Incidence of acquired thrombotic thrombocytopenic purpura in Germany: a hospital level study

A systematic literature search, following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, [20] was performed to determine the available published German evidence on the aTTP epidemiology. Details on the methodology of this systematic literature are provided as supporting information in the Additional file 1 of this manuscript.

Acquired TTP and hemolytic uremic syndrome (HUS) belong to a group of diseases known as thrombotic microangiopathies (TMAs), which present with platelet consumption, microangiopathic hemolytic anemia (MAHA) and organ dysfunction resulting from endothelial damage and microvascular thrombosis [21]. Whereas aTTP is primarily associated with central nervous system involvement and HUS is mainly seen as a disease of the kidneys, both are multi-organ diseases and may present with a quite comparable symptom complex and clinical picture, with ADAMTS13 activity as only reliable differentiation criteria [22]. Furthermore in Germany both aTTP and HUS are grouped in the same Diagnosis Related Group (DRG: L72Z named Thrombotic Microangiopathy and Hemolytic Uremic Syndrome) [23]. In Germany the tenth revision of the international classification of diseases (ICD-10) is usually used for coding purposes. As aTTP could either be coded as TMA (ICD-10 M31.1) or HUS (D59.3) national estimates on the annual frequency of both diagnoses (TMA and HUS) were determined from national hospitalization statistics [24‐26], and from the German DRG (G-DRG) browser [27‐29].

A hospital-level study was performed to determine the proportion of aTTP cases within a population of hospitalized TMA/HUS patients. To figure out the actual number and the proportion of aTTP cases coded within ICD-10 M31.1 (TMA) and ICD-10 D59.3 (HUS), a retrospective epidemiological data collection in Germany hospitals was performed. Based on the ICD-10 code descriptions, it was expected that most of the aTTP cases would be grouped within ICD M31.1 (named “Thrombotic Microangiopathy” including “Thrombotic thrombocytopenic purpura”). In order to identify the German key TMA hospitals, data from the quality reports of the hospitals (German: Qualitätsberichte der Krankenhäuser) [30], were determined as these include the number of TMA hospitalization cases (ICD-10 M31.1) on the hospital level. Using these data, the 30 German hospitals with the highest number of TMA diagnoses in 2013, 2014 and 2015 were determined, (this period refers to data used for hospital selection only) as TMA patients are usually referred to the centers of maximum care that are capable to provide all necessary TMA diagnostics and therapies. These 30 hospitals reflect mainly university hospitals and those were invited to participate in the study. Following the approval by the local ethic committees, the participating hospitals were requested to identify all hospitalization cases with the primary or secondary diagnosis of TMA (M31.1) or HUS (D59.3) recorded between 2014 and 2017, and to extract the following predefined information for each case:

Patients were considered to have a confirmed diagnosis of aTTP if ADAMTS13 activity levels were < 10% and/or the medical records explicitly mentioned the diagnosis of aTTP. The differentiation between initial and recurrent aTTP episodes was performed on the basis of medical records. By applying this aTTP definition, a conservative approach was chosen as other potential approaches (e.g., ADAMTS13 activity levels < 10% as only criteria) would potentially have led to an underestimation of aTTP cases.

On the basis of these data, the number and proportions of aTTP cases within the ICD-10 M31.1 and the D59.3 (primary and secondary diagnoses at the hospital level) were determined.

Finally, the proportion determined in the hospital-level study was projected to the national level to elucidate an incidence of aTTP for Germany. On the basis of the outcomes of the hospital-level study, we calculated the proportion of patients with a confirmed diagnosis of aTTP for the primary and secondary diagnosis of TMA and HUS, respectively. In order to reflect the statistical uncertainty related to the hospital study sample a logistic regression analysis, in consideration of the single hospitals as a random effect measure (random effect model), was performed to reflect the 95% confidence intervals (95%CI) around the mean estimates of the related proportions. These proportions and the related 95% CIs were then applied to the national cases for TMA/HUS to project a national estimate of aTTP in Germany. Thereafter, we transferred results into an overall aTTP incidence estimate including both first and recurrent episodes of aTTP.

The determined number of incident aTTP cases are also expressed as cases per million using the number of German inhabitants, based on data from the Federal Statistical Office, related to the year investigated. When reporting the incidence of initial (new) aTTP cases the term “incidence of aTTP” is used, whereas for the overall incidence (initial & recurrent) of aTTP cases the term “incidence of aTTP episodes” is used.

In total 340 studies were identified via the database/journal searches, and 296 abstracts were reviewed (database/journal search minus duplicates). From these, 16 articles were selected for full-text review, and 2 papers met the inclusion criteria. The flow chart of study selection is shown in Fig. 2. Detailed information on the systematic literature search strategies’ and outcomes by database / journal are provided in the supporting information (Additional file 1).

As a result of the systematic search we identified two publications [31, 32] that fulfilled all inclusion criteria, but none provided a national estimate on the aTTP incidence or prevalence for Germany. For more information on these studies please refer to the supporting information (Additional file 1).

The national estimates for TMA and HUS primary and secondary diagnoses, which are based on the national hospitalization statistics [24‐26] and on the G-DRG database [27‐29], are presented in Table 1, for the years 2013 to 2016.

Eight centers (≈27% of the top 30 TMA hospitals requested to participate) delivered data according to a predefined collection form. In total, during the time frame of 2014 to 2017 (hospital study data collection period), 600 hospitalization episodes related to the primary (ICDs M31.1 or D59.3 coded as primary diagnosis) or secondary diagnosis (ICDs M31.1 or D59.3 coded as secondary diagnosis) of TMA/HUS were identified and extracted. As these eight centers are major TMA centers, with a high number of TMA cases, the number of cases that were retrospectively recorded (n = 600) reflect approximately 10% of all German HUS / TMA cases observed in a comparable 4-year period (total n = 5906 year 2013 to year 2016) [24‐29]. The aggregated results by hospital over this four-year period are presented in Table 2.

Using logistic regression analysis, the proportion of patients with a confirmed diagnosis of aTTP was calculated for the primary and secondary diagnosis of TMA (M31.1) on the basis of the outcomes of the hospital-level study. As aTTP cases were only identified in TMA patients (M31.1) the related proportion was computed only for the primary (mean 37.3%; 95%CI: 30.8–43.8%) and secondary TMA (mean 5.1%; 95%CI: 1.7–8.6%) diagnoses. Furthermore the proportion of recurrent aTTP cases (mean 29.9%; 95%CI: 20.3–39.5%) in relation to all aTTP episodes was computed (Fig. 3).

By combining the national estimates (Table 1) and the proportion of aTTP cases, national estimates for acute aTTP episodes were calculated, which are depicted in Table 3.

Data are given for the total population (including children, adolescents and adults) and for adults only (18+). The average annual values (year 2014–2016) were 166 (95%CI: 127–205) acute aTTP episodes projected in adults (18+) and 172 (95%CI: 132–212) in the overall population. The majority of acute aTTP cases in the overall population occurred as initial manifestations (average annual values year 2014–2016: 121; 95%CI: 105–129).

Using the number of inhabitants in Germany of the related years (year 2016: total: 82.50 million; 18+: 69.05; year 2015: total: 82.20; 18+:68.80; year 2014: total: 81.20; 18+: 67.96) [33, 34] the number of incident acute aTTP episodes (reported in Table 3) was transferred into cases per million inhabitants (Table 4). In the overall population (adults and children) the average annual incidence (year 2014–2016) of aTTP was 1.47 (95%CI: 1.28–1.57) and the average incidence of aTTP episodes was 2.10 (95%CI: 1.60–2.58).