Introduction
Until now obligatory quality assurance measure, according to § 136ff Social Code Book
V, was suspended for treatment of patients with congenital heart disease (CHD) for
various reasons. Quality of care has a major impact on patients' long-term outcome
including quality of life, especially in CHD. Therefore, since 2012 the German Society
for Thoracic and Cardiovascular Surgery (DGTHG) and German Society for Pediatric Cardiology
and Congenital Heart Defects (DGPK) as the scientific societies execute a multicenter
registry study concerning interventional and surgical therapies in patients with CHD.
The structure of the database has been described in detail previously.[1 ] The voluntary German Registry for Quality Assurance in CHD (Nationale Qualitätssicherung
Angeborene Herzfehler) has been approved by the Charité's Ethics Committee (code number:
EA2/011/11). The structure of the registry, data acquisition, and evaluation is in
accordance with the guidelines of “Good Epidemiological Practice,”[2 ] “Good Hospital Practice,”[3 ] and the Declaration of Helsinki for medical research involving human subjects.[4 ] The registry contains detailed information on diagnoses and procedures, which enable
detailed risk categorization[5 ]
[6 ] in nearly 90% of cases. Severity of adverse events is assessed in a standardized
manner according to patient's outcome.[5 ]
[7 ] As each patient is distinctly identifiable by his own life-long lasting pseudonymization
number, analyses of the data can refer to patients, cases (hospital stays), or procedures.
Each year, the responsible scientific societies write an annual report based on two
types of analyses: the national report summarizes aggregated and anonymized results
of all participating institutions, compared with the previous year. In addition, separate
institution-related reports contain the analyses of each participating heart center
in comparison with the national results. Due to contractual arrangements, the institutional
report is confidential and not intended for public access. Upon request, each institution
can receive an electronic copy of its evaluated data for further use (e.g., transfer
to the European Congenital Heart Surgeons Association database: https://echsacongenitaldb.org ). The purpose of this report is to provide a public comprehensive annual update of
the activities and outcomes from the German National Report on CHD 2021.
Methods
Voluntary online data submission into the database with the goal of 100% coverage
as previously described.[1 ] All patients/parents gave written consent. All data sets were monitored with respect
to data integrity before case closure. Source data were not monitored. The annual
report includes only completed cases. Coding of diagnoses and procedures is based
on the International Pediatric and Congenital Cardiac Code (IPCCC).[8 ] Completeness of recorded procedures was estimated by comparison with the German
Heart Surgery Report 2020[9 ] and the online available Deutscher Herzbericht 2021 (www.herzstiftung.de ).
Risk Stratification
All cases intended as single interventional treatment were assigned to risk categories
1 to 4 according to the catheterization for Congenital Heart Disease Adjustment for
Risk Method.[5 ] All cases intended as surgical treatment were assigned to risk categories 1 to 5
according to the Society of Thoracic Surgeons (STS)-European Association for Cardio-Thoracic
Surgery (EACTS) mortality categories.[6 ] Details are provided in the Supplementary Material. Risk stratification of cases
with multiple procedures and those starting with a hybrid procedure was not possible
due to the lack of established risk categories for these procedures.
Key Performance Indicators
Length of hospital stay, the need and length of intensive care treatment, the need
and length of mechanical ventilation, procedure time, and the requirement of blood
transfusion are general key performance indicators presented in this report. In addition,
specific key performance indicators were analyzed: use of fluoroscopy, fluoroscopy
times, data from cardiopulmonary bypass (CPB), and the use of near-infrared spectroscopy.
Key Quality Indicators
In the German National Report mortality in-hospital, 30, and 90 days after first procedure
is analyzed. Adverse events are recorded based on the IPCCC nomenclature. In addition,
the number of cases without adverse events is analyzed. Severity of adverse events
was categorized according to patient's outcome by the supplying physician and according
to the definitions of adverse event severity for congenital cardiac catheterization[5 ] or the definition of major complications of the STS Congenital Heart Surgery Database
(STSCHSD) database.[7 ] Unplanned redo procedures and patient's death trigger automatically notification
of an adverse event with severity grades 4 or 5 (“major” or “catastrophic”). Observed
mortality and morbidity rates are compared with published data if available.[5 ]
[6 ]
Definitions for Adverse Event Severity in Cases with Interventions[5 ]
None: no harm, no change in condition, and may have required monitoring to assess
for potential change in condition with no intervention indicated.
Minor: transient change in condition, not life-threatening, condition returns to baseline,
required monitoring, required minor intervention such as holding a medication, or
obtaining laboratory test.
