In our previous study published in Clinical Genetics we estimated the frequency of the 1.4-Mb CMT1Aduplication in the Greek population at 25.9%, a figure significantly lower than reports from most other populations. We noted at the time that the reduced sensitivity of the long polymerase chain reaction (PCR) method used for detecting the duplication may have been partly responsible for underestimating the duplication frequency (1).To re-evaluate the duplication frequency with a method that is almost 100% sensitive, we screened 128 consecutive Greek CMT1 index cases presenting from 2006 to 2013 using Multiplex Ligation-dependent Probe Amplification (MLPA) (2, 3). Inclusion criteria were similar to our previous study (1). Seventy-eight cases were familial (60.9%) and 50 non-familial (39.1%). Informed consent was obtained from all participants.The duplication frequency using MLPA was 52.3%, higher than our previous estimation (p = 0.0001) and similar to frequencies reported in most populations (1). In familial cases the frequency was 66.6% (52/78) and in non-familial cases 30% (15/50). Subsequent screening of the 128 cases with long PCR revealed an overall sensitivity of 67.16%, lower than figures quoted in the literature (84-87%) (2, 4). In familial cases the sensitivity was lower (59.6%) compared to non-familial cases (93.3%, p = 0.014). This indicates that in familial Greek CMT1A cases the recombination event has ∼40% possibility of occurring outside the usual 3.2-kb recombination hotspot, rather than the usually quoted figure of <25%, possibly because of relative genetic isolation (5). In the Greek population, therefore, the sensitivity of long PCR depends on the characteristics of the patient sample. The more familial cases the lower the sensitivity. Given a 75.3% of familial cases in our original sample, this largely explains our earlier underestimation of the duplication frequency (1). In conclusion, a more accurate estimation of the CMT1A-duplication frequency in Greek patients is ∼50%, similar to other ethnic populations.