Background
The recently published monograph, Neonatal encephalopathy and neurologic outcome, from the American College of Obstetricians and Gynecologists calls for a root cause analysis to identify components of care that contributed to cases of neonatal encephalopathy to design better practices, surveillance mechanisms, and systems. All cases of infants born in New Zealand with moderate and severe neonatal encephalopathy were reported to the New Zealand Perinatal and Maternal Mortality Review Committee from 2010. A national clinical review of these individual cases has not previously been undertaken.
Objectives
The objective of the study was to undertake a multidisciplinary structured review of all cases of neonatal encephalopathy that arose following the onset of labor in the absence of acute peripartum events in 2010–2011 to determine the frequency of contributory factors, the proportion of potentially avoidable morbidity and mortality and to identify themes for quality improvement.
Study Design
National identification of, and collection of clinical records on, cases of moderate or severe neonatal encephalopathy occurring after the onset of labor in the absence of an acute peripartum event, excluding those with normal gases and Apgar scores at 1 minute, among all cases of moderate and severe neonatal encephalopathy at term in New Zealand in 2010–2011 was undertaken. Cases were included if they had abnormal gases as defined by any of pH of ≤ 7.2, base excess of ≤ –10, or lactate of ≥ 6 or if there were no cord gases, an Apgar score at 1 minute of ≤ 7. A clinical case review was undertaken by a multidisciplinary team using a structured tool to record contributory factors (organization and/or management, personnel, and barriers to access and/or engagement with care), potentially avoidable morbidity and mortality and to identify themes to guide quality improvement.
Results
Eighty-three babies fulfilled the inclusion criteria for the review, 56 moderate (67%) and 27 severe (33%), 21 (25%) of whom were deceased prior to hospital discharge. Eighty-four percent of 64 babies with cord gas results had one of pH of ≤ 7.0, base excess of ≤ –12, or lactate of ≥ 6; and 42% (8 of 19) without cord gases had 5 minute Apgar scores < 5. Excluding 5 babies who died within a day of birth, all but 1 baby were admitted to a neonatal unit within 1 day of birth. Contributory factors were identified in 84% of 83 cases, most commonly personnel factors (76%). Fifty-five percent of cases with morbidity or mortality were considered to be potentially avoidable, and 52% of cases were considered potentially avoidable because of personnel factors. The most frequently identified theme related to the use and interpretation of cardiotocography in labor.
Conclusion
A multidisciplinary case review of neonatal encephalopathy following apparently uncomplicated labor identified a high rate of potentially avoidable morbidity and mortality and issues amenable to quality improvement such as multidisciplinary training of staff in fetal surveillance in labor.
Reviews of care among cases of neonatal encephalopathy (NE) from Europe and South Africa have reported substandard pregnancy care, most often related to care provided by health professionals, including issues with risk assessment and fetal monitoring. Furthermore, the recently published monograph, Neonatal encephalopathy and neurologic outcome, from the American College of Obstetricians and Gynecologists calls for a root cause analysis to identify components of care that contributed to cases of NE to design better practices, surveillance mechanisms, and systems.
The Neonatal Encephalopathy Working Group (NEWG) of the Perinatal and Maternal Mortality Review Committee (PMMRC) of New Zealand has prospectively collected data on cases of moderate and severe NE at term in New Zealand since 2010. The PMMRC is required to report mortality and morbidity of mothers and babies in New Zealand and by critical analysis identify areas in which there may be potential for improvement of services and outcomes. However, to date there has been no detailed review of individual NE case records to elicit contributory factors and estimate potential avoidability.
New Zealand has approximately 65,000 births annually. The maternity model of care in 2010 involved a self-employed midwifery lead maternity carer for 80% of women, a private obstetrician for 6%, a general practitioner for 1.5%, and a hospital primary midwife, secondary care, or no antenatal care for 15%. Whereas the responsibility for care sits with the lead maternity carer, there are clear guidelines for referral to an obstetrician and responsibility for care may therefore change during pregnancy or in labor as required and on discussion with the woman and her caregiver. In 2011 in New Zealand, 3.3% of babies were born at home and 10.1% in primary facilities compared with 40.9% in secondary hospitals and 45.7% in tertiary level facilities.
