The corneal endothelium continuously dehydrates the stroma and is therefore responsible for its transparency. This is achieved by an osmotic gradient that is generated by membrane-bound ion transport systems. The 2 most important systems are the membrane-bound sodium/potassium–adenosine triphosphate (ATPase) sites and the intracellular carbonic anhydrase pathway. Activity of both pathways produces a net flux from stroma to aqueous humor.1 The barrier portion of the endothelium is somewhat unique in that it is permeable to a degree to permit the ion flux necessary to establish the osmotic gradient.2,3
Endothelial cells are especially sensitive to surgical damage. This is because of their exposed position as the innermost corneal layer that borders on the anterior chamber and their limited ability to regenerate. Thus, surgical damage to the endothelium can have severe consequences on a surgery's functional outcome.
The pump function of the corneal endothelium and the integrity of its cell-to-cell junctions are maintained by the ionic composition of the aqueous. Replacing the aqueous with an irrigating solution during surgery can therefore be important to endothelial survival. Several researchers have tried to develop an irrigating solution that is endotheliophilic during surgery.
Edelhauser and coauthors4,5 showed that the composition of the irrigating solution is more important to endothelial survival than irrigation time. Several in vivo studies using different irrigating solutions confirmed that postoperative corneal thickness and endothelial cell count do not depend on irrigation volume and time but rather on the solution's chemical composition.6–8 Postoperative corneal swelling was significantly less with a fortified balanced salt solution (BSS Plus®) than with pure Ringer's lactate. BSS Plus contains additional dextrose, glutathione, and bicarbonate and thus mimics the physiological composition of the aqueous.
In addition to the chemical composition of the irrigating solution and the intraoperative effect of dissipating heat from the vibrating phaco needle, the temperature of the irrigating solution has also been implicated as a significant factor in postoperative outcome. Several surgeons have suggested that cooling the irrigation solution may not only maintain intraoperative mydriasis but also prevent postoperative corneal swelling.9 Thus, some surgeons use irrigating solutions that have been refrigerated, while some use solutions adapted to room temperature.
This study evaluated the effects of temperature and composition of irrigation solutions during phacoemulsification on the corneal endothelium, stromal thickness, and pupil diameter.
Patients and Methods
This study comprised 80 consecutive patients who had routine phacoemulsification and posterior chamber intraocular lens (IOL) implantation at Klinik Dardenne. Mean age of the 33 men and 47 women was 68 years (range 23 to 85 years). During surgery, patients received BSS Plus (Alcon) or Ringer's solution Salvia (Clintec Salvia) as an irrigating solution (Table 1). Both solutions were used at room temperature (23°C) or cooled in a refrigerator to 8°C. The setup involved 4 cross-classified groups of 20 patients each with the following parameters: Ringer's solution cold (8°C); Ringer's solution at room temperature (23°C); BSS Plus cold (8°C); BSS Plus at room temperature (23°C).
Patient Allocation to Treatment Group
To ensure that differences among the groups were attributable to the treatment factors (solution composition and temperature) rather than other factors, the 4 groups were designed to be homogeneous with respect to age, sex, lens hardness, corneal risk (e.g., Fuchs' dystrophy, guttata) and surgical risk factors (e.g., synechias, undilatable pupils). Homogeneity was attained using the nonrandomized allocation method of Altmann,10 which minimizes the effect of such factors, particularly in studies with limited group sizes. According to Altmann's technique, every new patient with a 50% probability is allocated to a treatment group to obtain the best balance of unrelevant factors among treatment groups.
Table 2 shows the allocation according to Altmann's minimization principle. Five of the 80 patients had with cornea guttata, 2 had undilatable pupils, and 1 had a history of iridocyclitis.
All patients had preoperative and postoperative examinations including visual acuity, refraction, intraocular pressure (IOP), and slitlamp evaluation of the anterior segment. Special attention was paid to co-existing corneal endothelial disease, anterior and posterior synechias, iridocyclitis, uveitis, and small pupils. Lens hardness was graded on a 1 to 5 scale by the surgeon in all cases (H.-R. K.).
Central corneal thickness was measured using an optical pachymeter (Haag-Streit). In cases with postoperative IOP spikes, pachymetry was measured after the IOP was normalized. All pachymetry readings were taken between 10 am and noon. Corneal endothelial photographs were taken with the specular microscope attachment to the Zeiss photo slitlamp. Corneal endothelial cells were counted on paper prints as cells counts per millimeter squared (magnification ×200).
Intraoperative pupil diameter in each eye was recorded parallax-free with the intraocular measuring spatula of Engels after the incision, phacoemulsification, irrigation/aspiration, IOL implantation, and an intracameral injection of acetylcholine at the end of surgery.
For statistical analysis, IOL type, material, and power; the viscoelastic material used; ultrasound time and intensity; the volume of the irrigating solution; and intraoperative problems or complications were noted.
After pretreatment with gentamicin and ibuprofen eyedrops, pupils were dilated with cyclopentolate 1% and phenylephrine 10%. Anesthesia was achieved by peribulbar injection of a bupivacaine 0.5% and lidocaine 2% mixture with hyaluronidase (37.5 IU). Epinephrine (1:1000) and gentamicin (80 mg/L) were routinely added to the irrigating solution.