Moderate: transient change in condition may be life-threatening if not treated, condition
returns to baseline, required monitoring, required intervention such as reversal agent,
additional medication, transfer to the intensive care unit (ICU) for monitoring, or
moderate transcatheter intervention to correct condition.
Major: change in condition, life-threatening if not treated, change in condition may
be permanent, may have required an ICU admission or emergent readmit to hospital,
may have required invasive monitoring, and required interventions such as electrical
cardioversion or unanticipated intubation, or required major invasive procedures or
transcatheter interventions to correct condition.
Catastrophic: any death, and emergent surgery, or heart–lung bypass support (extracorporeal
membrane oxygenation [ECMO]) to prevent death with failure to wean from bypass support.
Definition of Major Complications in Cases with Operations
Major complications are defined according to complication codes in the STS Congenital
Heart Surgery Database (STSCHSD) Data collection form, Version 2.50 as cited in Jacobs
et al[7 ]:
Postoperative acute renal failure requiring temporary or permanent dialysis.
Postoperative neurologic deficit persisting at discharge.
Postoperative atrioventricular block requiring permanent pacemaker.
Postoperative mechanical circulatory support.
Phrenic nerve injury/paralyzed diaphragm.
Unplanned reoperation.
Any death.
All other recorded adverse events according to the IPCCC were rated as minor (most
frequent in 2020: postprocedural pulmonary infection 15.80.21, pleural effusion requiring
drainage 15.80.61, arrhythmia requiring drug treatment 11.00.30, postprocedural chylothorax
15.80.55, sternum left open: elective [planned] 15.03.57).
Evaluation and Data Presentation
All cases are analyzed with respect to the initial treatment and after allocation
to one of four groups: surgery, intervention, multiple procedures, and hybrid procedures.
In addition to this analysis of the entire cohort, we performed detailed analyses
of 15 index procedures containing 6 defined interventional and 9 defined surgical
procedures in specific subgroups of CHD.
Longitudinal Evaluation of Patient's Medical Careers
All patients who since 2012 received surgical correction for Fallot malformation and
all patients who received surgical or interventional treatment of native coarctation
of the aorta (CoA) were analyzed for requirement of further invasive treatment.
Results
Registry Data Report 2020
The number of patients, cases (number of hospital stays), and procedures, which constitute
the 2020 data basis of the registry, are shown in [Fig. 1 ]. Taking together all surgical and interventional cases, 23 German centers treated
6,051 cases under the terms of quality assurance. Eighteen participating centers provided
detailed surgical data from 2,795 single surgery cases and 336 cases with planned
multiple procedures, altogether 3,713 surgeries on CHD. In-hospital lethality was
1.6% in all single surgery cases and 5.7% in all cases with planned multiple procedures
([Fig. 2 ]). In comparison, the German Heart Surgery Report based on anonymous self-disclosure
of 78 institutions in 2020 counts 5,637 operations on CHD. Reported lethality was
2.7% in cases with extracorporeal bypass use and 2.1% in cases without.[9 ] In this quality registry data report, 23 participating centers provided detailed
data from 3,226 interventions compared with the reported number of 5,239 interventions
by anonymous self-disclosure of 32 institutions published in the Deutscher Herzbericht
2021 (www.herzstiftung.de ). Thus, overall completeness of quality recorded procedures nationwide was 66% for
surgical and 62% for interventional procedures and larger at the participating centers.
Fig. 1 Numbers of patients, cases, and procedures. Note that 9.4% of 5,432 patients were
treated two or more times in hospital. In 9.2% of the 6,051 cases, patients received
two or more procedures, which was initially planned only in 5.6% of the cases. The
rate of surgical procedures exceeded the rate of planned single surgical cases indicating
a major role of surgery in redo procedures.
Fig. 2 Key quality indicators in different case subgroups. (A ) Key quality indicators of 2,795 cases with single interventional procedures. Adverse
events were recorded in 5.5% of the cases. The rate of major or catastrophic adverse
events was 1.5% in all cases. Severity of 6.5% of adverse events recordings (0.4%
of all cases) was not classified. (B ) Key quality indicators of 2,887 cases with single surgical procedures. Adverse events
were recorded in 32.1% of the cases. The rate of major adverse events was 9.9% in
all cases. Severity of 1.3% of adverse events recordings (0.4% of all cases) was not
classified. (C ) Key quality indicators of 336 cases with multiple procedures. Adverse events were
recorded in 57.1% of the cases. In this case group subsequent procedures were present
in 100% by definition and were not conditioned by complications. Note: different x -axis scales. The definitions of major and minor adverse events for surgical and interventional
cases differ fundamentally and are not suitable for direct comparison.
