During the QI-project, 905 live infants were born in period 1, compared to 864 in period 2. Of these, 93 were singletons with GA ≥ 36 completed weeks and subject to assisted vaginal delivery in period 1, versus 119 in period 2. A flow chart of inclusion is shown in Figure 2. A closer investigation of the data revealed 17 protocol violations (assigned protocol not applied). These cases were excluded from statistical analysis.
A total of 195 cases were eligible for statistical analyses in periods 1 (n=92) and 2 (n=103). The characteristics of these are shown in Table 1.
Table 1 Background characteristics of mothers and infants
|
Period 1 (n=92)
|
Period 2 (n=103)
|
P-value
|
Nulliparous
|
66 (71.7)
|
74 (71.8)
|
0.99a
|
Gestational age (weeks)
|
40.1 ± 1.3
|
39.9 ± 1.3
|
0.21b
|
Birthweight (g)
|
3676 ± 570
|
3508 ± 437
|
<0.01c
|
Delivery mode
Ventouse
Forceps
Breech (Lövset’s manoeuvre)
|
85 (92.4)
2 (2.2)
5 (5.4)
|
94 (91.3)
2 (1.9)
7 (6.8)
|
0.77a
1,00b
0.69a
|
Indications, assisted delivery
Suspect asphyxiae
Maternalf
Breech presentation
|
50 (54.3)
37 (40.2)
5 (5.4)
|
61 (59.2)
36 (35.0)
7 (6.8)
|
0.49a
0.45a
0.69a
|
Cord blood gases, artery
Arterial pH
Arterial PCO₂ (kPa)
Arterial base deficit (mmol/l)
Cord blood gases, vein
Venous pH
Venous PCO₂ (kPa)
Venous base deficit (mmol/l)
|
73 (79.3)
7.20 ± 0.1
7.34 ± 1.4
5.93 ± 2.9
82 (89.1)
7.32 ± 0.1
5.08 ± 1.0
5.72 ± 2.5
|
83 (80.6)
7.20 ± 0.1
7.30 ± 1.3
5.84 ± 2.9
98 (95.1)
7.33 ± 0.1
5.05 ± 1.0
5.47 ± 2.5
|
0.81c
0.86c
0.84c
0.64c
0.80c
0.53c
|
Metabolic acidosisg
|
0 (0)
|
1 (0.9)
|
1.00d
|
Apgar scores
1 minute
5 minutes
10 minutes
|
8 (6.25 – 9)
9 (9 – 10)
10 (9 – 10)
|
8 (7 - 9)
9 (9 - 10)
10 (9 – 10)
|
0.89b
1.00b
0.73b
|
The results are reported as frequency (percentage), mean ± standard deviation (SD), median (inter-quartile range (IQR))
aCalculated using Pearson’s Chi-square Test
bCalculated using Independent-Samples Mann-Whitney U Test
cCalculated using Independent samples T-test, equal variances assumed.
dCalculated using Fisher’s Exact Test
eBased on information from continuous foetal monitoring (CTG alone or CTG with ST-analysis)
fProlonged 2nd stage, maternal exhaustion, hypertension / pre-eclampsia
gDefined as arterial pH < 7.00 and arterial base deficit ≥ 12 mmol/liter
There were no significant differences in most background variables, except from mean birthweight. At this institution, the preferred method for instrumentation in assisted vaginal deliveries is ventouse. None of the breech deliveries needed instrumentation by forceps. Cord blood analyses were not complete or successful for all cases.
After implementing the new protocol, there was a significant improvement in all cord clamping variables (Table 2). Mean cord clamping time increased by 43%. The main indication for ECC was the need to move infants to a resuscitation table for further assessment and care. This indication did not apply in period 2.
