Key words DeGIR registry data - interventional radiology - bleeding - module B - embolization
- nationwide availability
Introduction
Active bleeding is a potentially life-threatening condition that often occurs as a
result of trauma or as an iatrogenic complication. Tumor bleeding and coagulation
disorders are a less common cause [1 ]
[2 ]
[3 ]
[4 ]
[5 ]. As a result of the good, scientifically verified results and the rapid development
and availability of transcatheter vessel occlusion performed by radiologists in German
clinics, interventional radiological treatment of bleeding is to be given preference
over other alternatives in certain situations, e. g., based on location or in the
case of a coagulation disorder, difficult surgical access, a site having undergone
extensive prior surgery, or inoperability of the patient [3 ]. Intubation anesthesia is typically not necessary for interventional radiological
treatment. Therefore, angiographic evaluation with interventional radiological treatment
to stop the bleeding can be performed immediately after diagnosis of active bleeding.
In addition, interventional radiological treatment of bleeding has increasingly been
included in the guidelines of other disciplines, e. g. gastroenterology and gynecology,
at least as an option [6 ]
[7 ].
Finally, not only the effectiveness and quality of a certain technique but also its
availability and the structural distribution of centers with sufficient practical
experience for regional coverage of need are decisive for practical patient care.
This was recently evidenced by the rapid development of the interventional radiological
treatment of strokes by means of thrombectomy [9 ].
As part of a quality assurance program, the German Society for Interventional Radiology
and Minimally Invasive Therapy (DeGIR) has been recording vascular and non-vascular
interventions in a registry run in collaboration with the German Society for Neuroradiology
(DGNR) for over 25 years. The registry includes the following modules: Module A (methods
for opening and reconstructing vessels), module B (transcatheter vessel occlusion),
module C (diagnostic puncture, drainage, PTCD, TIPSS, etc.) Module D (oncological
methods, primarily tumor-specific embolization and ablation), module E (neurointerventions
for opening vessels) and module F (neurovascular embolization treatments).
The goal of this overview is to evaluate whether there is nationwide availability
of module B interventional radiological treatment of active bleeding for patients
in Germany. DeGIR module B interventions are considered particularly challenging special
interventions. The literature does not contain any numbers regarding the nationwide
availability of methods of surgical and endoscopic hemostasis. Therefore, it is not
the goal of this study to compare the interventional technique with these techniques.
Methods
Data collection
The analysis was based on DeGIR registry data from the years 2016 and 2017. The data
was recorded during this time using the BQS software (Institut für Qualität & Patientensicherheit
GmbH). Data regarding module B interventions entered on a voluntary basis was used
as a surrogate for determining performance and experience in relation to interventional
radiological hemostasis at clinics. DeGIR module B encompasses methods for occluding
vessels including coils, fluid embolization, particles, plugs, and the like.
The number of centers fulfilling the criteria for DeGIR certification as training
centers (at least 20 interventions per year) or that are already certified was determined
for further analysis. This essentially requires the completion and documentation of
at least 20 module B interventional radiological interventions and an interventionalist
with personal level 2 module B certification. This personal level 2 certification
requires proof of at least 100 completed module B interventions and 30 CME points.
“High-volume” centers were defined as having intervention numbers above the 9th decile.
Analysis of coverage
Data was assessed by state. For detailed analysis of coverage, without making the
data of individual clinics visible, the recorded module B interventions were broken
down by 40 smaller regions (districts, former districts and states: Arnsberg, Berlin,
Brandenburg, Braunschweig, Bremen, Chemnitz, Darmstadt, Dessau, Detmold, Dresden,
Düsseldorf, Freiburg, Gießen, Halle, Hamburg, Hannover, Karlsruhe, Kassel, Koblenz,
Cologne, Leipzig, Lüneburg, Magdeburg, Mecklenburg-West Pomerania, Middle Franconia,
Munster, Lower Bavaria, Upper Bavaria, Upper Franconia, Upper Palatinate, Rheinhessen-Pfalz,
Saarland, Schleswig-Holstein, Swabia, Stuttgart, Thuringia, Trier, Tübingen, Lower
Franconia, Weser-Ems).
Statistics
R Statistics (R version 3.5.3 (2019–03–11) -- “Great Truth”) was used for descriptive
statistics [8 ]. p = 0.05 was accepted as the level of significance.
Graphics
The following software was used to create graphics:
Creative Commons Attribution 3.0 License (www.geonames.org ), Geojson Germany (https://github.com/isellsoap/deutschlandGeoJSON ), https://www.destatis.de/DE/Service/Impressum/copyright-genesis-online.html (Federal Statistical Office of Germany (Destatis), https://krankenhausatlas.statistikportal.de/ ; data license dl-de/by-2-0, https://www.govdata.de/dl-de/by-2-0
Technical data:
Basic data:
Genesis-Online; data license dl-de/by-2-0), Folium/Geopandas/Shapely/Python (mapping).