Age and Gender Distribution
The majority of treatments were performed in children and adolescents ([Table 1 ]). Note that 10.2% of the cases were treated in newborns and 13.6% in adults. Males
were more affected by treatment for CHD than females.
Table 1
Age and gender distribution of all cases
Numbers
Percentage
Newborn
618
10.2
Infants (30 d-year)
1,532
25.3
Children and adolescents
3,077
50.9
Adults
824
13.6
Males
3,237
53.5
Females
2,814
46.5
Note: Assignment of patient's age was done at the date of the first procedure of every
case.
Case Complexity
This report for the sake of space disclaims the listing of cardiac and noncardiac
codiagnoses as well concomitant procedures. The numbers of elective redo cases with
previous single or multiple cardiac procedures are shown in [Fig. 3 ]. In 2020, first treatment for CHD in Germany was performed in 48% of interventional,
57% of surgical, and 62.5% of cases with multiple procedures.
Fig. 3 Pretreatment of patients in different case subgroups. (A ) Cases intended for single interventional treatment. (B ) Cases intended for single surgical treatment. (C ) Cases with multiple procedures. The rate and the distribution of different pretreatments
is expressed in the bar charts.
Key Performance Indicators
Details are listed in [Table 2A ] for all cases and for those with defined interventional ([Table 2B ]) or surgical ([Table 2C ]) index procedures. Interventional cases required the least resources. Hospital stay
was 2 days in median while the 75 percentile lay at 3 days. Intensive care treatment
was applied in 13.5% of the cases and most cases (83%) were performed under analgosedation
without mechanical ventilation. Note that 6.8% of the cases were performed without
fluoroscopy (mainly Rashkind procedures in dextro-transposition of the great arteries
and atrial septal defect [ASD] closures, see index procedures). Analysis of interventional
index procedures revealed significant differences among the different subgroups of
these patients. Most frequent intervention were ASD closure with 493 cases. For this
procedure, the rate of general anesthesia was 29%, thereby above average. Rare and
complex interventions like ventricular septal defect (VSD) closure and native CoA
treatment required longer in-hospital stay and in cases with native CoA the largest
numbers of ICU stays (29%) and intubations (20%). Beside the large spectrum of age
in this subgroup 25% of the patients were newborns or small infants younger than 4
months of age. Only half of the interventional units reported these complex procedures.
Percutaneous pulmonary valve implantation is the most laborious interventional index
procedure, which is expressed by longest procedure and fluoroscopy times. The median
procedure time of 76 minutes and the median fluoroscopy time of 10.1 minutes in the
analysis of all interventional cases was slightly higher compared with those other
interventional index procedures (compare [Table 2A ] and [B ]).
Tables 2
General and specific key performance indicators
(A) All cases
Number of procedures
Number of centers
Length of hospital stay (d)
ICU
Length of ICU stay (d)
Intubation
Mechanical ventilation (h)
Blood transfusion
Procedure time (min)
Fluoroscopy
Fluoroscopy time (min)
CPB
Perfusion time (min)
Cross-clamp
Cross-clamp time (min)
NIRS
Interventional cases
2,795
22
2.0 (2.0/3.0)
13.5%
1.0 (1.0/5.0)
17%
3.0 (2.0/4.0)
0.7%
76 (49–115)
93.2%
10.1 (5.5/18.4)
Surgical cases
2,887
19
11.0 (8.0–18)
94.4%
4.0 (2.0–8.0)
98.1%
