVE MATERNAL OUTCOME” is a bonafide work done by Dr.SARINA VINCENT AROKIA A. at R.S.R.M Lying in Hospital, Stanley Medical College, Chennai. This dissertation is submitted to Tamilnadu Dr. M.G.R. Medical University in partial fulfillment of university rules and regulations for the award of M.S.
Degree in Obstetrics and Gynaecology.
Prof. Dr. PONNAMBALA NAMASIVAYAM, MD, DA, DNB.
Dean
Stanley Medical College &
Hospital, Chennai – 600 001
Dr. K. KALAIVANI,
M.D., D.G.O., DNB.
Prof & Head of Department,
Dept. of Obstetrics and Gynaecology Government RSRM Lying In Hospital, Stanley Medical College ,Chennai
CERTIFICATE BY THE GUIDE
This is to certify that this dissertation entitled
“IMPLEMENTATION OF MEOWS IN PREDICTING
MATERNAL MORBIDITY AND USING IT TO IMPROVE MATERNAL OUTCOME” submitted by Dr.SARINA VINCENT AROKIA A., appearing for Part II MS, Branch II Obstetrics and Gynecology Degree Examination in May 2018, is a Bonafide record of work done by her, under my direct guidance and supervision as per the rules and regulations of the Tamil Nadu Dr. MGR Medical university, Chennai, Tamil Nadu, India. I forward this dissertation to the Tamil Nadu Dr. MGR Medical University Chennai, India.
Dr. K. KALAIVANI, M.D., D.G.O.,DNB
Professor and Head of Department, Dept. of Obstetrics and Gynecology Government RSRM Lying In Hospital
Stanley Medical College, Chennai
DECLARATION
I Dr.Sarina Vincent Arokia A., solemnly declare that the dissertation titled, “IMPLEMENTATION OF MEOWS IN PREDICTING MATERNAL MORBIDITY AND USING IT TO IMPROVE MATERNAL OUTCOME” is a bonafide work done by me at R.S.R.M. Lying in Hospital. Stanley Medical College, Chennai – during October 2016–to September 2017 under the guidance and supervision of Prof.Dr .K. Kalaivani M.D., D.G.O., DNB., Professor and Head of the department, Obstetrics and Gynecology. The dissertation is submitted to the Tamilnadu Dr. M.G.R. Medical University, in partial fulfillment of University rules and regulations for the award of M.S. Degree in Obstetrics and Gynecology.
Place: Chennai
Date: Dr.Sarina Vincent Arokia A.
ACKNOWLEDGMENT
I am grateful to PROF.DR. PONNAMBALA NAMASIVAYAM, MD, DA, DNB., Dean, Govt. Stanley Medical College for granting me permission to undertake this study. I take this opportunity to express my sincere and humble gratitude to Dr.K. KALAIVANI, M.D., D.G.O., DNB., Superintendent, Govt. R.S.R.M. Lying in Hospital who not only gave me the opportunity and necessary facilities to carry out this work but also gave me encouragement and invaluable guidance to complete the task I had undertaken. I am deeply indebted to her, the mover behind this study for her able guidance and inspiration and constant support without which this would not have been possible.
I am very grateful to the Registrar Dr.H.ANITHA VIRGIN KUMARI, M.D., D.G.O. for her invaluable advice, constant guidance and supervision during this study.
I am extremely grateful to all our Assistant Professors, for their advice and support during this study.
I sincerely thank my fellow postgraduates and friends for their support and cooperation.
this study would not have been possible.
I wish to specially thank My Husband, who has consolidated my confidence to overcome obstacles and supported me in my every endeavour.
Finally I thank Lord Almighty, who gave me the will power and showered blessings to complete my dissertation work.
CONTENTS
S.NO TITLE PAGE
NO
1. INTRODUCTION 1
2. AIM OF THE STUDY 5
3. MATERIALS AND METHODS 6
4. REVIEW OF LITERATURE 9
5. RESULTS 36
6. DISCUSSION 65
7. CONCLUSION 75
8. BIBLIOGRAPHY 9. ANNEXURES
PROFORMA
MASTER CHART
ABBREVIATIONS
CONSENT FORM
ETHICAL COMMITTEE APPROVAL FORM
This is to certify that this dissertation work titled IMPLEMENTATION OF MEOWS IN PREDICTING MATERNAL MORBIDITY AND USING IT TO IMPROVE MATERNAL OUTCOME of the candidate Dr.SARINA VINCENT AROKIA A. with Registration Number 221616052 for the award of MASTER OF SURGERY in the branch of OBSTETRICS AND GYNAECOLOGY.
I personally verified the urkund.com website for the purpose of plagiarism check. I found that the uploaded thesis file contains from introduction to conclusion pages and result shows 1 percentage of plagiarism in the dissertation.
GUIDE AND SUPERVISOR SIGN WITH SEAL
INTRODUCTION
Pregnancy and labour are normal physiological events. However it is associated with a risk of life to both mother and fetus when there is severe derangement of physiological parameters from normal. Early identification of critical illness during pregnancy and labour helps in preventing morbidity and mortality to both lives involved. Such illness during pregnancy are relatively rare and timely identification of them is compounded by the normal physiological changes associated with pregnancy and childbirth. Early identification of such events remain a challenge to healthcare professionals involved in the care of pregnant mothers.
Recording of physiological parameters is an integral part of care of pregnant mothers. This helps in early identification of problem mothers and thereby they can be referred to appropriate centres for care. Patients usually display abnormalities in simple physiological parameters before developing any serious illness. Patient’s outcome can be improved by identification of these early signs and escalation of treatment. MEOWS is an easy way of recording variables and thereby identifying pregnant mothers who require appropriate intervention. MEOWS chart includes 10 different variables which are recorded at various intervals. Small changes in the combined physiological variables measured by MEOWS may pick up deterioration earlier than an obvious change in individual variable.
Delay in diagnosis contribute to a large proportion of preventable
maternal deaths. MEOWS aids as a tool in the early recognition and measurement of the factors in assessing the condition of the deteriorating women.