After a temporal clear corneal incision was made, phacoemulsification was performed using a classic divide and conquer technique. Hydroxypropyl methylcellulose (Adathocel®) was used as a viscoelastic agent. All IOLs were foldable silicone or acrylic.
Regression and correlation analyses were calculated to analyze the relationships between the parameters. To examine the influence of temperature and of type of solution on endothelial cell counts and corneal thickness, a 2-factor analysis of variance (ANOVA) was used.
The 4 groups did not differ significantly with respect to age, sex, or risk factors. Intraoperative parameters such as ultrasound time, lens hardness, and irrigation volume had no influence on endothelial cell counts or pachymetry. As the 4 groups did not differ, contamination of the treatment effects with possible unrelated factors was ruled out.
Best corrected visual acuity increased from a mean of 0.27 preoperatively to a mean of 0.64 postoperatively. Neither the type of irrigation solution nor the solution's temperature had an effect on the postoperative visual function (Table 3).
A postoperative decrease in endothelial cell count proved the independence of temperature and type of solution (Figure 1).
The influence of temperature on the postoperative corneal thickness was not statistically significant (P > .05) (Figure 2). There was, however, a significant difference between the 2 solutions in corneal thickness (P ≤ .05). The mean thickness increased by 1.92% in BSS-treated eyes and by 4.23% in eyes treated with Ringer's solution.
Although the pupil was dilated by topical application of cyclopentolate 1% and phenylephrine 10% and intraocular instillation of epinephrine (1:1000) added to the irrigating solution, intraoperative miosis was a frequent problem. Pupils were most dilated when cold BSS Plus was used and least dilated when warm Ringer's solution was used. However, the changes in pupil diameter throughout surgery and among the 4 groups (i.e., temperature and type of solution) were not significant (Table 4).
Corneal endothelial safety is the guideline for developing intraocular irrigating solutions. Whereas balanced salt solutions are known to promote toxic cell alterations,11,12 certain additives such as adenosine and glutathione have positive effects on endothelial pump function.13 The ATPase-dependent pump function prevents stromal hydration because of hyperosmolarity. Dextrose is known to prevent swelling of cultured corneal epithelial cells. This was the rationale behind the development of BSS Plus, which contains oxidated glutathione and bicarbonate as well as dextrose in a balanced ionic solution.
The protective effect of BSS Plus on the corneal endothelium has been shown in several experiments. Early in vitro studies by Edelhauser and coauthors4,5 showed that bicarbonate, glutathione, and adenosine protected stromal thickness and endothelial integrity in isolated human corneas. Several clinical studies confirmed these in vitro results.6,14
These findings agree with our results. BSS Plus induced significantly less corneal swelling than Ringer's solution on the first postoperative day. However, it did not induce a difference in endothelial cell counts. In our opinion, the temporal thickening of the cornea is induced by surgical trauma and might be related to intermediate damage to the corneal endothelial cells with subsequent recovery. This is indicated by the finding that there was no long-term difference among groups in postoperative visual acuity and long-term corneal thickness. In our setup, this lack of a long-term advantage of BSS Plus over Ringer's solution may be the result of our short irrigation periods, low irrigation volumes, and relatively atraumatic surgeries. In patients with higher long-term corneal risks (hard lenses, reduced endothelial cell counts, corneal dystrophies) or in operations by less experienced surgeons, the significant but small BSS Plus advantage observed in our cases may prove a sight-preserving difference.
The second aspect of this study was the influence of irrigation temperature on corneal function and pupil diameter. Corneal temperature is well below body temperature. It decreases from the endothelium to the outer surface. The central corneal temperature is about 30.7°C to 35.0°C. Because of the cornea's avascularity, its temperature is transferred from the well-vascularized anterior uvea through the aqueous.15,16 The optimal temperature of solutions for intraocular surgery, especially phacoemulsification, is controversial.9,17 An increased temperature, which may result from obstruction of the irrigation flow, can cause intraoperative endothelial damage. Hausmann and Richard18 found an endothelial cell loss of up to 10% after an increase in aqueous temperature from 29.1°C to 33.4°C. They concluded that cooling the irrigation solution might prevent endothelial damage.
Therefore, heat generation by the phaco tip and the use of normothermal and hypothermal irrigating solutions might influence postoperative corneal results. If perfusion with a hypothermic solution is performed, the temperature within the anterior chamber quickly decreases to the level of the irrigating solution. However, anterior chamber and overall ocular temperatures reach normal level within a few minutes after cold irrigation is stopped.19 This is in agreement with unpublished data (H.-R. Koch, MD, C.F. Mathey, MD, H.J. Ensikat, PhD, M. Rössler, “Intraocular Pressure During Phacoemulsification,” film presented at the Symposium on Cataract, IOL and Refractive Surgery, Boston, Massachusetts, USA, April 1997). In vivo experiments using normothermal (23°C) and hypothermal (5°C) irrigating solutions in rabbits did not induce changes in endothelial morphology.20 Yagoubi and coauthors21 perfused the anterior chambers of rabbits and saw an increase in corneal pachymetry depending on solution composition, but no effect from the solution's temperature.