9.0 (6.0–42)
57.2%
202 (133–286)
76.7%
115 (76–168)
61.7%
69 (43–105)
81.1%
Multiple procedures
336
17
28 (17.5–54)
96.1%
15 (8–27)
73.8%
59 (10–190)
67.7%
61.5 (39.5–99.5)
86.7%
11.0 (6.2–18)
65%
147.5 (91–220)
46.5%
92 (57–134)
76.1%
(B) Cases with interventional index procedures
Age (y)
Number of procedures
Number of centers
Length of hospital stay (d)
ICU
Length of ICU stay (d)
Intubation
Mechanical ventilation (h)
Blood transfusion
Procedure time (min)
Fluoroscopy
Fluoroscopy time (min)
ASD
9.2 (5.5/37.9)
493
22
2.0 (2.0–3.0)
5.1%
1.0 (1.0–1.0)
29%
2.0 (2.0–2.0)
0.2%
47 (32–69)
77.7%
5.0 (2.3–8.0)
VSD
5.6 (3.8/7.9)
29
9
3.0 (2.0–3.0)
3.4%
13.8%
2.0 (2.0–2.0)
3.4%
85 (69–97)
100%
12.4 (9.5–17.6)
Native CoA
5.2 (0.3/16.7)
45
12
3.0 (2.0–5.0)
28.9%
5.0 (5.0–8.0)
20%
3.0 (3.0–20)
0%
64 (48–117)
100%
8.6 (5.0–12.4%)
Recurrent CoA
7.5 (0.8/14.8)
30
13
2.0 (2.0–3.0)
6.7%
0
0
0
60.5 (40.0–95.0)
100%
6.5 (4.0–11.5)
PDA
2.9 (0.9/5.1)
281
22
2.0 (2.0–3.0)
9.3%
3.0 (1.0–9.0)
9.3%
3.0 (2.0–18)
0.7%
53 (40–75)
100%
6.7 (4.6–10)
PPVI
18.3 (12.2/29.2)
88
16
5.0 (4.0–6.0)
14.8%
1.0 (1.0–1.0)
54.5%
4.0 (3.0–4.0)
1.1%
155 (115–209)
100%
22.4 (14.9–36.6)
(C) Cases with surgical index procedures
Age (y)
Number of procedures
Number of centers
Length of hospital stay (d)
ICU
Length of ICU stay (d)
Intubation
Mechanical ventilation (h)
Blood transfusion
Procedure time (min)
CPB
Perfusion time (min)
Cross-clamp
Cross-clamp time (min)
NIRS
ASD
4.0 (2.0/6.7)
180
18
8.0 (6.0–9.0)
100%
2.0 (2.0–3.5)
100%
6.0 (4.0–8.0)
36.7%
153 (119–180)
100%
53 (41.5–65)
77.8%
24 (20–31)
85.6%
VSD
0.4 (0.3/0.7)
261
18
10 (8–13)
100%
4.0 (3.0–6.0)
100%
12.0 (6.0–54)
82%
175 (150–221)
100%
93 (71–113)
99.6%
58.5 (43–73.5)
94.3%
AVSD
0.5 (0.4/1.3)
205
17
12 (9.0–18)
100%
5.0 (3.0–9.0)
100%
24 (7.0–83)
76.1%
236 (187–288)
100%
140 (105–186)
97.1%
100 (71–131)
96.1%
Native CoA
0.1 (0.0/0.3)
119
17
10.0 (7.0–15)
100%
4.0 (3.0–7.0)
100%
11.0 (5.0–48)
38.7%
121 (87–162)
26.9%
114 (50–136)
19.3%
39 (25–56)
79.8%
SimpleTGA
0.0 (0.0/0.0)
71
17
21 (17–28)
100%
13.0 (6.0–17)
100%
73 (24–120)
93
287 (244–339)
100%
173 (150–222)
100%
112 (93–134)
97.2%
Fallot
0.5 (0.4/0.8)
143
18
13 (9.0–18)
100%
5.0 (3.0–8.0)
100%
22 (7.0–52)
90.2
254 (204–305)
100%
155 (114–192)
100%
94 (72–124)
96.5%
TCPC
3.8 (3.2/4.6)
157
16
17 (12–24)
100%
5.0 (3.0–9.0)
100%
8.0 (6.0–10.0)
60.5%
253 (185–332)
99.4%
93.5 (68.5–136)
12.7%
51.5 (19–84)
91.1%
Norwood I (< 90 d)
0.0 (0.0/0.0)
75
15
47 (29–70)
100%
21 (11–33)
100%
130 (72–199)
96%
315 (263–414)
100%
182 (147–228)
100%
57 (47–75)
88%
Pulmonary valve replacement
16.3 (9.6/24.2)
112
17
9.0 (7.0–14)
99.1%
3.0 (2.0–5.0)
100%
8.0 (6.0–12)
57.1%
271 (221–372)
100%
111.5 (87.5–166)
38.4%
66 (48–105)
87.5%
Abbreviations: ASD, atrial septal defect; AVSD, atrioventricular septal defect; CoA,
coarctation aortae; CPB, cardiopulmonary bypass; ICU, intensive care unit; NIRS, near-infrared
spectroscopy; PDA, patent ductus arteriosus; PPVI, percutaneous pulmonary valve implantation;
TCPC, total cavopulmonary connection; TGA, transposition of great arteries; VSD, ventricular
septal defect.
Data are given as numbers, percentages or median (25. – 75. percentiles).
Surgical cases required more resources. Intensive care treatment in 94.4% of the cases
with a median length of stay of 4 days and blood transfusions in 57.2% were required.
Again, analysis of index procedures revealed significant differences. Patients with
ASD, VSD, or native CoA were far younger than patients in the interventional subgroups.