Maternal early obstetric warning system includes 3 components, early warning criteria, prompt reporting & bedside evaluation. An abnormal parameter requires prompt reporting to a senior care provider followed by bedside evaluation by that care provider who should have the ability to activate resources in order to initiate emergency diagnostic and therapeutic interventions as needed. This should be followed by planning and implementation of diagnostic work-up. Close follow-up of patient’s status by a senior care provider is required until abnormality resolves or parameter judged to be benign etiology or patient is determined to be potentially critically ill and care is escalated.
Several variations of early warning charts are available for decades in the developed countries for use in adult non-pregnant patients. They are used for evaluation of physiological parameters to identify patients who can become severely ill and may require additional interventions or escalated care. However these cannot be used in pregnant patients as the normal physiological changes in pregnancy are not taken into consideration in the construction of these charts. The changes in physiology seen in normal pregnancy means, that any scoring system may need to be modified for this group of patients. Modification of such scoring charts for evaluation of pregnant mothers in the antenatal and
postnatal period has become available since a few years. They are used to record physiological parameters throughout pregnancy and upto 6 weeks of postnatal period to identify problem patients who require referral or intervention. During established labour MEOWS chart should be replaced by partogram and clear guidelines already exist regarding frequency of observations and normal physiological parameters in labour.
Tracking patients clinical responses may indicate potential deterioration providing a trigger for escalation of clinical care. In addition, it also provides guidance about patients recovery and return to stability thus facilitating a reduction in frequency and intensity of clinical monitoring and intervention.
MEOWS can be used in community or primary health set up by midwives starting as early as the pregnancy is diagnosed and can be repeatedly recorded during every antenatal visit. When an abnormal parameter is identified additional help is taken in the form of referral to higher centres or tertiary institutions where the MEOWS can again be used in addition to clinical judgement to confirm the altered health status and to provide appropriate intervention or care. MEOWS chart is colour coded for the easy identification of abnormal health status and to identify patients for referral by midwives.
Escalation of care in patients with high MEOWS score helps in improving obstetric outcome. MEOWS also helps in optimal utilisation of
available health resources particularly in developing countries like India where health resources are scarce. Implementation of use of such a score or system requires dedicated and informed health care system that acknowledges and practices such tools and provide guidelines for its use.
AIMS AND OBJECTIVES
This study aims at evaluating the use of MEOWS score as a simple bedside tool and establish its credentiality in predicting maternal morbidity and mortality. It emphasises the importance of early recognition of deterioration of apparently sick patients, ensures better communication within members of multidisciplinary team, and providing prompt attention to the unwell woman.
MATERIALS AND METHODS
MEOWS score is a score attributed to sum of all parameters documented on the MEOWS observation chart. Physiologicl parameters of 500 study subjects attending 3 peripheral health centres (Sanjeevarayanpet, Kondithope and Chennai harbour health post) from where higher numbers of referrals are made to RSRM, will be recorded on MEOWS chart. The village health nurses of these centres will be instructed to fill the charts after recording vital parameters of the patients.
A “TRIGGER” was defined as a single markedly abnormal (1 red trigger) or 2 simultaneously mildly abnormal observations (2 amber triggers).
Patients who score one or more triggers are referred to RSRM. The score of the patients are confirmed and they are followed up throughout the duration of stay in our hospital till discharge. Based on outcome at discharge, category 1 (normal and recovered without morbidity) and category 2 (recovered with morbidity or mortality) will be defined.
Performance of MEOWS chart will be evaluated and calculation of risk of morbidity will be done.
Inclusion criteria:
1. All pregnant women >28 weeks pregnancy presenting to primary health centre for antenatal checkup.
2. Antenatal women >28 weeks pregnancy, who are not in active labour.
3. Postnatal women upto 42 days of delivery, coming to PHCs for follow up.
4. Antenatal women >28 weeks pregnant, for whom health workers are conducting home visits.
5. Postnatal women upto 42 days of delivery, for whom health workers are conducting home visits.
Exclusion criteria:
1. Obstetric women who are already in intensive care settings.
2. Women receiving care in high dependency units whose risk factors are already identified like severe anemia, gestational hypertension etc.
3. Women in active labour (for whom one to one care with partogram is required).
Sample size: 500
Procedure and investigation details:
Everytime a set of observation is performed on an antenatal or postnatal woman, MEOWS score is calculated and recorded in MEOWS chart. The values of observations are translated to a summary score, with a critical threshold value above which medical review and intervention is needed.
Data collection:
Preformatted proforma enclosed in appendix.
REVIEW OF LITERATURE
Most deaths occurring during the process of pregnancy and childbirth are preventable. Most of these deaths occur in the developing countries.
Again maternal mortality rates are high in rural areas and in low socioeconomic populations. Huge difference in the maternal mortality rates between developed and developing countries reflect the fact that escalation of care can lead to a decrease of maternal mortality rates in developing countries. Certain age groups are also at increased risk – adolescent mothers and elderly gravidas (Ann K. Blanc et al, 2013).
In the past 2 decades maternal mortality rates have decreased by about 44% worldwide. This could be achieved due to the improved access to healthcare facilities and improved maternal healthcare policies in most of the developed and developing countries in the world. The WHO target of achieving maternal mortality rates of less than 70 per 1,00,000 live births, as part of Sustainable Development Agenda by the year 2030 is achievable only by implementation of systematic care of mothers before conception, during pregnancy and after childbirth all over the world (Yamey G et al, 2014).
HISTORY OF MEOWS SYSTEM
In the United Kingdom an inquiry was launched into the safety of maternity services by The King’s Fund in the year of 2007 (Smith A et
al, 2007). To their surprise a high number of failures were identified in the maternal healthcare system leading to increase in the maternal mortality. This observation by the inquiry panel of experts who are from outside the maternity services brought a fresh prespective regarding ethical issues, regulation of maternal services and various part of patient safety. Their recommendations about how safely maternity services can be implemented is published in their report titled “Safe Birth is Everybody’s Business (O’ Neill O, 2008). These recommendations for safe maternal care services were derived from previous works on patient safety that were widely accepted in clinical areas other than maternity care services. The methodology is a retrospective investigation into the incident and learning from past mistakes rather than a prospective approach of putting reliable systems of care in place to ensure safety of mothers.
