Purpose: The aim of this paper is to investigate the limits of LET monitoring of therapeutic carbon ion beams with miniaturized microdosimetric detectors. Methods: Four different miniaturized microdosimeters have been used at the 62 MeV/u 12C beam of INFN Southern National Laboratory (LNS) of Catania for this purpose, i.e. a mini-TEPC and a GEM-microdosimeter, both filled with propane gas, and a silicon and a diamond microdosimeter. The −yD (dose-mean lineal energy) values, measured at different depths in a PMMA phantom, have been compared withLE¯TD (dose-mean LET) values in water, calculated at the same water-equivalent depth with a Monte Carlo simulation setup based on the GEANT4 toolkit. Results: In these first measurements, no detector was found to be significantly better than the others as a LET monitor. The −yD relative standard deviation has been assessed to be 13% for all the detectors. On average, the ratio between −yD and LE¯TD values is 0.9 ± 0.3, spanning from 0.73 ± 0.08 (in the proximal edge and Bragg peak region) to 1.1 ± 0.3 at the distal edge. Conclusions: All the four microdosimeters are able to monitor the dose-mean LET with the 11% precision up to the distal edge. In the distal edge region, the ratio of −yD to LE¯TD changes. Such variability is possibly due to a dependence of the detector response on depth, since the particle mean-path length inside the detectors can vary, especially in the distal edge region.