Influence of atmospheric pressure on postoperative decompensation of intraocular pressure in perfluorethane‐filled eyes after vitrectomy

Abstract

Editor, Besides variation of atmospheric pressure because of altitude, variations of atmospheric pressure because of weather conditions could be a reason for intraocular pressure (IOP) decompensation in perfluorethane (C2F6)-filled eyes after pars plana vitrectomy (PPV). A sudden lowering in atmospheric pressure of about 20 hPa (15 mmHg) would result in an IOP increase of 15 mmHg, which would be clinically relevant. Included were 266 eyes of 253 patients (65 ± 11 years) that underwent PPV with 14% C2F6 tamponade. The following subgroups were defined for analysis: in 139 eyes (52.3%), no combined or previous cataract surgery was conducted, and no glaucoma was present (Ph); in 106 eyes (39.8%), combined cataract surgery was conducted without the presence of glaucoma (Ps). In seven eyes (2.6%), no cataract surgery was conducted but glaucoma was present (PhG); in 14 eyes (5.3%), combined cataract surgery was conducted in the presence of glaucoma (PsG). Statistical significance of the differences between subgroups was evaluated using analysis of variance, followed by Scheffè’s method. Only phakic eyes without prior surgery were included. Eyes that experienced trauma or eyes with different tamponade were excluded. PPV was conducted as 20-gauge vitrectomy in general anaesthesia. IOP was determined preoperatively and at the first postoperative day. Data of atmospheric pressure were provided by the Leibnitz Institute of Marine Sciences, Kiel, Germany. On the first postoperative day, the gas filling was approximately 80% in all eyes. 118 patients (44.4%) experienced an IOP increase of 10 mmHg or more. Preoperatively, the mean IOP was 14.5 ± 2.9 mmHg, postoperatively it was 25.1 ± 9.3 mmHg. Patients with glaucoma displayed a higher base line level of IOP, which was statistically significant only in the group PhG compared with Ph (p = 0.003) and Ps (p = 0.009). The difference in IOP increase, however, did not reach statistical significance between any of the subgroups (p = 0.37; p = 0.97; p = 0.81). In the combined patient collective, a small but statistically significant positive correlation of the preoperative and postoperative IOP (Pearson correlation coefficient R = 0.2; p = 0.001) was found, similar to the results of others (Wong et al. 2011). No statistical significant correlation between the differences in pre- and postoperative IOP and the differences in pre- and postoperative atmospheric pressure could be found (R = 0.03; p = 0.6). A subgroup of 41 patients who experienced a postoperative drop in atmospheric pressure of at least 10 hPa (7.5 mmHg) did not display any statistically significant correlation between pre- and postoperative atmospheric pressure and IOP, either (R = 0.07; p = 0.64; Fig. 1). Correlation between pre- and first-day postoperative IOP difference and pre- and first-day postoperative atmospheric pressure difference in 41 patients, which experienced a drop of the atmospheric pressure of ≥10 hPA. During flight ascent in high altitudes, the cabin pressure can decrease about 200 mmHg in a few minutes (Dieckert et al. 1986), whereas meteorological alterations of the atmospheric pressure change slower and only in a magnitude of 20–30 mmHg. Nevertheless, in a gas-filled sphere with rigid walls, even a slow decrease in the ambient air pressure would result in an increase in IOP. However, such a model sphere does not reflect the situation in the eye correctly. The fact that even an alteration of 200 mmHg of the surrounding atmospheric pressure in a time frame of 1 hr results in a much lower increase in IOP can most likely be attributed to the existence of compensatory mechanisms, such as increased aqueous humour drain, scleral expansion and choroid compression as described by Lincoff (Lincoff et al. 1989). Accordingly, during flights (Kokame & Ing 1994) or travelling in high altitude (Ferrini et al. 2010), a transient IOP decompensation in gas-filled eyes can be seen, but no influence of the meteorological variation in the atmospheric pressure could be found in this study.