In total, 245 cases were reported to the German pediatric surveillance (ESPED) during the study period. 21 cases were false or double reports, a simultaneous report for AIS and CSVT was given in one case and four questionnaires were not returned. 22 cases were diagnosed using ultrasound and 2 cases were identified by CT. These files were excluded from further analysis in order to include only cases validated by MRI. 51 cases with CSVT and 144 with AIS remained for further analysis. Based on these data, the relative risk of AIS was 2.8 [95% CI 2.1; 3.9] times higher compared to CSVT.
Prior to MRI, cerebral ultrasound was performed in 188 (96.4%) cases. In 60% of cases, AIS or CSVT has been recognized on initial ultrasound.
Among 144 AIS cases only one brain area was affected in 36 (25%) patients. The most frequent affected areas were parietal, temporal, frontal and occipital lobe. In 73 cases more than one area was affected. In 35 cases physicians did not specify the affected brain area, but indicated whether the infarction was right, left or bilateral. 75 AIS cases were depicted on the left side while 43 cases were identified on the right side and n = 19 cases were affected on both sides (n = 7, side not specified).
Among 51 CSVT cases, a single sinus was affected in 23 cases (45%). Venous thrombosis was detected in 7 out of 23 cases in the transvers sinus and in 16 of 23 cases in the superior sagittal sinus. In 24 cases (47%) multiple sinuses were affected, predominantly in the combination of superior sagittal and transverse sinus.
Risk factors and case characteristics of AIS and CSVT patients
No differences in anthropometric data and perinatal characteristics were detected (Table 1). Both entities were more common in boys (AIS 96/144, 66.7%; p < 0.0001; CSVT 34/51, 66.7%; p = 0.02). The ratio of preterm (AIS 13.2%; CSVT 29.2%) infants exceeded the average premature birth rate in Germany (∼8%) [10]. However, CSVT was identified more often in premature infants compared to AIS cases (p = 0.02) although gestational age was not significantly different between both groups (Table 1).
Table 1. Characteristics of AIS and CSVT patients (n=number of questionnaires with available information)
|
AIS
|
|
CSVT
|
p-value
|
n
|
|
|
n
|
|
General infant characteristics
|
Sex
male
female
Gestational age [weeks+days]
premature infants ( < 37+0)
gestational age preterm infants
[weeks+days]
Multiples
Caucasian ethnicity
Stroke, thrombosis, cardiovascular events or other stroke related events in family history
|
144
140
140
19
143
140
144
|
96 (66.7)
48 (33.3)
39+3 (35+4; 40+2)
19 (13.2)
34+5 (33+3; 35+4)
8 (5.6)
135 (96.4)
20 (13.9)
|
|
51
48
48
14
50
51
51
|
34 (66.7)
17 (33.3)
39+1 (35+4; 40+3)
14 (29.2)
33+0 (30+6; 34+5)
3 (6.0)
49 (96.1)
11 (21.6)
|
1.00
0.21
0.02
0.09
1.00
1.00
0.20
|
Anthropometric infantile data
|
|
Head circumference at birth [in cm]
Length at birth [in cm]
Birth weight [in Gramm]
SGA (birth weight < 10P)
LGA (birth weight > 90P)
|
139
142
143
138
138
|
34 (33; 35.5)
51 (48; 53)
3240 (2760; 3690)
25 (18.1)
15(10.9)
|
|
43
46
48
47
47
|
34 (32; 35)
50 (47; 53)
3200 (2305; 3690)
6 (12.8)
5 (10.6)
|
0.17
0.15
0.43
0.40
0.96
|
Peripartum characteristics
|
|
Delivery mode
spontaneous delivery
caesarian section
vaginal-operative delivery
Umbilical artery pH ≤ 7.1
5-minute-Apgar score < 7
Perinatal asphyxia
Intubation/mask ventilation during
initial care
Hypothermia treatment
|
142
141
141
144
144
144
|
53 (37.3)
71 (50.0)
18 (12.7)
19 (13.5)
16 (11.4)
19 (13.2)
30 (20.8)
6 (4.2)
|
|
44
43
43
51
51
51
|
13 (29.6)
26 (59.1)
5 (11.4)
13 (30.2)
12 (27.9)
13 (25.5)
21 (41.2)
7 (13.7)
|
0.56
0.01
0.01
0.04
0.004
0.02
|
Maternal factors
|
Maternal age [in years]
Age ≤ 18 or age ≥ 35
Primiparity
|
138
138
144
|
30 (27; 34)
32 (23.2)
80 (56.0)
|
|
46
46
51
|
33 (29; 35)
18 (39.1)
28 (54.9)
|
0.01
0.04
0.93
|
|
|
|
|
|
|
|
|
|
|
|
quantitative variables are expressed as median (IQR). Categorical variables are expressed as n (%). P-values are obtained from chi-square or fisher exact tests for categorical data and from Mann–Whitney U test for continuous variables, significant p-Values after correction for multiple testing with Benjamini-Hochberg Procedure are printed in bold
Clinical signs representing neonatal asphyxia (umbilical artery pH ≤ 7.1, Apgar score < 7 at five minutes or need for intubation/mask ventilation following birth) were more frequent in infants suffering from CSVT compared to AIS cases. In addition, the proportion of infants treated with therapeutic hypothermia was significantly higher in CSVT cases (Table 1).