The King’s Fund report was not the only report that have pointed out reasons for failure in maternity care services leading to increased morbidity and mortality which are preventable to a major extent. A series of Confidential Enquiries into Maternal Deaths have been done in the United Kingdom in the past 2 decades. They investigated the cause for all maternal deaths happening in the United Kingdom and have identified some of the critical underlying causes for maternal deaths (Wilkinson H et al, 2011). In 2008 a report was published by the Healthcare Commission after reviewing the functioning and efficacy of maternity
services. Based on their work the National Health Service Litigation Authority identified factors that reduces the likelihood of adverse events and these are reflected in Clinical Negligence Scheme for Trusts (CNST) standards in NHSLA (Wilson J et al,1997, Winn SH et al, 2007, Nicholson S et al, 2009). Valuable contributions in this regard is provided also by the Royal College of Obstetricians and Gynaecologists and the Royal College of Midwives in the form of formulating and promoting guidelines for better quality of maternal care during prenatal, antenatal and postnatal period. The King’s Fund was committed to see through all these recommendations. These recommendations are implemented and tested in multidisciplinary maternity services centre by the King’s Fund Comittee in partnership with the Royal College of Obstetricians and Gynaecologists, Royal College of Midwives, Healthcare Commission, NHS Litigation Authority, Centre for Maternal and Child Enquiries.
Number of difficulties encountered during this process were overcome with time. Shortage of workforce leading to frustrations among the staff who were working on their own initiatives to translate guidelines into practical tools, who did not have enough time to train subordinates felt overburdened by inspections and regulations. Action plans were produced based on every national report, however implementing them in the clinical practice were practically difficult. It was clear that teams needed support to make these changes happen. Hence Safer Births Programme was launched with the combined expertise of The King’s Fund and their partners. Twelve trusts were selected to join the programme. The lessons
and experiences from those organisations were also taken into account.
Most of them were clear about their shared objectives of safer care for both mothers and babies and their roles in the wider team. They communicated effectively, had trainings together where they work, including in the community. They ensured that staff with the right level of experience are available to meet the demands, recorded the informations clearly and used protocols to ensure that guidelines are followed. They collected and reviewed information on their performance and are supported by their respective boards. Although all 12 teams could not achieve all of these, they collectively acknowledged that they have to go a long way to sustain the changes they have made. Those maternity services that continued to face challenges in implementing the changes also arrived at the conclusion that with time, commitment , leadership and a bit of skill from training and experience, they can know how to execute the changes at the right time and place in a proper way.
Early Warning Scores have been used successfully in the care of critically ill patients. However these scores could not be used in obstetric patients because of the normal physiological changes that occur during pregnancy and childbirth. Hence a modification of the Early Warning Score has been introduced in obstetric population in the United Kingdom to decrease maternal mortality and morbidity by improving early detection of clinical signs of deterioration in those who develop critical illness during motherhood. This modified score, the MEOWS have
predictive ability for obstetric conditions like sepsis, haemorrhage and pre-ecclampsia (Say L et al, 2014). The parameters of the MEOWS chart accounts for this.
The existing Joint Commission standardised requirements of hospitals to have protocols for identifying early warning signs of deterioration by the staffs and to seek assistance if this occurs. Specific changes in maternal vital signs and clinical condition should trigger a predetermined response. Ever since its implementation, this tool has proven to decrease mortality in the obstetric patient. Evaluation of the efficacy of MEOWS is still continuing.
MEOWS is in widespread use in obstetric units across the United Kingdom and attempts are going on to make this gold standard countrywide. MEOWS is not widely accepted in United States as it is in United Kingdom. However, Maternal Early Warning Criteria which was drawn from MEOWS by deleting temperature and pain scores and adding oliguria as an additional measure, is being increasingly used in United States as a tool to reduce maternal mortality (Mhyre JM et al, 2014).
Validation studies of MEOWS have shown high sensitivity in predicting morbidity (89%) and reasonable specificity (79%) supporting its use for obstetric patients (Singh S et al 2012). The philosophy behind MEOWS is same as that of the majority of validated obstetric emergency training courses, i.e, recognising the risk much earlier before it could become life threatening followed by implementing a coordinated approach to manage
emergencies. This requires early communication and coordination among involved healthcare workers within a multidisciplinary team.
MEOWS encourages innovative, multidisciplinary, evidence-based practice. MEOWS provides a standardised assessment of maternal well being and can be scored using a paper based or electronic scoring chart. It has been incorporated into the rapid response policy of United Kingdom Health Services and thereby clarifies the roles and responsibilities of each staff involved in maternal care. The use of a pathway-specific maternal early warning tool results in significant reductions in both severe maternal morbidity and composite mortality.
Although several early warning tools are available for the monitoring of well being of obstetric population, the Modified Early Warning System (MEOWS) has been proposed in the United Kingdom.
Similarly the National Council for Patient Safety recently proposed the use of the Maternal Early Warning Criteria (MERC) in the United States.
The four most common causes of maternal morbidity namely haemorrhage, sepsis, pre-ecclampsia and cardiovascular dysfunction are well addressed by most of these early warning tools (Say L et al, 2014).
Only MEOWS has been prospectively tested to evaluate if their use will result in decreased maternal morbidity.
Critical illness is uncommon but potentially devastating complication of pregnancy (Baskett, 2008). It may be devastating, not
only for the woman who becomes ill, but also for her family and for those healthcare professionals responsible for her care. At its most extreme, critical illness may lead to the death of the woman during pregnancy or shortly afterwards. The Confidential Maternal Deaths Enquiry published in 2012 confirmed that Ireland continues to have a low maternal mortality ratio by international standards (O’ Hare MF 2015). However, there is no room for complacency and efforts to improve the quality of clinical care in the maternity services must be continually renewed.