The aim of the current study was to undertake a structured multidisciplinary review of individual cases to determine whether there were contributory factors to death and morbidity from NE, whether death or the severity of morbidity could have been prevented, and to identify areas in which improvements to maternity and neonatal care in New Zealand might reduce the rate and morbidity from NE.
The study aimed to undertake multidisciplinary clinical case review of pregnancies resulting in NE, in which the encephalopathy was considered to be the result of events occurring during the intrapartum period but not because of acute peripartum events.
The specific objectives of the review among this group of cases were to determine the rate of contributory factors (organization and/or management, personnel, and barriers to access and/or engagement with care), the rate of potentially avoidable mortality or morbidity, and to identify key areas in which quality improvement initiatives should focus resource.
Materials and Methods
Cases in 2010–2011 for multidisciplinary review were identified from the national database of term babies born with moderate or severe NE. The time period 2010–2011 was chosen because full data were available for this cohort. NE was defined, for the national collection, as a clinically defined syndrome of disturbed neurological function within the first week of life in the term infant (≥ 37 weeks’ gestation), manifested by difficulty in initiating and maintaining respiration, depression of tone and reflexes, subnormal level of consciousness, and often seizures. Only Sarnat moderate and severe cases were included.
Cases were identified and reported by pediatricians via the New Zealand Pediatric Surveillance Unit methodology, a well-established and effective mechanism for collection of data on uncommon pediatric disorders. To ensure full case ascertainment, key clinicians were also identified and asked to provide details of all cases that met the definition in their hospitals. Term neonatal hypoxic peripartum and/or hypoxic ischemic encephalopathy deaths in the PMMRC data set, validated against national health and mortality data sets, were also included if not already identified via the New Zealand Pediatric Surveillance Unit.
The cases were limited to those born by emergency cesarean delivery (CD), operative vaginal, or unassisted vaginal birth after laboring without an identified acute event in labor (cord prolapse, abruption, uterine rupture, shoulder dystocia, vasa previa, head entrapment with breech, maternal collapse, or arrest), who had abnormal (or no) blood gases (defined as any of pH of ≤ 7.2, base excess of ≤ –10, or lactate of ≥ 6) at birth and/or an Apgar score at 1 minute of ≤ 7. These cases were chosen for review because there was no obvious cause for the NE and it was hypothesized significant learnings might arise from review.
The case reviews followed the methodology developed by the PMMRC for the review of maternal mortality since this was used for maternal morbidity and perinatal mortality, using a checklist tool ( Table 1 ) to identify contributory factors and potentially avoidable mortality/morbidity. The review utilized copies of all case notes and data collected from the submitted forms covering antenatal care, intrapartum care, neonatal resuscitation, and early neonatal management.