To our knowledge there are no published data on the effects of different irrigating solutions and temperatures during phacoemulsification under clinical conditions in humans. Our results agree with the unpublished data reported above, showing that a temperature lower than room temperature does not affect postoperative corneal parameters nor prevent corneal burns.
Pupil diameter during phacoemulsification was also not influenced by the cold stimulus. It was more dependent on intraoperative mechanical irritation. In our series, we were able to achieve stable pupil dilation after adequate preoperative pharmacological treatment both at cold and room temperatures.
Thus, the benefit of hypothermia during phacoemulsification remains questionable. On the other hand, we agree with the results of several authors20,21 that moderate hypothermia over a limited time does not appear to cause corneal damage. Therefore, we conclude that the ionic composition of irrigating solutions is important in the outcome of phacoemulsification and that there is no detectable effect of the irrigating solution temperature.
1. Riley MV. Transport of ions and metabolites across the corneal endothelium. In: McDevitt DS, ed, Cell Biology of the Eye. New York, NY, Academic Press, 1982; 2:53-95
2. Watsky MA, McDermott ML, Edelhauser HF. In vitro corneal endothelial permeability in rabbit and human: the effects of age, cataract surgery and diabetes. Exp Eye Res 1989; 49:751-767
3. Watsky MA, McCartney MD, McLaughlin BJ, Edelhauser HF. Corneal endothelial junctions and the effect of ouabain. Invest Ophthalmol Vis Sci 1990; 31:933-941
4. Edelhauser HF, Van Horn DL, Hyndiuk RA, Schultz RO. Intraocular irrigating solutions: their effect on the corneal endothelium. Arch Ophthalmol 1975; 93:648-657
5. Edelhauser HF, Gonnering R, Van Horn DL. Intraocular irrigating solutions: a comparative study of BSS Plus and lactated Ringer's solution. Arch Ophthalmol 1978; 96:516-520
6. Benson WE, Diamond JG, Tasman W. Intraocular irrigating solutions for pars plana vitrectomy: a prospective, randomized, double-blind study. Arch Ophthalmol 1981; 99:1013-1015
7. Matsuda M, Tano Y, Edelhauser HF. Comparison of intraocular irrigating solutions used for pars plana vitrectomy and prevention of endothelial cell loss. Jpn J Ophthalmol 1984; 28:230-238
8. Rosenfeld SI, Waltman SR, Olk RJ, Gordon M. Comparison of intraocular irrigating solutions in pars plana vitrectomy. Ophthalmology 1986; 93:109-114; discussion by HF Edelhauser, 114-115
9. Ernest PH. Corneal lip tunnel incision. J Cataract Refract Surg 1994; 20:154-157
10. Altmann DG. Practical Statistics for Medical Research. New York, NY, Chapman and Hall, 1991; 74-106
11. Harper JY Jr, Pomerat CM. In vitro observations on the behavior of conjunctival and corneal cells in relation to electrolytes. Am J Ophthalmol 1958; 46(1, pt 2):269-276
12. Merrill DL, Fleming TC, Girard LJ. The effects of physiologic balanced salt solutions and normal saline on intraocular and extraocular tissues. Am J Ophthalmol 1960; 49:895-898
13. McCarey BE, Edelhauser HF, Van Horn DL. Functional and structural changes in the corneal endothelium during in vitro perfusion. Invest Ophthalmol 1973; 12:410-417
14. Nuyts RMMA, Edelhauser HF, Holley GP. Intraocular irrigating solutions: a comparison of Hartmann's lactated Ringer's solution, BSS and BSS Plus. Graefes Arch Clin Exp Ophthalmol 1995; 233:655-661
15. Schwartz B, Feller MR. Temperature gradients in the rabbit eye. Invest Ophthalmol 1962; 1:513-521
16. Efron N, Young G, Brennan NA. Ocular surface temperature. Curr Eye Res 1989; 8:901-906
17. Koch H-R. Was ist zu tun bei intraoperativer Verbrennung des korneoskleralen Wundrandes durch den Phakotip? Auswirkungen und Behandlung abhängig von der Inzisionstechnik. Ophthalmo-Chirurgie 1992; 4:172-174
18. Hausmann N, Richard G. Investigations on diathermy for anterior capsulotomy. Invest Ophthalmol Vis Sci 1991; 32:2155-2159
19. May DR, Freedland RJ, Charles S, et al. Ocular hypothermia: anterior chamber perfusion. Br J Ophthalmol 1983; 67:808-813
20. Mannis MJ, Miller RB, Carlson EC, et al. Effect of hypothermic perfusion on corneal endothelial morphology. Br J Ophthalmol 1983; 67:804-807
21. Yagoubi MI, Armitage WJ, Diamond J, Easty DL. Effects of irrigation solutions on corneal endothelial function. Br J Ophthalmol 1994; 78:302-306