Table 2 Comparison of cord clamping variables and indications for ECC
|
Period 1 (n=92)
|
Period 2 (n=103)
|
P-value
|
Cord clamping time in seconds
< 60 seconds (ECC)
60-179 seconds (IMCC)
≥180 seconds (DCC)
|
320 ± 243
12 (13.0)
12 (13.0)
68 (73.9)
|
457 ± 375
1 (1.0)
3 (2.9)
99 (96.1)
|
< 0.001a
< 0.01b
< 0.01b
< 0.001b
|
Indications for ECC
None / tradition
Infant moved to resuscitation tabled
Cord problems
Maternal complications
|
(n=12)
1 (1.1)
11 (12.0)
0 (0)
0 (0)
|
(n=1)
0 (0)
0 (0)
1 (1.0)
0 (0)
|
0.47c
< 0.001c
1.00 c
|
The results are reported as mean ± standard deviation (SD), frequency (percentage)
aCalculated using Independent samples T-test, equal variances not assumed
bCalculated using Pearson’s Chi-square Test cCalculated using Fisher’s Exact test
dtraditional resuscitation table in period 1, LifeStart in period 2
The only case of ECC in period 2 was due to cord snapping when positioning the infant for transitional support. Ventilation support was given on LifeStart™. The infant was transferred to NICU due to respiration problems and Apgar scores 5-8-8, and the case was reported as an adverse event. The other adverse event was a case of converting care from LifeStart™ to traditional resuscitation table because of a tight cord knot. The cord was clamped at 60 seconds to move the infant for ventilation support, despite 1 minute Apgar score of 9. The cases were investigated in order to avoid future cases, and information on proper care strategies were provided to relevant personnel.
Variation in prevalence of ECC during the QI-project
A Run Chart visualises the variation in prevalence of ECC for periods 1 and 2 (Figure 3). The blue line reflects proportions of ECC from baseline measurement, through different QI activities to prepare for implementation of the new protocol, to completed implementation.
During simulation training, the prevalence dropped to 8.3% and further to zero, already before implementing the new protocol in November 2018. After protocol revision in December 2018, there were no cases of ECC registered. This fulfils the Run chart requirements of a “shift”, indicating that the improvement was not the result of chance.
Primary outcome
Multivariate logistic regression was performed to assess the overall impact of the new delivery room protocol on the primary outcome (prevalence of ECC) when adjusted for possible confounders and mediators (Table 3). The proportion of ECC was reduced by 94% in period 2, compared with period 1 (OR=0.06, 95% CI 0.01 – 0.49, p < 0.01). The only covariate reaching statistical significance was 1-minute Apgar score ≤ 5 (p<0.001) When adjusted for, this did not alter the impact of the new delivery room protocol.
Table 3 Logistic regression analyses of overall impact of the new delivery room protocol on the prevalence of ECC, adjusted for possible confounding factors
Covariate
|
Unadjusted OR
|
95% CI
|
Adjusted OR
|
95% CI
|
P-value
|
Old protocol
|
Ref.
|
|
|
|
|
New protocol
|
0.07
|
0.01 – 0.51
|
0.06
|
0.01 – 0.49
|
< 0.01
|
Nulliparous
|
0.88
|
0.26 – 2.97
|
1.54
|
0.29 – 8.19
|
0.62
|
Gestational age
|
1.19
|
0.75 – 1.87
|
1.08
|
0.56 – 2.09
|
0.81
|
Birthweight (g)
|
1.00*
|
1.00 - 1.00
|
1.000
|
1.00 - 1.00
|
0.76
|
Suspect asphyxia
|
0.88
|
0.28 – 2.71
|
0.84
|
0.21 – 3.37
|
0.80
|
1-min Apgar ≤ 5
|
13.73
|
4.08 – 46.22
|
17.61
|
4.46 – 69.50
|
< 0.001
|
Analysed for all assisted vaginal deliveries in periods 1 and 2 (N=195)
*differences only visible with 3 decimals.
OR, odds ratio CI, confidence interval
After controlling for infants placed directly on mother’s chest after delivery, multivariate logistic regression was performed to assess the direct impact of the new, mobile resuscitation table (Table 4). OR’s were adjusted for the same confounding factors. The likelihood of ECC for infants needing help on a resuscitation table was reduced by 98% when using LifeStart™, compared to a traditional resuscitation table (OR=0.02, 95% CI 0.00 – 0.16, p <0.001). No covariates reached statistical significance.