Critical illness in pregnancy may be due to conditions unique to pregnancy, due to conditions exacerbated by pregnancy or due to coincidental conditions. This is reflected in the classification of maternal deaths into direct, indirect and coincidental deaths (Wilkinson H et al, 2011). The conditions unique to pregnancy include obstetric haemorrhage, pre-eclampsia/eclampsia, pulmonary embolism, chorioamnionitis/endometritis, uterine rupture, APH and acute fatty liver of pregnancy.
It is estimated that for every maternal death there are around nine women who develop severe maternal morbidity (Plaat and Naik, 2011).
In a study of severe maternal morbidity for 2004 – 2005 in the three Dublin maternity hospitals, the rate of severe maternal morbidity was 3.2 per 1,000 maternities (Murphy et al, 2009). The commonest cause was haemorrhage. A national review of postpartum haemorrhage in Ireland
over 11 years from 1999 to 2009 found that there were increasing rates of atonic postpartum haemorrhage (Lutomski et al, 2011).
As critical care has evolved worldwide, attempts to identify the deteriorating patient clinically at the most early has led to the introduction in hospitals of Early Warning Scores (Smith et al, 2013). In Ireland, this has led to the National Early Warning Score (NEWS) being developed in collaboration with the HSE Acute Medical Critical Care Programme. The NEWS was also the first guideline to be approved by the National Clinical Effectiveness Committee and it was launched in March of 2013 by the Ministry of Health.
In 2008, a hospital report following a maternal death due to infection recommended the introduction of MEOWS in Our Lady of Lourdes Hospital, Drogheda. The use of MEOWS has also been recommended by the Confidential Maternal Enquiry Reports both in the UK and Ireland (McClure et al, 2011).
In the last two ‘Saving Mother Lives” reports substandard care was identified where signs and symptoms were not recognised at time and acted upon. Both reports recommended that a National Obstetric Early Warning Scoring System should be introduced and used for all obstetric women, including those being cared for outside the obstetric setting (Al- Foudri H et al, 2010, Bamber JH, 2015). The MEOWS demonstrated a
much higher sensitivity than non-obstetric early warning systems that are currently used in the adult population.
Two countries Nigeria, and India are estimated to account for over one third of all maternal deaths worldwide, with an approximate 58,000 maternal deaths (19%) and 45,000 maternal deaths (15%) respectively (Kent A, 2010). To achieve upon the momentum generated by MDG 5, a transformative new agenda has been laid out as part of the Sustainable Development Goals to reduce the global MMR to less than 70 per 1,00,000 live births by 2030. The recent World Health Organisations publication concentrates towards ending preventable maternal mortality, establishes a national target that no country should have an MMR greater than 140 per 1,00,000 live births and outlines a strategy for achieving these targets by 2030. Planning for improving maternal health require, accurate and internationally comparable measures of maternal mortality.
The SDG calls for achieving this global goal will require countries to reduce their MMR by atleast 7.5% each year between 2016 and 2030 (Gething PW, 2016, Bongaarts J, 2016). MEOWS system can help a lot in achieving this by reducing the avoidable maternal deaths.
The causes of maternal mortality in India and strategies for reducing maternal mortality are multifactorial and variable across different socioeconomic strata and different educational level of the society. Global Maternal mortality was 2,89,000 in 2013, a 47% decline from levels in 1990. Maternal mortality is very high in Southern Asia and
Subsaharan Africa contributing for 24% and 62% of the global burden in 2013. About 80% of maternal deaths are caused by direct obstetric causes such as, haemorrhage, infection and hypertensive disorders, ruptured uterus, hepatitis and anaemia (Say Let al, 2014). As such 50% of maternal deaths due to sepsis are related to illegal induced abortion.
MMR in India has not declined significantly in the past 15 years. Age, parity, unplanned pregnancy, birth interval, socioeconomic status, nutritional status and related illegal abortion are the factors that contribute to high maternal mortality (Kent A 2010). In 1985 WHO reported that 63 – 80% of maternal deaths due to direct obstetric causes and 88 -98% of all maternal deaths due to indirect causes could probably have been prevented with proper handling. In India, improper coordination between varied levels in the delivery system and fragmentation of care accounts for the poor quality of maternal health care (Prakash A et al, 1991).
Gupta et al conducted a study in Rajasthan to ascertain the magnitude of maternal mortality rate and their causes. The study was conducted in the state of Rajasthan in India, covering 25,926 households in 411 villages. It has two major components: a community-based household survey and a case-control study with cases and controls sampled from the same population. A total of 32 maternal deaths and 6,165 live births were identified. The group of women who died during pregnancy or delivery (cases) is compared with a group of women who gave birth and survived (controls). MMR was estimated to be 519 (95%
confidence interval). Haemorrhage was the chief cause (31%) of maternal deaths; the other causes were obstructed labour, severe anemia, puerperal sepsis and abortion. Young age at childbirth (Odds ratio – 2.6; 95% CI – 1.9-3.2) and poverty (OR – 2.5; 95% CI – 1.6-3.4) were independently associated with increased risk of maternal death (Gupta SD et al, 2010).
Rationale behind our study
In a developing country like ours where the resources are scarce, implementation of MEOWS will help in reducing maternal morbidity and mortality to a greater extent. Midwives and village health nurses in the peripheral health sector are the first point of contact for the pregnant mother during the initial periods of illness or discomfort. Early recognition of alteration in certain physiological parameters and prompt recognition of danger signs in these patients can help midwives and village health nurses to refer them to higher centres for confirmation of diagnosis and care.
India aims not only to achieve the millennium development goal of reducing maternal mortality, but also to maintain the declining trend.
Hence investigating a method which would help the peripheral health sector personnel to easily identify women who need additional care would greatly help in achieving this mission. MEOWS system helps for this by providing a single chart to plot important parameters and facilitate early referral of potentially ill patients.
