Abstract
Purpose :
Decreased outflow facility, C, causes intraocular pressure (IOP) elevation in glaucoma. The in vivo mouse model is a key tool for glaucoma research but reported values of C in vivo vary by an order of magnitude. There is thus an urgent need for a precise and repeatable method to measure C in mice in vivo.
We posit that a major source of variation is the time-dependent changes to aqueous humour dynamics that occur under anaesthesia and propose a new approach to address this.
Methods :
Our ‘Corrected method’ uses both IOP during perfusion, P(t), and the IOP that would exist under anaesthesia in the absence of perfusion, S(t). S(t) cannot be directly measured but can be accurately inferred via repeated measurements of the flow rate from the perfusion system Q at a single pressure S0, interlaced between measurements at different P values. It can be shown that Q(t) is given by Q(t)=C(P(t)-S(t)). Linear regression can be used to extract C. This approach compares to the ‘Typical method’, which treats S(t) as constant.
To compare methods, we carried out multiple studies with humidity controlled iPerfusion systems (1), and isoflurane anaesthetised C57BL/6J mice. We consider fold increase in Precision of the Corrected Method based on the 95% margin of error (ME) for relevant parameters:
Precision Increase = Typical method ME / Corrected method ME
Results :
In N=16 independent eyes at Imperial College, the average facility was 6.9 [5.9, 8.1] nl/min/mmHg with the Typical method vs 5.9 [5.4, 6.6] nl/min/mmHg with the Corrected method, a 1.8-fold increase in precision. In N=9 different mice, we found a 2.2-fold increase in precision on the % difference between contralateral eyes. For N=9 mice with contralateral eyes treated with Netarsudil solution vs control perfusate, we observed a 3.5-fold increase in precision on the % difference between contralateral eyes (Figure).
To demonstrate repeatability, the average facility in N=16 independent eyes as measured on different systems at Columbia University by another researcher yielded 5.9 [4.6,7.5] nl/min/mmHg.
Conclusions :
The new technique greatly increases precision by accounting for anaesthetic effects. Measured values of C are reproducible across institutions and comparable to values measured for ex vivo and postmortem perfusions (4.3 nl/min/mmHg & ~4.2 nl/min/mmHg respectively (1)).
[1] Madekurozwa et al. 2022, doi.org/10.1016/j.exer.2022.109103
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.