Global Epidemiology of Gaucher Disease: an Updated Systematic Review and Meta-analysis

Background: Gaucher disease [GD], an autosomal recessive lysosomal storage disorder, is characterized by progressive lysosomal storage of glucocerebroside in macrophages predominantly in bone, bone marrow, liver, and spleen. Meta-analysis of global GD epidemiology was not available before this study. Methods: To provide a systematic review and meta-analysis of birth prevalence and prevalence of GD in multiple countries. MEDLINE and EMBASE databases were searched for original research articles on the epidemiology of GD from inception until July 21, 2021. Meta-analysis, adopting a random-effects logistic model, was performed to estimate the birth prevalence and prevalence of GD. Results: Eighteen studies that were screened of 1874 records were included for data extraction. The studies that fulfilled the criteria for inclusion involved 15 areas/countries. The global birth prevalence of GD was 1.5 cases [95% confidence interval: 1.0 to 2.0] per 100,000 live births. The global prevalence of GD was 0.9 cases [95% confidence interval: 0.7 to 1.1] per 100,000 inhabitants. Conclusions: This is the first comprehensive systematic review that presented quantitative data of GD global epidemiology. Quantitative data on global epidemiology of GD could be the fundamental to evaluate the global efforts on building a better world for GD patients.


Literature Search Strategy
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement was the guideline for this systematic review and meta-analysis. 6 The complete checklist could be found in Additional file 1. The study strategy adopted to identify studies was as follows: EMBASE and MEDLINE were searched by terms ["incidence", "prevalence", "epidemiolog*" and Gaucher disease"] from inception until July 21, 2021. Endnote X7 was used to manage citations. Detailed literature search strategy for different databases was provided in Additional file 2.

Inclusion and Exclusion Criteria
Studies fulfilled all of the following criteria were selected: (1) the case collection was based on a field survey; (2) the study was based on population samples rather than volunteers; (3) the study had definite numerator [number of patients] and denominator [number of live births or inhabitants].
Studies that fulfilled any of the following criteria were excluded: (1) study without information available for metaanalysis was excluded; (2) conference abstract was excluded; (3) study that focused on one specific population from one area/country was excluded.  9 When P < 0.1 for the Q test or I 2 > 50%, the signs for substantial heterogeneity, were received, random-effects model was used; otherwise, fixed-effects model was performed. In addition, funnel plot was used to describe potential publication bias. Figure 1 showed the process of identifying eligible epidemiological studies. Eighteen studies, all of which met inclusion criteria and were not excluded by exclusion criteria, were selected and then subjected to quality assessment. Five [28%] studies were rated as high quality, 9 [50%] studies were considered to be of medium quality, and 4 [22%] studies were assessed as low quality (Table 1). Details about the quality of each included study were reported in Additional file 3.

RESULTS
The studies that fulfilled the criteria for inclusion involved 15 areas/countries.   (Fig. 3).
Although the range of birth prevalence and prevalence of GD was large, no qualitative difference in study methodology that could justify its impact on the pooled estimates was observed. No publication bias was found based on the funnel plot and Begg's test for birth prevalence and prevalence of GD [P value = 0.274 and 0.389; Fig. 5].

DISCUSSION
The upper limit in the definition of rare disease ranges from 5 to 76 cases per 100,000 people. 28 According to the definition of rare disease, quantitative epidemiology of GD in this study (1.5 cases [95% CI: 1.0 to 2.0] per 100,000 live births) confirmed that GD was a rare disease. 29 GD is extremely common in Ashkenazi Jews. Goldblatt and Beighton 30 reported that the prevalence of GD in the South African Jewish population would be 1:4000. Through population-based genetic screening programs, birth prevalence of GD was predicted to be 1:450. 31 In Ashkenazi Jews, GD may not be considered as a rare disease; however, epidemiology data of GD in one race could not represent other races.
Because of the founder effect, the data from a specific population from one area/country may affect the accuracy of global epidemiology of GD, so 3 studies that focused on 1 specific population were excluded: (1) Goldblatt and Beighton 30  When studies were screening and quality assessing, "incidence" was misused to present the frequency of GD among births. It is easy to distinguish the difference between incidence and prevalence. The numerator and denominator of incidence are the numbers of disease onsets and the number of healthy individuals [a population at risk] during periods of observation. The numerator and denominator of prevalence are the total number of cases and the number of population at a certain moment. 34 For new born screening of genetic diseases, including GD, patients were already there, so incidence is not suitable for frequency of GD among births. Birth prevalence, the prevalence at birth, was more suitable to present the frequency of GD among births. 8,35 Theoretically, prevalence should be not far from birth prevalence. 36 In this review, birth prevalence of GD (1.5  Taiwan Asia cases [95% CI: 1.0 to 2.0] per 100,000 live births) was higher than prevalence of GD (0.9 cases [95% CI: 0.7 to 1.1] per 100,000 inhabitants). Following reasons could explain such phenomenon: (1) it is very hard to find all GD patients; (2) the life span of GD patients is not long enough as a normal person. 37 Although birth prevalence was affected by many factors, including diagnostic technology, prenatal diagnosis, and termination of pregnancy, birth prevalence of GD may be more accurate than prevalence to calculate the number of GD patients for now.
To our best knowledge, there was only one comprehensive review to represent GD epidemiology. 5 In this review, the author used incidence and birth prevalence at the same time to show the frequency of GD among births, which would confuse the reader. In the part of "incidence", 11 studies were used to review "incidence" of GD. 5 Among these 11 studies, 8 studies were included in this review to calculate the birth prevalence of GD. 10,[12][13][14]18,19,21,22 In the part of "prevalence", prevalence of GD was reviewed on the basis of 9 studies. 5 Among these 9 studies, 3 and 2 studies were included in this review to calculate prevalence 10,19,38 and birth prevalence of GD, 11,16 respectively.
Pooled birth prevalence of GD in Europe was lower than the data in Oceania; however, the highest birth prevalence of GD was reported in Austria from Europe [5.8 cases per 100,000 live births]. 18 The lowest birth prevalence of GD in Europe, 0.2 cases per 100,000 live births, was reported in Turkey. 14 The difference of GD birth prevalence between Austria and Turkey may be explained by the proportion of Ashkenazi Jews in the two countries. 30,31 Three studies of GD birth prevalence in Asia were all from China, which has a low proportion of Ashkenazi Jews. If pooled birth prevalence of GD in Asia contained data from West Asia, the birth prevalence of GD in Asia may be higher.
According to pooled birth prevalence of three subtypes of GD, proportion of patients with GD 1 in total patients with GD is~83%, which is consistent with the review of Stirnemann et al [prevalence of GD1: 90%-95% in Europe and North America]. 1 There are two other studies that reported birth prevalence of 3 subtypes of GD; however, cases of GD 1 patients were separated into 2 groups [early and late], meanwhile cases of GD 2 patients and GD 3 patients were mixed. 11,16 These two studies were excluded.
Quantitative data of GD global epidemiology is the fundamental to evaluate global efforts that build a better world for GD patients, including more accurate The broadly varied data of life expectancy did not reduce the effect that higher pooled global data could reflect global efforts to expand life expectancy of citizens worldwide.

Limitations
There are several limitations of this report: (1) <30% of studies were assessed as high-quality, highlighting the need for high-quality study about epidemiological evidence of GD; (2) more than half of the studies were from Europe (56%). Reports from other continents were underrepresented, which might cause bias for calculating global epidemiology of GD.