MEOWS System
Various types of early warning systems are in use in different parts of the world with slight changes in parameters recorded. A number of different modifications has been made to most of these warning systems to accommodate newer parameters or to delete existing parameters in order to improve outcome. The one used in NHS is in use since 2003 and was modified later. The currently used MEOWS system includes 10 parameters. They are
1. Respiratory rate 2. Pulse rate
3. Oxygen saturation 4. Temperature 5. Systolic BP 6. Diastolic BP 7. Protienuria
8. Neurological response 9. Pain score
10. Lochia
This chart is used by staff who are attending antenatal and postnatal mothers during both routine and emergency visits. A score of 3 or more triggers a call out cascade which gives instructions about, further monitoring, review and need for referral.
Good communication among different cadres of healthcare professionals and prompt referral has been the cornerstone for successful management of pregnant mothers when they show any symptoms or signs of deflection from normal pathway of pregnancy. Trainee doctors and midwives should have a low threshold for referring patients to higher centres, and the tool used for assessment of mothers is designed in such a way to facilitate this. The referral should be directed to a senior member of the care taking team and on an urgent basis bypassing local commissioning rules for referral. The referring health personnel must receive a response in return of referral. The recipient health personnel must respond promptly in assessing the health condition of the patient, and if not, the sender must follow it up. The authors of ‘Saving mothers lives: Reviewing maternal deaths to make motherhood safe – 2006-08’
also recommend the routine use of national modified early obstetric warning score (MEOWS) chart in all pregnant or postpartum women who become unwell and require obstetric services care. They also suggest that these charts should not only be used in community and obstetrics units , but also in other departments of the hospitals where pregnant and postpartum patients are likely to be managed for their health problems, like in the emergency department.
The frequency and method of documentation of clinical parameters in hospitalized pregnant and postpartum mothers are described below.
Frequency is determined by risk status, diagnosis, reason for admission
and initial observations on admission. An individual plan of care should be made by the midwife and the doctor which should specify the frequency of physiological observations and where they are documented.
Maternal Early Obstetric Warning System has got 3 major components 1. Early Warning Criteria
2. Bedside evaluation 3. Prompt reporting
The Early Warning Criteria included in the MEOWS system has been enumerated previously and is described below in further detail. A complete set of observations is made whenever the patient is evaluated and charted. The value of the observations are then translated into a summary score which has a critical threshold, above which medical review and interventions is required. It is believed that small changes in the combined physiological variables measured by MEOWS may pick up deterioration earlier than the obvious change in an individual variable.
Early detection will trigger subsequent prompt intervention that will either reverse further physiological decline or facilitate timely referral to appropriate centre.
The use of MEOWS does not demand critical care or define treatment, but is a tool to aid in the early recognition and management of the deteriorating woman. However no diagnostic tool can replace the
actual physical examination of a woman and clinical assessment of her condition. In some cases of maternal collapse there may be no prior warning symptoms and signs, although they may have risk factors that make this more likely. Often there are clinical signs that precede collapse.
The use of an early warning score is supported by NICE guidelines (NIfHaC E, 2007).
Respiratory rate:
Respiratory rate is the most sensitive indicator of deteriorating physiology and must be recorded in all women everytime a full set of observation is made. Respiratory rate is a mandatory observation as it is the best marker of a sick woman and is the first observation that will indicate a problem or deterioration in condition.
Pulse rate:
Tachycardia is the key parameter for early detection of critical illness in maternal obstetric patients. Pulse should be measured manually to asses rate, volume and regularity. Pulse rate can be monitored via a saturation probe on the finger. But if the women is peripherally shut down, as in cases of hemorrhage, the pulse oxymetry probe will not detect the pulse rate accurately. Moreover pulse oxymeter does not give an idea regarding regularity and pulse volume. Nail paints commonly used by women also affects waveform and accuracy. Tachycardia is often an early sign of deterioration and can indicate an impending collapse and
hemodynamic abnormality. A tachycardic women should be considered hypovolemic until proven otherwise. Hence reading of pulse should be done manually and with utmost care not to miss an early sign of deterioration.
Oxygen saturation:
Hand held pulse oxymeter is an easy and invaluable method of monitoring blood oxygen levels, particularly in patients with respiratory distress or abnormal respiratory rates and pattern. If a patient is receiving oxygen, then the rate of administered oxygen should be documented underneath the saturations.
Blood pressure:
Blood pressure should be recorded using correct cuff size, especially in obese women. All pregnant women with a systolic blood pressure of 160mm of Hg or more require antihypertensive treatment.
Also a reading of 160mm of Hg triggers a red score. Pregnant women can lose upto 30-40% of total circulating blood volume before any change in blood pressure can be appreciated while recording. Hence falling BP should be regarded as a late sign of deterioration. BP recording should preferably be done manually using an aneroid BP apparatus as electronic recordings can underestimate BP readings to the tune of 5%. If electronic recording of BP is found to be raised, it should be rechecked manually atleast once using an aneroid BP machine.
Width(cm) Length(cm) Arm circumference(cm)
Normal 12.0 – 13.0 23 Upto 33
Large adult 12.5 – 13.0 35 Upto 42
Urine output and proteinuria:
Urine output should be charted on a separate fluid balance chart in admitted patients. Postoperative women and high risk mothers should have fluid balance chart maintained. The optimum urine output is 1ml/kg/hr and the minimum urine output is 0.5ml/kg/hr. Urine output is one of the few signs of end organ perfusion. When a fluid balance chart is used, it should be accurately filled in with both measured input and output. Proteinuria is an early sign of preeclampsia. Qualitative measurement of urinary protein content using protein dipstick is an easy way of assessing proteinuria at bedside. A urine protein dipstick test result of >2+ triggers a red score in the MEOWS chart and warrants further evaluation for gestational hypertension.
Conscious level:
Conscious level should be assessed on all women and recorded using AVPU scale.
A – Alert and conscious
V – Responds to voice P – Responds to pain U – Unresponsive
A fall in AVPU score should always be considered significant.
Temperature:
Temperature change may not necessarily be an effective measure of deterioration. A fall or rise in temperature may indicate sepsis, and a sepsis screen and appropriate antibiotic therapy should be considered.
