We investigated the presence of autoregulation on the ONH using different parameters of the pulse waveform determined by LSFG. It is very important to investigate it, because dysfunction of autoregulation is related to the cause of various ocular diseases including glaucoma [1–3]. Our results showed that the BOS, FAI, and RI of the pulse waveforms changed significantly in response to changes in the OPP induced by an elevation of the IOP in normal subjects.
The BOS demonstrated the constancy of the blood flow during a beat, and the RI indicated the resistance to flow in the arterial vessels. Considering the formula for calculating the BOS and RI, these parameters should have a strong inverse relationship [20, 32]. Takeshima et al reported that there was a significant increase in the BOS and a significant decrease in the RI in glaucoma patients, and a decrease in the IOP after trabeculectomy. In addition, the reduction in the IOP was significantly associated with the changes in BOS and RI . We evaluated the inverse changes by an IOP elevation, and our results showed that the opposite changes in those waveforms, e.g. a significant BOS reduction and a significant RI increase were observed after IOP elevation by 20 or 30 mmHg. In addition, the degree of the BOS decrease and the RI increase was greater with the higher increase of the IOP. Kiyota et al. reported similar changes in these parameters on the ONH and the choroid after artificial experimental IOP elevation . Our results corroborate those previous reports [32, 33].
Representative MBR pulse waveforms of the ONH before (a), immediately after (b), and 10 min after IOP elevation are shown in Fig. 7. These waveforms show that the blood flow is reduced and becomes unstable during the IOP elevation. These results are in good agreement with earlier color Doppler imaging -derived findings that the retrobulbar central retinal artery blood flow velocity decreased and the RI increased during the IOP elevation . The FAI was increased during the IOP elevation, and it was calculated from the maximum change of all frames (1/30 s) in a rising curve [20, 35]. The degree of the maximum MBR was not severely reduced compared with the minimum MBR, and the large difference between the maximum and minimum MBRs caused the lower BOS and higher RI and FAI. A situation in which there is not a totally decreased MBR during the IOP elevation, should be associated autoregulation on the ONH.
During the 10 minutes of IOP elevation by 30 mmHg in Experiment 2, the BOS and RI of the pulse waveform and the MBR recovered to the baseline values during the 10 min IOP elevation. However, the OPP was remained reduced. Interestingly, both BOS and RI significant recovered at 10 min after IOP elevation, but the MBR significant recovered as soon as 1 min after the IOP elevation. The result of MBR confirmed previous findings that blood flow on the ONH recovers just after an experimental increase in IOP in rabbits  and cats . These results indicate that the blood flow on the ONH is autoregulated. The blood flow on the ONH originates from the short posterior ciliary artery which suggests that a local autoregulatory system might exist in the deep region of the ONH.
Autoregulation is caused by changes in the blood vessel tone and in the resistance to blood flow, and it is related to dilatations of the blood vessels in spite of changes in the OPP. Myogenic and metabolic mechanisms control the blood flow on the ONH when the OPP changes . The myogenic responses are activated after changes in the diameter of the blood vessel lumens and the intraluminal pressure. Our results showed an increase in the RI immediately after the IOP elevation which probably occurs because the pressure on the ocular vessels causes a decrease in the vascular capacity . There was a large difference in the percentage change in the RI (+ 50.4%) and the MBR (-25.0%) at an IOP elevation of 30 mmHg. The myogenic responses should be activated when the IOP is decreased by the IOP elevation. Accordingly, the myogenic responses would develop to regulate blood flow on the ONH immediately after an increase of the IOP.
Both the BOS and RI recovered at 10 min during the IOP elevation, Furthermore, the percentage reduction of BOS and RI were significant correlated with that of the OPP, but then was not significantly during the 10 min IOP elevation. These results imply the presence of another autoregulatory process. A long duration of IOP elevation results in hypoxia of the tissues which causes accumulation of metabolites  and the release of nitrous oxide (NO) . These changes induce a vasodilatation of the retinal vessels. Our study showed that the decreasing BOS and the increasing RI immediately after IOP elevation significant recovered during the 10 min IOP elevation indicating that the activation of some metabolic mechanisms during a longer period of IOP elevation on the ONH even though the OPP remained reduced.
The blood flow recovered even though the IOP was elevated because of decreasing resistance to blood flow, but it did not fully return to baseline level during the 10 minutes measuring period. A longer period of IOP elevation measurements is needed to determine whether the ONH blood flow fully recovers. However, this it is difficult to do because of ethical issues. Eyes with poor autoregulation are at much greater risk of developing ONH ischemic damage than those with efficient autoregulation. Thus, under the abnormal conditions in which autoregulation of the blood flow is lost, such as in eyes with diabetic retinopathy  and with glaucoma , it is important to pay more attention to the increase in the IOP in such diseases.
There are limitations in our study. First, the variables were measured only with 20 and 30 mmHg elevations. Many studies reported on the effects of stepwise elevations of the IOP [13, 14, 42–45]. Second, the ONH blood flow largely recovered due to autoregulation after 10 minutes of IOP elevation, but it did not fully return to the baseline levels. However, times longer than 10 min were painful, and it could not be extended for ethical reasons. Third, our study had many myopic eyes. The morphological features of the optic disc of myopic and non-myopic eyes are different  which might affect the results. Fourth, only relatively young subjects were studied, and the results cannot be extrapolated to elderly subjects. Fifth, our sample size was relatively small. Further studies with a larger number of subjects including non-myopic subjects, a wider range of ages, and IOP elevations in a step-by-step manner are needed.