In vivo measurement of time-resolved autofluorescence at the human fundus

An experimental setup for measurement of time-resolved autofluorescence of the human eye fundus is demonstrated. The method combines laser scanning technique and time-correlated single photon counting. The light source is a laser diode, delivering pulses of about 100 ps duration at a repetition rate of 40 MHz. The excitation wavelength is 446 nm and the cutoff wavelength of fluorescence detection is at 475 nm. The autofluorescence can be determined with a spatial resolution of 80 x 80 microm2 and 25 ps time resolution. The fluorescence decay is optimally approximated by a biexponential model. The dominating lifetime tau1 is shortest in the macula (320 to 380 ps) and reaches 1500 ps in the optic disk. The lifetime tau2 varies between 2 ns and 5 ns, but the spatial distribution is more homogeneous. Respiration of 100% oxygen for 6 min leads to changes in the fluorescence lifetime pointing to detection of coenzymes. Diagrams of lifetime tau2 versus tau1 are well suited for comparison of substances. Such lifetime clusters of a 20 deg macular field of a young healthy subject and of a patient suffering from dry age-related macular degeneration overlap only partially with tau2-tau1 clusters of lipofuscin.

[Alterations in autofluorescence decay time in the fundus after oxygen provocation]

BACKGROUND

Alterations in the equilibrium of redox pairs of co-enzymes give information about the metabolic state in the citric acid cycle as well as in the respiratory chain. Fluorescence properties are different between the reduced and oxidised states of co-enzymes so that a change of the oxygen partial pressure can be sensitively detected in the tissue. Therefore, it was investigated whether the autofluorescence of co-enzymes is detectable in the living fundus.

METHOD

The provocation of metabolism was realised by inspiration of 100% oxygen. The time-resolved autofluorescence was detected by a single photon counting technique. The decay behaviour of autofluorescence was approximated by a bi-exponential model. For evaluation of the results, histograms of decay times tau(1) and tau(2) were calculated in defined ranges around the macula and optic disc before, during, and after inhaling oxygen.

RESULTS

The calculated decay times corresponded in the macula to the decay times of FAD and NADH(+) given in the literature. Connective tissue in the optic disc also showed fluorescence.

CONCLUSIONS

Changes in the histograms of decay rates demonstrate that provocation of metabolism is detectable by measurement of time-resolved autofluorescence. This method reveals the evaluation of metabolism at the cellular level as a new diagnostic possibility.