Septicemic shock can be particularly difficult to recognise. A collapsed septic patient may exhibit all the signs and symptoms of hypovolemia, but if there is no positive response from fluid resuscitation after 10 minutes, then septicemic shock should be considered and a full sepsis screen ordered urgently.
Pain score:
Women who experience unexplained pain should be evaluated thoroughly and referred to higher level of care centres. Intense epigastric pain is more likely a marker of imminent eclampsia and needs admission and continuous monitoring. Other differential diagnosis considered in pregnant mothers of severe abdominal pain are pelvic abscess (where patient have severe abdominal pain), abruption placenta, uterine rupture
and cardiovascular disease. When assessing pain, clinical bias should be avoided and proper judgement is recommended. Pain is assessed using a 4 tier pain scale with 0 corresponding to no pain and 3 corresponding to intolerable pain. A pain score of 2 or 3 triggers a red signal.
Lochia:
The vaginal discharge in puerperium which lasts for 4 to 6 weeks is an important marker of sepsis in postnatal period and is also noted while examining a postnatal period and is also noted while examining a postnatal woman. MEOWS chart in some countries do not have lochia as a parameter.
Guidelines for the use of MEOWS:
Effective warning systems include clear expectations for observation, predefined criteria for an abnormality, and a protocol to trigger a response if any abnormality is detected. The MEOWS is calculated by scoring a full set of observations carried out routinely by staff which includes all the above described variables. Of all the variables the heart rate, BP and respiratory rate are the sensitive indicator in the same order.
MEOWS scoring:
Score 3 2 1 0 1 2 3
Temperature
(C) <35 35 -37.4 37.5- 39 >39
Systolic BP
(mm Hg) <70 71 – 79 81-89 90 – 139 140-149 150-159 >160 Diastolic BP
(mmHg) <45 46 – 89 90-99 100-109 >110
Pulse
(per min) <40 40-50 51–100 101-110 111-129 >130
Respiratory rate (per min)
<8 9 -14 15-20 21-29 >30
O2 saturation <95 95-100
AVPU score Alert Responds
to voice
Responds
to pain Unconscious
Urine output
(ml/Hr) <10 <30 Not
measured
Pain score Severe
pain Moderate Mild
pain No pain
Lochia Abnormal Normal
If the pulse rate is higher than systolic BP then score 2 for pulse.
Frequency of scoring MEOWS in hospitalised patients:
Everytime a set of observation is performed in either ante or postnatal women, MEOWS should be calculated and recorded in the observation chart. All women presenting to Triage who are having baseline observations carried out should have a MEOWS calculated and documented in the records every 12th hourly. In order to empirically derive an early warning score to predict maternal death, physiologic data from obstetric patients were used.
Antenatal:
Frequency of observations will depend on the nature of the admission or as indicated by the lead clinician. Full set of observations should be carried out twice daily atleast 12 hours apart minimum.
Delivery room:
All women should have a set of MEOWS observations documented in the records on admission in delivery room. Currently women who are in labour need not have MEOWS repeated. Regular observations should still be documented on the partogram as usual. However a score attributable to the baseline observations on admission and recorded in the records. High risk women receiving care in delivery room should have the MEOWS score documented on the mega chart. Normal MEOWS observation chart should be started once the mega chart is no longer used.
Recovery room:
The MEOWS chart should be initiated in the recovery room by the recovery practitioner prior to transfer to postnatal ward.
Postnatal ward:
All women should have a full set of observations on admission to the postnasal ward and should have this repeated a minimum of 12 hours apart. A MEOWS score should be attributed to every set of observations.
The frequency of observations will depend on the nature of the admission or as indicated by the lead practitioner.
Others:
All obstetric inpatients must have a full set of observations and a MEOWS calculated at every level of transfer to a new area, i.e, for example for transfer from recovery room to postnatal ward. The MEOWS chart used in one area should be transferred with the patient to the next area in order to help identify changes in trends of observations.
Triggering on MEOWS:
A trigger is defined as a single markedly abnormal observation (red trigger), or the combination of two simultaneous mildly abnormal observations (two amber triggers).
PARAMETER RED TRIGGER AMBER TRIGGER Temperature <35 or >38C 35 – 36C
Systolic BP <90 or >160 mm Hg 150 – 160 mm Hg 90 – 100 mm Hg Diastolic BP >100 mm Hg 90 – 100 mm Hg Heart rate <40 or >120 bpm 100 – 120 bpm
40 – 50 bpm Respiratory rate <10 or >30/min 21 – 30/min O2 saturation <95%
Neurological score U – Unresponsive
P – Responds to pain V – Responds to voice
Proteinuria >2+ 2+
Pain score - 2-3
Lochia - Heavy, offensive.
It is important to remember that when the women triggers she requires referral to appropriate level doctor, monitoring, review or repeat of investigations and a plan of care. Recognition of deterioration in condition does not necessarily mean diagnosis, but does mean investigation and appropriate level referral involving a multidisciplinary approach. Any woman who triggers 4 should have their full set of observations frequently repeated depending on the diagnosis.
Actions to be taken when a woman is triggering on MEOWS chart:
1. Immediate midwifery measures of escalation pathway, she should know which level of clinician you are alerting. Inform labour ward coordinator.
Make sure you have all the information you need to inform the physician to be written on hand notes, charts, blood results etc.
2. State the current problem giving the observation findings and state which ones are triggering. If raised systolic or diastolic BP, report any signs such as headache, nausea, vomiting or upper epigastric pain. Be clear about your expectations of the clinician that the women requires a bedside review in less than 10 minutes.