Results
Registry data from 242 clinics in Germany was evaluated for the analysis of DeGIR
module B. 16,763 module B interventions in 2016 and 16,399 in 2017 were recorded in
the DeGIR registry via voluntary entry. Changes over the course of the two years were
not statistically significant. The median number of interventions performed per facility
was 41 in 2016 and 40 in 2017. The DeGIR requirement for certification as a module
B training center was met by 160 clinics in 2016 and 162 in 2017. The total number
of interventions performed at high-volume centers was constant in 2016 and 2017 (over
500 interventions per year at 23 clinics).
Availability of care
Normalized to one million inhabitants, an average of 211 interventions were performed
in 2016 and 200 in 2017 in Germany (standard deviation = 101 and 109, respectively).
The median was 202 and 222, respectively, per one million inhabitants. The combined
result for 2016 and 2017 was a median of 425 per one million inhabitants (standard
deviation = 205). The interquartile range (IRQ) was 293–521 interventions per one
million inhabitants with the lowest value in Bremen (87) and the highest in Saarland
(888). [Fig. 1 ] provides an overview of the recorded interventions per million inhabitants for every
state.
Fig. 1 Performance of different states. In Fig. 1 the number of interventions is illustrated
by a bar plot for each state in Germany. The number is a summary of 2016 and 2017
and normalized to one million citizens. The red line illustrates the median of 425.
The lower dashed red line illustrates the first quartile (293) and the upper red line
the third quartile (521).
The analysis of the districts and former districts yielded an average of 430 interventions
in 40 regions per year (calculated for 2016 and 2017) with a standard deviation of
302. The median is 413 interventions per year. There are no districts without recorded
interventions. [Fig. 2 ] shows the absolute number of interventions by state and region as well as the development.
[Fig. 3 ] shows the combined number of interventions per million inhabitants for every state
from the years 2016 and 2017. However, individual smaller regions have a smaller number
of hospitals in which interventional radiological treatment of acute bleeding can
be offered.
Fig. 2 Widespread distribution of interventions and changes. Fig. 2 shows the number of
interventions in 2016 and 2017 for the states and regions. In A–C the absolute numbers are shown in increasing shades of green and the changes from
2016 and 2017 in increasing shades of blue (D ). A shows the absolute number of interventions in 2016 for each state and B shows the numbers from 2017. C illustrates the summarized numbers for each region from 2016 and 2017. D shows percentage changes in interventions between 2016 and 2017 on the state level;
small changes and negative tendencies were encoded as white areas.
Fig. 3 Performance of the different states per one million citizens. Nationwide distribution
of transcatheter vessel occlusion (2016 and 2017) on the state level per one million
citizens in Germany.
Development between 2016 and 2017
In cases recorded between 2016 and 2017, there were no statistically relevant changes.
However, there were some significant differences locally on the district and former
district level as described above. On average, the percentage change was + 0.02 %
with a standard deviation of 37 %. The maximum decrease in a region between the two
years was 96 % while the maximum recorded increase in a region was 88 %. [Fig. 2D ] shows the development between 2016 and 2017 for the individual states.
Discussion
The analysis of the DeGIR registry data from the years 2016 and 2017 regarding the
comprehensive Germany-wide availability of transcatheter vessel occlusion (module
B) for assessing the interventional radiological treatment of acute bleeding showed
that i) there is good nationwide availability on a state level and ii) the distribution
of experience regarding the necessary procedures is almost equally high among the
individual regions. White dots on the map coincide with regions that have a lower
density of hospitals and thus have relatively poor coverage, see [Fig. 4 ].
Fig. 4 Distribution of clinics in Germany https://www.destatis.de/DE/Service/Impressum/copyright-genesis-online.html (Statistisches Bundesamt (Destatis), https://krankenhausatlas.statistikportal.de/ ; Datenlizenz dl-de/by-2-0, https://www.govdata.de/dl-de/by-2-0 .
Interventional radiological treatment of acute bleeding has increasing clinical value
and also a significant benefit for patients since the chances of successfully correcting
this life-threatening condition are very high. In addition, interventional radiological
treatments are minimally invasive with high efficiency and effectiveness and low procedural
stress for the patient. In contrast to open surgery and often also endoscopic procedures,
endovascular treatment is usually highly targeted and is gentle to surrounding tissue
structures. In addition, intubation anesthesia is sometimes not necessary [9 ].
In addition to the effectiveness and level of risk of a method, its availability plays
a major role in the practical situation and especially in patient care in general.
