The prevalence of Pompe disease in Austria found in the present study, 1:350,914, is considerably lower than that reported for Europe, 1:283,000 [4]. Although the prevalence of Pompe disease varies [2] it is reasonable to assume that there are still some undiagnosed individuals with Pompe disease in Austria. In the paediatric cohort, children with the severe, classical course of IOPD are generally less prone to be missed due to the rapidly progressive, life-threatening course. In contrast, late-onset patients with slowly progressive myopathy but without cardiac involvement experience not only a significant diagnostic delay but the diagnosis may also be missed.
The frequency of undiagnosed LOPD has been investigated in several studies [6, 12,13,14,15,16,17] and in the largest of these studies, 2.4% of patients with hyperCKemia or limb-girdle muscular dystrophy (LGMD) of unknown origin were found to have LOPD [6]. This suggests that there are also some undiagnosed LOPD patients in Austria, and is supported by pts 19, 20 and 21, who were diagnosed by whole exome sequencing for diagnostic workup of pauci- or asymptomatic hyperCKemia.
The clinical symptoms of the Austrian LOPD cohort are in line with the typical findings in Pompe disease [18,19,20]. Core features are proximal lower limb and axial weakness, frequently in combination with respiratory weakness, while scapular winging is rare. Although such a pattern of weakness is also seen in LGMD, especially prominent axial and respiratory weakness should alert the clinician to search for Pompe disease. Despite these suggestive clinical features, mean diagnostic delay was 8.8 years, which is similar to previous reports [18,19,20,21]. Several factors have been found to influence the diagnostic delay, such as presenting signs and symptoms, year of diagnosis and age at symptom onset [18,19,20,21]. Due to the small sample size, these factors cannot be statistically analysed in our population, but it seems that age is an important factor as the mean diagnostic delay was 18.6 years in patients currently older than 40 years and only 2.8 years in patients younger than 40 years. As Pompe disease is a progressive disorder, which causes significant morbidity and increased mortality [7], early diagnosis and treatment are essential [22]. It remains to be seen whether simplified diagnostic procedures such as the dried blood spot test [23], easily accessible genetic testing and campaigns to increase awareness will further reduce the diagnostic delay in the near future.
Besides typical clinical findings, electromyographic (EMG) abnormalities, in particular spontaneous activity and myotonic discharges, have been considered to be suspicious for Pompe disease [24]. Results from EMG studies were available in only 15 of our patients and myotonic discharges and abnormal spontaneous activity were observed in only two and six patients, respectively. Only a few studies reported the frequency of these EMG abnormalities in Pompe disease, and some found them frequently [25, 26] others only rarely [27, 28] in their population. This discrepancy might be explained by the muscles studied as Kassardjian et al. [26] clearly showed that the highest diagnostic yield can be achieved when sampling from paraspinal muscles and the tensor fasciae latae muscles. This might also explain the low rate of EMG abnormalities in our cohort as these muscles have not been investigated.
The c.-32-13T > G mutation was the most frequent mutation (85%) found in Austrian LOPD patients, which equals reports from nearby European countries [19, 27,28,29,30]. Similar to other studies, we also found some new mutations, which were all predicted to be pathological and Pompe disease has been verified by reduced enzyme activity either in the dried blood spot test or in muscle tissue. In contrast, the c.-32-13T > G mutation was only seen in 50% of IOPD. The known mutations found in IOPD are predicted to be severe with the exception of c.784G > A, which is predicted to be potentially less severe (http://cluster15.erasmusmc.nl/klgn/pompe/mutations.html). In this case the second, yet undescribed frameshift mutation c.1057delC, p.(Gln353Serfs*39) probably defines the rather severe phenotype.
The response to ERT could be evaluated in only about half of the patients who underwent yearly standardized assessments. Recently published European guidelines [11] recommend a standardized regular monitoring to gain more knowledge about the disease course under ERT to facilitate informed, cost-sensitive treatment decisions. As can be seen from the figures, the effects of ERT vary between patients and as nicely described by [22] there are obvious responders and non-responders. In the patients we were able to analyse, the 6minWT improved or remained unchanged in only two of six, while FVC improved or remained unchanged in seven of ten. This is somewhat different to treatment responses reported in a large meta-analysis [7]. However, a comparison between the published results and those presented here seems invalid, as our population is small and the observation periods vary greatly.
To summarize, in this study we report the phenotypes and genotypes of Austrian IOPD and LOPD patients and their response to treatment. The genotype corresponds to that described in other European countries and the reported clinical features corroborate the typical phenotype in LOPD with limb-girdle and axial weakness in the majority of patients. The incidence in Austria, however, is lower than that reported for Europe, suggesting that some patients go undiagnosed. Despite the fact that all patients have access to ERT, the recommended yearly standardized assessment is only performed in about half.