In our study population mothers of infants with AIS were younger and more likely to be either ≤ 18 years or ≥ 35 years. There were no significant differences in further maternal characteristics (primiparity, rate of abortion), potential risk factors related to pregnancy (chorioamnionitis, gestational diabetes, hypertensive pregnancy disorder or oligohydramnios) or maternal behavior (smoking or alcohol consumption during pregnancy) (Table A1, Additional file).
All associations remained significant after p-value adjustment for multiple comparison with the Benjamini-Hochberg procedure except for perinatal asphyxia (corrected p-value = 0.05) (Benjamini-Hochberg p-values not reported).
The most common symptom in both groups was clinical seizure activity (AIS 109/144, 76%; CSVT 35/51, 69%, Table 2). The type of seizure however, differed between CSVT and AIS cases with focal seizures being more prevalent in newborns with AIS (p < 0.0001). This was followed by apneic spells (47/144, 33%) in the AIS group and insufficient drinking/poor suck (16/51, 31%) in infants suffering from CSVT. Lethargy, as a clinical sign of encephalopathy was also more common in cases with CSVT (p < 0.0001).
Table 2. Frequency of symptoms
|
AIS (n=144)
|
CSVT (n=51)
|
p-value
|
|
Seizures
focal
generalised
not specified
|
109 (75.7)
66 (60.6)*
11 (10.1)*
32 (29.4)*
|
35 (68.6)
7 (20.0)*
18 (51.4)*
10 (28.6)*
|
0.32
<0.0001
|
|
Hypoglycaemia
|
8 (5.6)
|
6 (11.8)
|
0.14
|
|
Electrolyte imbalances
|
2 (1.4)
|
3 (5.9)
|
0.11
|
|
Suspicious limb movement
|
31 (22.2)
|
11 (21.6)
|
0.92
|
|
Unspecific symptoms
|
76 (52.8)
|
34 (66.7)
|
0.09
|
|
respiratory dysfunction
|
21 (14.6)
|
11 (21.6)
|
0.25
|
|
apnoea
|
47 (32.6)
|
11 (21.6)
|
0.14
|
|
insufficient drinking/ poor suckling
|
26 (18.1)
|
16 (31.4)
|
0.05
|
|
muscular hypotonic
|
24 (16.7)
|
15 (29.4)
|
0.05
|
|
symptoms suggesting septicaemia
|
13 (9.0)
|
8 (15.7)
|
0.19
|
|
lethargy
|
6 (4.9)
|
12 (23.5)
|
<0.0001
|
|
results are given as n (%), sum of percentage >100 because of multiple symptoms per patient
p-values are obtained from chi-square or fisher exact tests, significant p-Values after correction for multiple testing with Benjamini-Hochberg Procedure are printed in bold, *percentage related to total number of seizures
|
Time to onset of symptoms and diagnosis
Information on date of onset of symptoms and date of diagnose were available for n = 37 CSVT and n = 121 AIS cases. Median age at onset of symptoms in infants with AIS was 1 day after birth (IQR 0–2 / range 0–13 days). The median age at onset of symptoms was 2 days in the CSVT group and delayed compared to AIS cases (IQR 0–10 / range 0–17 days; Mann–Whitney U test p = 0.0002; log rank test p < 0.0001). Accordingly, CSVT cases were diagnosed later in life with a median of 10 days (IQR 4–14 / range 0–22) whereas AIS was identified by cerebral imaging with a median of 3 days of life (IQR 1–5 / range 0–19) (Mann–Whitney U test and log rank test p < 0.0001; Fig. 1) irrespectively of frequency and type of symptoms.