The DeGIR registry data shows high nationwide availability on the state level. The
radiologists performing interventional radiological procedures to treat acute bleeding
also have the necessary experience. The results coincide with the neuroradiology results
regarding DeGIR module E data [10 ]. The large number of centers that are suitable for certification as a DeGIR training
center or are already certified and also the large number of high-volume centers indicate
that the conditions are good for training young radiologists interested in interventional
radiology. There are no comparable numbers for surgical or endoscopic methods of hemostasis
in the literature. Therefore, we were not able to compare various techniques. It is
also not possible to make a statement about the percentage distribution of the use
of different hemostasis techniques. However, this was also not the goal of the present
study. Instead, the aim was to examine whether nationwide availability of interventional
radiological techniques for treating acute bleeding is ensured in Germany. Based on
the current data, the answer is yes even though there are limitations regarding the
interpretation of the voluntarily recorded registry data as discussed in the following.
Interpretation of the registry data
DeGIR module B data was used in the present analysis as a surrogate for experience
and performance distribution for interventional radiological emergency care in Germany.
It is important to note that performance data is recorded on a voluntary basis. Therefore,
it can be concluded that the number of interventions and experience with the corresponding
treatments are significantly higher than the present numbers indicate. The significant
fluctuation in the number of recorded interventions in some regions in the years 2016
and 2017 with a constant Germany-wide total number of interventions is a result of
multiple parallel effects. In numerous cases, the fluctuations exceed the variance
of incidence of the indications for interventional radiological treatment to be assumed.
The reason for the local fluctuations of up to approx. +-90 % is ultimately unclear
and can possibly be explained by inconsistent recording of performance data in individual
clinics. Since the recording of performance data in the DeGIR registry is currently
performed on a voluntary basis and also entails a certain time requirement, the recorded
number of interventions can vary greatly. However, in individual cases, the availability
of the technique could have changed thus explaining the significant local fluctuations.
The switch of an experienced interventional radiologist to another clinic would be
a further possible explanation for the fluctuation. The acceptance of the minimally
invasive technique by referring physicians and a change among these colleagues could
also be responsible for the fluctuation.
Over the years more and more radiology clinics have entered data in the DeGIR registry.
Only after a longer stagnation it will be possible to evaluate the development of
the interventional treatment of acute bleeding based on the absolute number of module
B interventions.
Availability of care
On the whole, the analysis shows good nationwide availability on the state level in
Germany of transcatheter vessel occlusion performed by interventional radiologists.
However, given an average of approximately 200 in both 2016 and 2017, the number of
interventions in every state per one million inhabitants varies greatly in some cases
(see [Fig. 3 ]). In individual regions, e. g. Mecklenburg-West Pomerania, Bremen and Saarland,
Hamburg and Thuringia, the deviation from the average is more than a standard deviation.
As discussed above, this is due to fluctuations in registry data. Status as a city
state must also be considered as a reason for the deviation in performance data. Large
city states can recruit patients from neighboring states while smaller city states
or even states can be subject to greater statistical fluctuations due to the consequently
smaller amount of data in this regard and individual centers have a major influence
on registry data. Based on the current data, the availability of individual interventionalists
at a particular center cannot be ruled out as a factor since the individual interventions
in the registry cannot be broken down by physician for reasons of privacy. Differences
in established clinical processes must also be considered as a further cause of local
fluctuations. In some clinics, cases of active bleeding tend to be referred to surgery
so that these cases are treated but do not undergo interventional therapy and are
thus not entered in the registry.
Some regions have a lower density of hospitals offering interventional radiological
treatment in cases of acute bleeding. In individual cases, this means longer transport
times and routes for patients requiring interventional radiological treatment for
acute bleeding. Therefore, in the reverse conclusion, the high number of certified
vascular centers in Saarland could be responsible for the unusually high number of
interventions in the nationwide comparison.
The future of DeGIR/DGNR quality registry data
The DeGIR has been continually improving and revising web-based recording of performance
data since 1994 to simplify data entry and to reduce the time needed to enter data
while maintaining suitability for quality assurance. In 2018, the web-based solution
provided by Samedi was implemented. In addition to structural data, indication data,
procedural data, and parameters regarding result quality, the registry includes data
regarding patient radiation exposure as a result of the particular intervention. Therefore,
the registry data could also be used for mandatory dose reporting. Moreover, § 137
of the Social Code states that quality assurance is mandatory. This requirement could
be satisfied at the same time when entering data in the registry. Furthermore, the
recorded data provides the foundation for multi-level personal and institutional DeGIR/DGNR
certification in the various treatment modules of interventional radiology. Mandatory
nationwide data entry in the future would increase the value of results from analyses
regarding various issues.