Table 4 Logistic regression analyses of the direct impact of the new, mobile resuscitation table on the prevalence of ECC
Covariate
|
Unadjusted OR
|
95% CI
|
Adjusted OR
|
95% CI
|
P-value
|
Old protocol
|
Ref.
|
|
|
|
|
New protocol
|
0.02
|
0.00 – 0.13
|
0.02
|
0.00 – 0.16
|
< 0.001
|
Nulliparous
|
0.84
|
0.23 – 3.05
|
0.44
|
0.06 – 3.30
|
0.42
|
Gestational age
|
1.22
|
0.74 – 2.01
|
0.95
|
0.46 – 1.96
|
0.89
|
Birthweight (g)
|
1.001*
|
1.000 – 1.002
|
1.001*
|
0.998 – 1.003
|
0.57
|
Suspect asphyxia
|
0.73
|
0.22 – 2.41
|
0.63
|
0.12 – 3.25
|
0.58
|
1-min Apgar ≤ 5
|
5.40
|
1.55 – 18.83
|
2.19
|
0.42 – 11.39
|
0.35
|
Analysed for all infants placed on a resuscitation table in periods 1 and 2 (N=83)
*differences only visible with 3 decimals.
OR, odds ratio CI, confidence interval
Secondary outcomes
All infants were initially stimulated by drying and rubbing their back with preheated towels. Heat-loss prevention was also provided for all cases; in period 1 by an overhead warmer; in period 2 by a heating mattress. 20% of the infants in period 2 had pulse-oximetry attached, compared to zero in period 1. ECG-monitors were not available throughout the project. No infants needed full resuscitation in either period. Univariate analysis showed that infants in period 2 were significantly less likely to be placed on mothers’ chest (within the first minute of life), despite better 1-minute Apgar scores. However, they were less likely to receive ventilation support or to be transferred to NICU (Table 5).
Table 5 Comparison of immediate transitional care and secondary outcomes for infants placed on a resuscitation table* in periods 1 and 2
|
Period 1 (n=92)
|
Period 2 (n=103)
|
P-values
|
Infants placed on mothers chest
|
69 (75.0)
|
43 (41.7)
|
< 0.001a
|
Infants placed on resuscitation table
|
23 (25.0)
|
60 (58.3)
|
< 0.001a
|
Short-time outcomes*
Apgar 1 < 5
Apgar 5 < 7
Apgar 10 < 9
Ventilation support
Transfer to NICU
|
(n=23)
12 (52.0)
4 (17.4)
9 (39.0)
15 (65.2)
6 (26.1)
|
(n=60)
12 (20.0)
3 (5.0)
9 (15.0)
33 (55.0)
7 (11.7)
|
< 0.01a
0.09b
0.05a
0.40a
0.17a
|
The results are reported as frequency (percentage)
aCalculated using Pearson’s Chi-square Test bCalculated using Fisher’s Exact Test
*traditional resuscitation table in period 1, LifeStart in period 2
Infants in period 2 were less likely to have 5-minute Apgar < 7 or 10-minute Apgar<9, but after adjusting for possible confounding factors (parity, GA, birthweight, suspect asphyxia and low 1-minute Apgar score) by multivariate logistic regression analysis, no significant difference could be demonstrated for any of the short-time outcomes (Table 6). Pre-existing indications (maternal diabetes) and late-onset complications (infection, heart conditions, drugs) resulting in subsequent interventions on a resuscitation table or transfer to NICU were excluded from analysis, since this did not interfere with cord clamping practice
Table 6 Logistic regression analyses of immediate care and short-time outcomes for infants placed on a resuscitation table* in periods 1 and 2
|
Unadjusted estimate
|
Adjusted estimate
|
OR
|
95 % CI
|
P-value
|
OR
|
95 % CI
|
P-value
|
Old protocol
Ventilation support
|
Ref.
0.65
|
0.24 - 1.77
|
0.40
|
1.22
|
0.38 - 3.97
|
0.74
|
Low Apgar scores
Apgar 5 < 7
Apgar 10 < 9
|
0.25
0.27
|
0.05 - 1.22
0.09 - 0.82
|
0.08
0.02
|
0.79
0.57
|
0.10 - 6.37
0.15 - 2.17
|
0.82
0.41
|
Transfer to NICU **
|
0.37
|
0.11 - 1.27
|
0.11
|
0.67
|
0.16 - 2.76
|
0.58
|
Adjusted for parity, gestational age, birthweight, suspect asphyxia and 1-minute Apgar score ≤ 5
*traditional resuscitation table in period 1 (n=23), LifeStart in period 2 (n=60)
**due to prolonged ventilation support or unsuspected compromised infant at birth
OR, odds ratio CI, confidence interval