Swelling of the collagen-keratocyte matrix of the bovine corneal stroma ex vivo in various solutions and its relationship to tissue thickness

Aim: The mammalian corneal stroma, like some other connective tissues, can absorb fluid, swell and become oedematous. Since studies on the corneal stroma have been carried out with different types of preparations and solutions, inter-study comparisons are very difficult. A study was thus undertaken on a standardised preparation to assess the relative magnitude of this swelling and its relationship to thickness of the preparations. Methods: From selected recent post-morterm eyes of adult cattle, stroma preparations were cut from the central part of the cornea. These preparations were immersed in various solutions of known pH and osmolality, and the time-dependent changes in wet mass were assessed over 9h at 37 degreesC. The relative rates and magnitude of the swelling of the tissue were then compared. Results: A reference value for stromal swelling was obtained by incubation in a 35 mM bicarbonate-buffered mixed salts solution equilibrated with 5% CO2-air (pH 7.60) where a 3.39-fold increase in wet mass and a 4.58-fold increase in thickness was realised in 9h, at an initial rate of 76+/-3%/h. The swelling was essentially the same in an organic buffer-mixed salt solution (pH 7.5) but progressively greater in phosphate-buffered saline (pH 7.5), a ran ge of phosphate buffers (10-67 mM, pH 7.5), NaCl solutions (0.025-1%) and with gross swelling observed in water (where a 15.9-fold increase in wet mass occurred along with a 25-fold increase in thickness, at an initial rate of 643+/-62%/h). Overall, the wet mass changes were strongly related to thickness (P < 0.001). Conclusions: The results confirm that the selection of solution(s) for studies on corneal stromal swelling is critical. The swelling (oedema) is lower in a physiologically-relevant solution (similar to the aqueous humour of the eye). This indicates that the swelling tendency of the corneal stroma has been overestimated in the past, and that a similar discrepancy may also exist for studies on other connective tissues ex vivo when non-physiological experimental solutions are used. (C) 2000 Harcourt Publishers Ltd.