3. Immediate midwifery measures to be continued, increase observation frequency to ¼ hourly. Explanation of plan of care to the woman and relatives is required. Ensure you have senior midwife help and consider location of the woman. Arrangements may need to be made to transfer to labour ward or ICU. Monitor saturatuion levels. Administer O2 via facemask if required, assess patency of airway. If the women is awake and talking, ask her about any signs or changes she perceives. Check whether the iv lines are running and ensure that there is no extravasation or swelling at puncture site. Check the drug chart and ensure medications have been administered, report time of delay of any
drugs, especially antihypertensives. Decide the optimum position of patient in the bed, whether sitting upright or lowering bed end. If antenatal, apply left lateral tilt 15 – 30 degrees and commence CTG. Get blood results from the laboratory. Bring ECG machine, haemacue and arterial blood gas analyser to bedside. Arrange for blood, if required. Maintain records in notes detailing plan of care.
The midwife and labour room coordinator should ensure that appropriate level of clinician attends the case and consider escalation depending on patient condition. It is important to care for the woman in the most appropriate clinical area. If this is not possible, then a delay in transfer must not delay immediate investigations. Full review of the notes including history taking and examination should be done.
SCORE
MEOWS≤2
Continue Current Treatment Plan
MEOWS-3
Informcoordinator and senior staff. Repeat observation. Review with senior staff
consider medical officer review
MEOWS≥4
Inform coordinator and senior staff, contact senior resident within
30 mins, contact registrar and senior obstetric registrar
Consider ABC
A and B-O2 Therapy
C-IV Access, Fluid resuscitation
Catheterisation
Left Lateral tilt
Increase frequency of observations
Fetal Monitoring
ECG, Analgesia
Required investigations MEOWS≥6
Inform coordinator and senior staff, contact senior resident within
30 mins, contact registrar and senior obstetric registrar,
Anaesthetic registrar
If Patient deteriorates, fails to respond, contact consultant obstetrician and consultant anaesthetist on call.
Alert obstetric emergency team
RESULTS
STATISTICAL ANALYSIS
The collected data were analysed with IBM.SPSS statistics software 23.0 Version. To describe about the data, descriptive statistics, frequency analysis, percentage analysis were used for categorical variables and the mean & S.D were used for continuous variables.
The Hosmer–Lemeshow test is used for goodness of fit in logistic regression for morbidity risk prediction model. To assess the relationship between the variables Spearman's Correlation was used. To find the significance in categorical data Chi-Square test was used. In all the above statistical tools the probability value .05 is considered as significant level.
AGE
Age Frequency Percent
Up to 20 yrs 80 16.0
21 - 25 yrs 216 43.2
26 - 30 yrs 136 27.2
31 - 35 yrs 56 11.2
Above 35 yrs 12 2.4
Total 500 100.0
The above table shows the frequency distribution of different age groups. The frequency and the percentage of the age groups are: Up to 20yrs - 80 and 16%, 21-25yrs - 216 and 43.2%, 26-30yrs - 136 and 27.2%, 31-35yrs - 56 and 11.2%, above 35yrs - 12 and 2.4%.
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0
Upto 20 yrs 21 - 25 yrs 26 - 30 yrs 31 - 35 yrs Above 35 yrs 16.0
43.2
27.2
11.2
2.4
Age range
PRIMI Vs MULTI
OB CODE Frequency Percent
MULTI 323 64.6
PRIMI 177 35.4
Total 500 100.0
The above table shows the frequency distribution of gravid order of the study subjects. The frequency and the percentage of the obstetric code are: Multigravida has a frequency of 323 and percentage 64.6%.
Primigravida has a frequency of 177 and percentage 35.4%
65%
35%
OB Code
MULTI PRIMI
CCU ADMISSION
CCU Frequency Percent
NO 403 80.6
YES 97 19.4
Total 500 100.0
The above table shows the frequency distribution of CCU admission. The frequency and percentage of CCU admissions are 97 and 19.4%. 403 out of 500 (80.6%) study subjects did not require CCU admissions.
81%
19%
CCU
NO YES
HDU ADMISSION
HDU Frequency Percent
NO 205 41.0
YES 295 59.0
Total 500 100.0
The above table shows the frequency distribution of HDU admissions. The frequency and percentage of HDU admissions are 295 and 59%. 205 out of 500 (41%) study subjects did not require any HDU admissions.
41%
59%
HDU
NO YES
IONOTROPIC SUPPORT
IONOTR SUPPORT Frequency Percent
NO 478 95.6
YES 22 4.4
Total 500 100.0
The above table shows the frequency distribution of patients requiring ionotropic support. The frequency and percentage of study subjects requiring ionotropic support are 22 and 4.4%. 478 out of 500 (95.6%) study subjects did not require ionotropic support.
96%
4%
IONOTR SUPPORT
NO YES
BLOOD TRANSFUSION
Blood Trans Frequency Percent
NO 341 68.2
YES 159 31.8
Total 500 100.0
The above table shows the frequency distribution of patients requiring blood transfusion. The frequency and percentage of blood transfusion are 159 and 31.8%. 341 out of 500 (68.2%) of patients did not require any blood transfusion.
68%
32%
BLOOD TRANS
NO YES
BLOOD PRODUCTS TRANSFUSION
Blood Products Frequency Percent
NO 460 92.0
YES 40 8.0
Total 500 100.0
The above table shows the frequency distribution of blood products requirement among study subjects. The frequency and percentage of blood products transfusion are 40 and 8%. 460 out of 500 (92%) patients did not require transfusion of any blood products.
92%
8%
BL PRODUCTS
NO YES
VENTILATOR SUPPORT
Ventilator Frequency Percent
NO 419 83.8
YES 81 16.2
Total 500 100.0
The above table shows the frequency distribution for ventilatory support. The frequency and percentage of ventilator support are 81 and 16.2%. 419 out of 500 (83.8%) of study subjects did not require any ventilator support.
84%
16%
VENTILATOR
NO YES
MAGSULF THERAPY
MgSO4 Frequency Percent
NO 349 69.8
YES 151 30.2
Total 500 100.0
The above table shows the frequency distribution of MgSO4 usage in the study subjects. The frequency and percentage of MgSO4 treatment are 151 and 30.2%. 349 out of 500 (69.8%) of study subjects did not require treatment with MgSO4.