VSD closure was the most frequent surgical procedure with 261 cases. Case performance
in VSD closure appears highly standardized with narrow interquartile ranges for perfusion
and aortic clamp times as well as for the duration of hospital stay. The complex nature
of univentricular heart treatment is visible in the information of Norwood and total
cavopulmonary connection (TCPC) cases. Main diagnosis in Norwood cases was hypoplastic
left heart syndrome (HLHS) in 81.3%. Eighteen patients (24%) received bilateral pulmonary
arterial banding before the Norwood procedure. Norwood patients had longest perfusion
times (median 182 minutes), longest mechanical ventilation (median 130 hours), longest
ICU stay (median 21 days), and longest in-hospital stay (median 47 days). Three out
of 18 pediatric congenital cardiac surgery units did not perform Norwood operations
in 2020. HLHS was also the most frequent main diagnosis in TCPC cases (32.5%). TCPC
cases showed short ventilation times with small interquartile range (median 8 hours),
but an elongated and case variable lengths of hospital stay (median 17 days, 12–24
days, 25 and 75 percentile, respectively).
Key Quality Indicators
In-hospital mortality is specified in detail in [Table 3 ]. It was 1.6% in isolated surgical and 0.5% in interventional treatments. Cases with
multiple procedures carried the highest mortality of 5.7%. The 336 cases of this group
compounded 48.5% newborns, 25.9% infants, and 25.6% children, adolescents, and adults.
In 27.7% of the cases, patients received not only one but two or more subsequent procedures,
in total 528 procedures including 308 operations and 220 interventions. In the subgroup
analysis of surgical index procedures, the Norwood procedure sticks out with 18.7%
mortality. Overall in-hospital mortality rate in all 2.795 interventional cases (0.5%)
is higher compared to defined interventional index procedures (two cases of death
[0.2%] out of 966 cases).
Table 3
In-hospital mortality—all cases and index procedure cases
Numbers
Percentage
All cases
85/6,051
1.4
Interventional cases
15/2,795
0.5
Surgical cases
47/2,887
1.6
Multiple procedures
19/336
5.7
Hybrid procedures
No data
Cases with interventional index procedures
ASD
0/493
0
VSD
0/29
0
Native CoA
0/45
0
Recurrent CoA
0/30
0
PDA
2/281
0.7
PPVI
0/88
0
Cases with surgical index procedures
ASD
0/180
0
VSD
0/261
0
AVSD
2/205
1.0
Native CoA
0/119
0
Simple TGA
0/71
0
Fallot
1/143
0.7
TCPC
0/157
0
Norwood I (< 90 d)
14/75
18.7
Pulmonary valve replacement
1/112
0.9
Abbreviations: ASD, atrial septal defect; AVSD, atrioventricular septal defect; CoA,
coarctation aortae; PDA, patent ductus arteriosus; PPVI, percutaneous pulmonary valve
implantation; TCPC, total cavopulmonary connection; TGA, transposition of great arteries;
VSD, ventricular septal defect.
Note: Interventional index procedures represent 966 out of 2,795 interventions (35%)
and account for 2 out of 15 in-hospital deaths (13%). Surgical index procedures represent
1,323 out of 2,887 surgical cases (46%) and account for 18 out of 47 in-hospital deaths
(38%).
Observed mortality in all five STS-EACTS mortality categories was lower compared with
the published mortality rates from the performance data set of the STS-EACTS group
for 2002 to 2007 ([Table 4 ]).[6 ]
Table 4
Comparison of observed and expected in-hospital mortality according to STS-EACTS mortality
categories
STS-EACTS mortality category
Numbers (deaths/patients)
Observed mortality (%)
STS-EACTS mortality 2002–2007[a ] (%)
Category 1
7/1,204
0.6
0.8
Category 2
6/712
0.8
2.6
Category 3
5/396
1.3
5.0
Category 4
11/303
3.6.
9.9
Category 5
12/60
20.0
23.1
Abbreviation: STS-EACTS, Society of Thoracic Surgeons-European Association for Cardio-Thoracic
Surgery.
a Observed mortality rates for performance data set in: O'Brien et al.[6 ]
Analysis of cases without adverse events is demonstrated in the bar charts of [Fig. 2 ], focusing on the severity categorization of recorded adverse events and in the sunburst
diagram of [Fig. 4 ], focusing on the adverse event rates related to the risk categorizations and the
type of intended procedure.