| Factors | NE (n = 83) | Deceased (n = 21) | Survivors (n = 62) | |||
|---|---|---|---|---|---|---|
| % | n/N | % | n/N | % | n/N | |
| Contributory factor | 84 | 70/83 | 86 | 18/21 | 84 | 52/62 |
| Organization/management factors | 37 | 31/83 | 33 | 7/21 | 39 | 24/62 |
| Poor organizational arrangements of staff | 4 | 3/83 | 0 | 0 | 5 | 3/62 |
| Lack of policies, protocols, or guidelines | 6 | 5/83 | 0 | 0 | 8 | 5/62 |
| Inadequate numbers of staff | 5 | 4/83 | 5 | 1/21 | 5 | 3/62 |
| Poor access to senior clinical staff | 7 | 6/83 | 10 | 2/21 | 6 | 4/62 |
| Delay in procedure (eg, cesarean delivery) | 10 | 8/83 | 5 | 1/21 | 11 | 7/62 |
| Delayed access to test results or inaccurate results | 8 | 7/83 | 10 | 2/21 | 8 | 5/62 |
| Other | 16 | 13/83 | 19 | 4/21 | 15 | 9/62 |
| Personnel factors | 76 | 63/83 | 76 | 16/21 | 76 | 47/62 |
| Knowledge and skills of staff were lacking | 34 | 28/83 | 43 | 9/21 | 31 | 19/62 |
| Delayed emergency response by staff | 27 | 22/83 | 24 | 5/21 | 27 | 17/62 |
| Failure to maintain competence | 5 | 4/83 | 5 | 1/21 | 5 | 3/62 |
| Failure of communication between staff | 12 | 10/83 | 14 | 3/21 | 11 | 7/62 |
| Failure to seek help/supervision | 28 | 23/83 | 38 | 8/21 | 24 | 15/62 |
| Failure to offer or follow recommended best practice | 58 | 48/83 | 57 | 12/21 | 58 | 36/62 |
| Lack of recognition of complexity or seriousness of condition by caregiver | 55 | 46/83 | 52 | 11/21 | 56 | 35/62 |
| Other | 1 | 1/83 | 5 | 1/21 | 0 | 0 |
| Barriers to access or engagement with care | 24 | 20/83 | 24 | 5/21 | 24 | 15/62 |
| Infrequent care or late booking | 8 | 7/83 | 10 | 2/21 | 8 | 5/62 |
| Declined treatment or advice | 4 | 3/83 | 5 | 1/21 | 3 | 2/62 |
| Obesity having an impact on delivery of optimal care (eg, USS) | 5 | 4/83 | 5 | 1/21 | 5 | 3/62 |
| Environment (eg, isolated, long transfer, weather prevented transport) | 5 | 4/83 | 0 | 0 | 6 | 4/62 |
| Other | 8 | 7/83 | 10 | 2/21 | 8 | 5/62 |
The review teams were multidisciplinary and included pediatricians, neonatologists, neonatal nurse practitioners, obstetricians, and midwives. Every case review included at least 1 representative from pediatrics, obstetrics, and midwifery. Clinicians did not perform a review of cases if they had been involved directly in the care. In some cases team members may have had prior knowledge or were aware of local circumstances related to a specific case and accordingly were encouraged to contribute any relevant further information. All reviewers had previous experience with the described process of case review.
A summary and time line were prepared for each case by a research assistant. In addition, the case notes were reviewed by 2 members of the multidisciplinary team (1 maternity and 1 pediatric), and these reviewers then presented the case to the full panel prior to discussion and then scoring using the checklist tool ( Table 1 ) and discussion of specific issues or potential recommendations.
Contributory factors were defined as modifiable components of the health system and issues of quality of care covering a broad spectrum of management, personnel, and access/engagement with care that had an impact on the outcome of the case. Each of these domains includes a checklist of 4-11 items ( Table 1 ). One or multiple categories and subcategories could be selected in any individual case.
Potentially avoidable mortality or morbidity was determined where the absence of any of the contributory factors could have resulted in survival or in less severe morbidity. Potentially avoidable mortality or morbidity was determined following the assessment of contributory factors, and the team decided which contributory factor or factors had led to a case being defined as potentially avoidable. The team was also asked to identify case-specific issues to inform the development of recommendations to drive quality improvement.
When all of the review meetings were complete, a meeting was held to review the overall findings and to prioritize the issues raised from individual case reviews into thematic areas to develop recommendations or areas for further development.
Data collected by the NEWG at the notification of the cases were made available by the PMMRC for the purposes of this study and were merged with the data from the clinical reviews for further analysis. The univariate analyses were performed using STATA version 9.2 (StataCorp, College Station, TX). The frequencies of the contributory factors were compared using χ 2 tests.
Institutional review board approval for this study was not required because the case reviews were completed under the New Zealand Public Health and Disability Act of 2000 (NZPHD Act). The PMMRC was established under the NZPHD Act, which provides the PMMRC with the power to request information relevant to the committee’s functions of reviewing and reporting on morbidity and mortality with a view to reducing both. The NZPHD Act outlines the PMMRC’s responsibility for ensuring the information provided is kept confidential and all publications are grouped and individuals are not identifiable.