70%
30%
MGSO
4NO YES
NASG GARMENT USE
NASG GARMENT Frequency Percent
NO 487 97.4
YES 13 2.6
Total 500 100.0
The above table shows the frequency distribution of NASG garment usage in the study subjects. The frequency and percentage of NASG garment usage are 13 and 2.6%. 487 out of 500 (97.4%) of study subjects did not require NASG garment.
97%
3%
NASG GARMENT
NO YES
MORBIDITY
MORBIDITY Frequency Percent
NO 67 13.4
YES 433 86.6
Total 500 100.0
The above table shows the frequency distribution of maternal morbidity among study subjects. The frequency and percentage of maternal morbidity are 433 and 86.6%. 67 out of 500 (13.4%) did not face any morbidity.
13%
87%
MORBIDITY
NO YES
REFERENCE MEOWS
REF MEOWS Frequency Percent
Low 292 58.4
Medium 181 36.2
High 27 5.4
Total 500 100.0
The above table shows the frequency distribution of referral MEOWS score. The frequency and percentage of referral MEOWS are:
LOW has a frequency of 292 and percentage 58.4%. MEDIUM has a frequency of 181 and percentage 36.6%. HIGH has a frequency of 27 and percentage 5.4%.
58%
36%
6%
Reference MEOWS
Low Medium High
HOSPITAL MEOWS SCORE
Hospital Meows Frequency Percent
Low 306 61.2
Medium 171 34.2
High 23 4.6
Total 500 100.0
The above table shows the frequency distribution of hospital MEOWS score. The frequency and percentage of hospital MEOWS score are: LOW has a frequency of 306 and percentage 61.2%. MEDIUM has a frequency of 171 and percentage 34.2%. HIGH has a frequency of 23 and percentage 4.6%.
61%
34%
5%
MEOWS Score
Low Medium High
Descriptive
Statistics Age Ref Meows
Meows Score
Baby Wt
Duration Of Stay
Morbidity Score
Mean 25.17 3.31 3.22 2.55 8.70 1.92
Median 24.00 3.00 3.00 2.60 8.00 2.00
Std.Deviation 4.60 1.21 1.21 0.63 2.74 1.51
Range 25 9 9 3.60 26 7
Minimum 17 1 1 0.60 2 0
Maximum 42 10 10 4.20 28 7
OB CODE OB CODE
Total MULTI PRIMI
M Score
Low Count 197 109 306
% 64.4% 35.6% 100.0%
Medium Count 112 59 171
% 65.5% 34.5% 100.0%
High Count 14 9 23
% 60.9% 39.1% 100.0%
Total Count 323 177 500
% 64.6% 35.4% 100.0%
The above table shows the comparison between MEOWS score with obstetric code. Shows no statistical significance with p=0.902 >
0.05.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Low Medium High
OB Code
Primi Multi
DURATION OF HOSPITAL STAY Duration of Stay
Total
< 7 DAYS
>= 7 DAYS
M Score
Low Count 90 216 306
% 29.4% 70.6% 100.0%
Medium Count 11 160 171
% 6.4% 93.6% 100.0%
High Count 0 23 23
% 0.0% 100.0% 100.0%
Total Count 101 399 500
% 20.2% 79.8% 100.0%
The above table shows the comparison between MEOWS score with duration of stay. Shows high statistical significance with p=0.0005 ≤ 0.01
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Low Medium High
Hospital stay days
< 7 DAYS >= 7 DAYS
CCU ADMISSIONS
CCU
Total NO YES
M Score
Low Count 286 20 306
% 93.5% 6.5% 100.0%
Medium Count 110 61 171
% 64.3% 35.7% 100.0%
High Count 7 16 23
% 30.4% 69.6% 100.0%
Total Count 403 97 500
% 80.6% 19.4% 100.0%
The above table shows the comparison between MEOWS score with CCU admissions. Shows High statistical significance with p=0.0005
≤ 0.01
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Low Medium High
CCU
No Yes
HDU ADMISSIONS
HDU
Total NO YES
M Score
Low Count 127 179 306
% 41.5% 58.5% 100.0%
Medium Count 62 109 171
% 36.3% 63.7% 100.0%
High Count 16 7 23
% 69.6% 30.4% 100.0%
Total Count 205 295 500
% 41.0% 59.0% 100.0%
The above table shows the comparison between MEOWS score with HDU admissions. Shows High statistical significance with p=0.009
≤ 0.01
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Low Medium High
HDU
No Yes
IONTOROPIC SUPPORT
IONOTR Support
Total NO YES
M Score
Low Count 301 5 306
% 98.4% 1.6% 100.0%
Medium Count 159 12 171
% 93.0% 7.0% 100.0%
High Count 18 5 23
% 78.3% 21.7% 100.0%
Total Count 478 22 500
% 95.6% 4.4% 100.0%
The above table shows the comparison between MEOWS score with ionotropic requirement. Shows High statistical significance with p=0.0005 ≤ 0.01
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Low Medium High
IONOTR Support
No Yes
BLOOD TRANSFUSION
BLOOD TRANS
Total
NO YES
M Score
Low Count 231 75 306
% 75.5% 24.5% 100.0%
Medium Count 98 73 171
% 57.3% 42.7% 100.0%
High Count 12 11 23
% 52.2% 47.8% 100.0%
Total Count 341 159 500
% 68.2% 31.8% 100.0%
The above table shows the comparison between MEOWS score with blood transfusion requirements. Shows High statistical significance with p=0.0005 ≤ 0.01
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Low Medium High
Blood tranfusion
No Yes
BLOOD PRODUCTS TRANSFUSION
BL Products
Total NO YES
MScore
Low Count 297 9 306
% 97.1% 2.9% 100.0%
Medium Count 150 21 171
% 87.7% 12.3% 100.0%
High Count 13 10 23
% 56.5% 43.5% 100.0%
Total Count 460 40 500
% 92.0% 8.0% 100.0%
The above table shows the comparison between MEOWS score with blood products requirement. Shows High statistical significance with p=0.0005 ≤ 0.01
0%
20%
40%
60%
80%
100%
Low Medium High
BL Products
No Yes