Fig. 4 Case distribution and risk-related outcome for adverse events. This sunburst diagram
analyzes data from a total of 6,018 cases. It visualizes the distribution of 4,745
cases without complications (79%) and 1,273 cases with adverse events related to the
type of procedure and risk classification. Risk classes 1–5 for surgery refer to STAT
mortality categories and are not comparable to the interventional procedure-type risk
categories 1–4. A case-related risk classification is not applicable in cases with
multiple procedures. Displayed event rates in the outer circle refer to the particular
dedicated risk category. Surgical risk categories are shown in the right half of the
diagram and assorted according to their frequencies clockwise from top to down. Most
surgeries were performed in lower risk categories 1 and 2 with highest freedom of
recorded complications in risk category 1 (76%). Lowest rates of freedom from recorded
complications were 52% in risk category 4 and 18% in risk category 5 (smallest group
representing only 60 cases [2%] in the surgical group). Interventional risk categories
are shown in the left half of the diagram and assorted clockwise from down to top.
Most interventions were performed at medium procedural risk categories 2 and 3. Freedom
from recorded complications ranged from 97% in category 1 to 91% in category 4. Cases
with multiple procedures showed with a 43% freedom rate of recorded complications
a lower score compared with cases with surgical or interventional treatment. Labeling:
Inner circle: cath – interventional cases; multiple – cases with ≥ 2 procedures at
different times; surgery – surgical cases. Middle circle: r – risk category (1 lowest;
4[5 ] highest); nc – not classified. Outer circle: no – cases without recorded complications;
event – cases with recorded complications.
The German National Quality Assurance Report on CHD contains detailed age and risk-related
outcome reports on key quality indicators for all combined case groups and all 15
index procedures, in detail not considered in this overview. These detailed tables
allow each participating center detailed comparison with their own separate and confidential
institution-related report.
For interventional cases, observed rates of all adverse events are in similar range
as compared with published data ([Table 5 ]).[10 ] Adverse event rates for surgical cases could not be compared with published data
in this report because defined major events[7 ] have not been analyzed with respect to STS-EACTS risk categories (reporting was
only on all adverse events) ([Table 6 ]).
Table 5
Rate of any adverse event according to procedural risk classification in interventional
cases
Procedural risk category
Numbers (any AE/patients)
Observed rate of any AE[a ] (%)
Expected rate of any AE (%)[b ]
Category 1
7/219
3.2
5.2
Category 2
52/1,265
4.1
13
Category 3
58/686
8.5
19
Category 4
26/295
8.8
25
Abbreviations: AE, adverse event; IPCCC, International Pediatric and Congenital Cardiac
Code.
a Any adverse event (AE) as defined for pediatric and congenital interventions—not
to compare with the IPCC code list and the STSCHSD Data collection form for surgical
procedures.
b Observed adverse event rates for performance data set in: Bergersen et al.[10 ]
Table 6
Rate of any adverse event according to STS-EACTS mortality categories in surgical
cases
STS-EACTS mortality category
Numbers (any AE/patients)
Observed rate of any AE[a ] (%)
Category 1
288/1,204
23.9
Category 2
235/712
33
Category 3
170/396
42.9
Category 4
145/303
47.9
Category 5
49/60
81.7
Abbreviations: AE, adverse event; IPCCC, International Pediatric and Congenital Cardiac
Code; STS-EACTS, Society of Thoracic Surgeons-European Association for Cardio-Thoracic
Surgery.
a Any adverse event (AE) as documented from the IPCC code list. 67.7% of all adverse
events were rated “minor,” 31% were rated “major” according to complication codes
in the STSCHSD Data collection form, Version 2.50, as cited in Jacobs et al.[7 ] 1.3% of all AE were not classified.
Longitudinal Data Analysis
Tetralogy of Fallot
The database included 1,632 patients following repair of Fallot over 9 years from
2012 to 2020. Eighteen percent of the patients received palliative procedures prior
to surgical repair. Until the end of 2020 20.8% of the patient's redo procedures were
recorded. The flowchart of [Fig. 5 ] describes patient's medical career and the bar chart of [Fig. 6 ] visualizes the frequency of redo cases in relation to the time interval after Fallot
repair. The largest number of redo cases and redo procedures took place in the first
year after initial repair. The list of the five most frequent types of operations
comprises delayed closure of sternum and weaning from ECMO indicating complex postoperative
courses ([Fig. 5 ]). Patients with palliative treatment prior to repair were more often subjects of
redo procedures (40.3% vs. 16.5%, respectively).
Fig. 5 Comparison of redo procedures after repair of tetralogy of Fallot with or without
preceding palliation. In this 9-year follow-up observation in patients after repair
of Fallot, the rate of redo procedures was 40.3% compared with 16.5% after previous
palliation.