Results
In the years 2010–2011, 149 cases of NE were reported (1.27 per 1000 term births). Of these cases, 96 (64%) were assessed as moderate and 53 (36%) as severe. Thirty-six (24%) babies died before discharge from the hospital. Descriptive data are available in the PMMRC Annual Reports 2011 and 2012.
Eighty-three babies, 56 (67%) moderate, and 27 (33%) severe fulfilled the inclusion criteria for the review, 21 (25%) of whom were deceased prior to hospital discharge. Excluding 5 babies who died within a day of birth, all but 1 baby were admitted to a neonatal unit within 1 day of birth. One baby was admitted on day 2. Forty-five (54%) had magnetic resonance imaging performed between day 1 and day 28 of life, and of these 18 (40%) were reported as moderate or severely abnormal and 26 (58%) normal or mildly abnormal. Mode of birth, blood gas results, and Apgar scores at 1, 5, and 10 minutes are shown in Table 2 .
| Clinical characteristic | NE (n = 83) | Deceased (n = 21) | Survivors (n = 62) | |||
|---|---|---|---|---|---|---|
| % | n/N | % | n/N | % | n/N | |
| Sarnat stage | ||||||
| Moderate | 67 | 56/83 | 5 | 1/21 | 89 | 55/62 |
| Severe | 33 | 27/83 | 95 | 20/21 | 11 | 7/62 |
| Gestation (mean [SD]) | 39.5 (1.3) | 39.3 (1.6) | 39.6 (1.2) | |||
| Birthweight (mean [SD]) | 3412 (520) | 3327 (710) | 3439 (441) | |||
| Mode of birth | ||||||
| Cesarean delivery | 30 | 25/83 | 29 | 6/21 | 31 | 19/62 |
| Normal vaginal delivery | 49 | 41/83 | 57 | 12/21 | 48 | 29/62 |
| Operative vaginal delivery | 18 | 15/83 | 14 | 3/21 | 19 | 12/62 |
| Vaginal breech | 2 | 2/83 | 0 | 3 | 2/62 | |
| Apgar score | ||||||
| At 1 min (median [IQR]) | 2 (1-4) | 1 (0-3) | 2 (1-4) | |||
| At 5 min (median [IQR]) | 4 (2-6) | 2 (0-4) | 4 (3-6) | |||
| At 5 min < 5 | 58 48/83 | 76 (16/21) | 52 (32/62) | |||
| At 5 min < 7 | 82 68/83 | 90 (19/21) | 79 (49/62) | |||
| At 10 min < 5 | 29 (24/83) | 67 (14/21) | 16 (10/62) | |||
| Cord blood gases (n = 64) | ||||||
| (any of pH ≤ 7.0, BE ≤ –12, lactate ≥ 6) | 84 (54/64) | 93 (14/15) | 82 (40/49) | |||
Contributory factors were identified in 84% of cases of unexpected NE associated with hypoxia at birth and did not vary by survival status ( P = .8) ( Table 1 ). There were personnel factors present in 76% of cases, organization/management in 37%, and barriers to access or engagement with care in 24% and these proportions did not vary by survival status ( P = .7, P = 1.0, and P = 1.0, respectively).
In 28 cases (34%), contributory factors were identified from 2 subcategories and in 8 cases (10%) from all 3 subcategories. Failure to offer or follow recommended best practice and failure to recognize the complexity or seriousness of the condition were the most common personnel factors and were each found in more than 50% of cases. Lack of skills of staff and delay in emergency response were also common, found in 34% and 27% of cases, respectively. No subcategory of organization, management, or barrier to access or engagement with care factors was present in more than 10% of cases reviewed ( Table 1 ).
Fifty-five percent of reviewed cases were assessed as potentially avoidable ( Table 3 ). Overall, 52% of cases were assessed as potentially avoidable because of personnel issues. The rate of potentially avoidable death/morbidity did not vary significantly by whether babies died or survived.