Fig. 6 Recorded proportion of patients with repeat treatments in relation to the total number
of patients with Fallot correction recorded in this period. This figure demonstrates
the relative probability to require a redo case in relation to the time after tetralogy
of Fallot (TOF) repair. Lower part in each column: patients with one redo case; above:
patients with 2, 3, … up to 5 redo procedures. After corrective surgery dated from
2012 to 2020, patients have not reached adolescent age. Note: This is no systematically
collected information. Since no data are available on the completeness of follow-up,
these figures must be interpreted as “minimum proportion.” Redo cases are counted
when entered into the registry database (upper line of the table). The scale basis
(lower line of the table) gives the maximal possible number of patients in the specific
time period (patients discharged after TOF repair). Thus, the probability may be underestimated
due to missing cases or due to overestimation of the basis population.
Coarctation
Over 9 years, from 2012 to 2020, the registry recorded 1,864 patients with primary
treatment of native coarctation. Data from 1,821 patients could be analyzed. Primary
treatment has been transcatheter intervention in 30% and surgery in 70% of the patients.
Native CoA was treated in 786 newborns (43.2%), 455 infants (25.0%), 454 children
and adolescents (24.9%), and 126 adults (6.9%). Types of procedures dispersed in 202
balloon dilations (10.8%), 353 stent implantations (18.9%), 960 operations without
CPB (51.5%), and 306 operations using CPB (16.4%). The distribution of procedures
according to age is demonstrated in [Fig. 7 ].
Fig. 7 Distribution of different coarctation of the aorta (CoA) treatment strategies to
age groups. This sunburst diagram shows the distribution of primary native CoA treatment
in 1,864 patients with respect to age and type of treatment. Frequencies were shown
with descend numbers in clockwise direction. Most patients (43.2%) were treated at
newborn age but 126 patients (6.9%) not until adulthood. Operations without cardiopulmonary
bypass (CPB) are most frequent in newborns and infants while stent procedures dominated
treatment in children and adults. Relative procedure frequencies in the outer circle
refer to the particular dedicated age group.
Until the end of 2020, redo procedures in 574 patients (30.8%) were recorded. A total
of 385 (67.1%) of the patients received 1 and 189 (32.9%) ≥ 2 subsequent procedures;
exclusively interventional treatment in 48.8%, exclusively surgery in 31.7%, and both
treatment strategies in 19.5%. The flowchart of [Fig. 8 ] describes patient's medical career and the bar chart of [Fig. 9 ] visualizes the frequency of redo procedures in relation to the time after primary
procedure in native CoA.
Fig. 8 Nine years' follow-up observation in patients with native coarctation of the aorta.
Subsequent procedures during the primary case occurred in 13.4% after balloon dilations
and in 12.7% after cardiopulmonary bypass (CPB) operations. After discharge from primary
care, redo cases with follow-up procedures were frequent in all treatment arms but
more frequent after primary interventional treatment (44.1% and 35.7% in this 9-year
observation period so far).
Fig. 9 Recorded proportion of patients with repeat treatments in relation to the total number
of patients with primary treatment of native coarctation of the aorta (CoA) recorded
in this period. This figure demonstrates the relative probability of redo cases in
relation to the time after primary care of native CoA. Lower part in each column:
patients with one redo case; above: patients with 2 or 3 redo procedures. The majority
of redo cases occurred in the first year after primary treatment and should be interpreted
in the context of 43.2% newborn and 25% infant cases. Probability of redo cases after
the first year levels off between 2 and 3%/year. Note: This is no systematically collected
information. Since no data are available on the completeness of follow-up, these figures
must be interpreted as “minimum proportion.” Redo cases were counted when entered
into the registry database (upper line of the table). The scale basis (lower line
of the table) gives the maximal possible number of patients in the specific time period
(patients discharged after primary treatment of native CoA). Thus, the probability
may be underestimated due to missing cases or due to overestimation of
Discussion
This multicenter registry was initiated by the scientific associations DGTHG and DGPK[1 ] based on the scientific awareness, that surgical and interventional treatment of
CHD represent complementary parts of common treatment concepts.[11 ]
[12 ] The data of our registry support this thesis. In 2020, one-third to one-half of
the cases received interventional or surgical treatment or even both prior to the
actual procedures ([Fig. 2 ]). In 5.6% of all cases, patients received scheduled multiple procedures (mostly
combinations of surgery and intervention). Several CHD diagnoses can be treated either
by interventional or surgical methods.[13 ]
[14 ]
[15 ] Detailed analysis of the index procedures (ASD, VSD, CoA) in the German Report exposes
the heart team access in patients with CHD. Surgical and interventional ASD, VSD,
and CoA patients obviously even by age represent a different spectrum of the diseases
([Table 2 ]). The necessity of evidence-based heart team decisions especially becomes obvious
in the long-term observations ([Figs. 5 ]
[6 ]
[7 ]
[8 ]
[9 ]). Prerepair palliation in Fallot patients was performed by various interventional
and surgical treatment options.[16 ]
[17 ] Palliated patients obviously represent the more complex spectrum of CHD with further
impact on redo cases and redo procedures after repair of Fallot—requiring a large
variety of interventional and surgical methods.[18 ] Coarctation patients present themselves at different age with other symptoms and
unlike anatomy of the aortic disease. Though the vast majority of patients underwent
primary treatment of native coarctation in newborn and infant age, 6.9% of the patients
in our report received primary treatment in adulthood.[19 ] While stent procedures dominated in older patients, surgery was the domain in newborn
and infants. However, our data show that balloon dilations and stent implantations
may represent therapeutic options in rare individual patients to start treatment of
coarctation even in newborn and infants. In the future, long-term assessment will
become increasingly more important in CHD since the quality of treatment frequently
does not become apparent within the first months after the procedure.[20 ]
[21 ]
Combined case group analysis provides a real, live description of the time and effort
and the performance of surgical and interventional treatment in Germany.[22 ] Cases planned with single interventions required the least resources. Surgical therapy
needed far more intensive care resources and transfusion medicine. Analysis of subgroups
underlines the integrity of our data sets. As expected, the rates of intensive care
treatments, mechanical ventilation, and CPB usage were 100% in some of the procedures
in contrast to the information on native CoA treatment, which was performed according
to different surgical strategies.[20 ]
[21 ] Various planned treatment strategies become visible by analyzing defined surgical
and interventional procedures. Overall, the key performance indicators ([Table 2 ]) mirror the how-to-do standard in Germany. As an example, the rate of general anesthesia
was higher in ASD interventions, which can be explained by the need for transesophageal
echocardiography guidance.[23 ] Fluoroscopy times are still an issue in interventional procedures.[24 ]
[25 ]
[26 ] Mean fluoroscopy in most of the index procedures were below average as well as the
in-hospital mortality rates. This may be evidence for an increasing number of complex
interventions underrepresented in the interventional index procedures.
Key quality indicators were the freedom of adverse events and in-hospital mortality
rates ([Table 3 ], [Figs. 3 ] and [4 ]). The overall in-hospital mortality of all cases was low with 1.6%.[22 ] When compared with published data, observed mortality and morbidity in this report
([Tables 3 ]
[4 ]
[5 ]
[6 ]) suggest overall good quality of medical treatment. However, this has to be carefully
interpreted with the given limitations of data acquisition and with respect to the
fact that we only report observed mortality and morbidity and did not compare with
calculated expected mortality and morbidity rates.[27 ]
[28 ]
[29 ] As expected, mortality was highest after Norwood I procedures.[30 ] These patients carried the largest morbidity burden as well, which was expressed
by all key performance indicators ([Table 2 ]). Adverse event rates can be related to increasing risk classification ([Fig. 4 ]). As obvious in Norwood patients, adverse event rates were related to patient's
age too. However, demonstrating this data would have blown up the sunburst diagram
in [Fig. 4 ]. It is the nature of complex heart disease that there is remaining risk for major
or catastrophic adverse events ([Fig. 3 ]). This was more frequent in surgical cases but again it has to be emphasized that
interventions and operations must not be understood as competing, but as complementary
therapeutic options. Procedural results, process numbers, and event rates are frequently
unsuitable for direct comparison between these treatment modalities.
Completeness, validity, and further developments of this registry will depend on continued
efforts of the DGTHG and the DGPK in close collaboration with all pediatric cardiac
and heart surgical departments in Germany. It will be of outstanding importance to
increase patient safety even further and to ascertain the continued high quality of
invasive procedures for the treatment of CHD.
Limitations
This registry is limited by its all-comers registry design and the voluntary participation
of patients and institutions. Currently, in this report we analyze at least 66% of
the surgical and 62% of the interventional procedures nationwide with larger completeness
from the participating centers. The range of reported lethality in cases with single
surgery or planned multiple procedures compares to anonymous self-disclosure data
in the current German Heart Surgery Report.[9 ] However, the number of not reported cases might significantly affect the real postinterventional
and postoperative mortality and morbidity. Though data integrity is monitored before
case closure, no source data monitoring could be performed in this registry so far.
This registry does not execute active follow-up tracking of patients.
