Intraocular pressure changes following 20G pars-plana vitrectomy

Acta Ophthalmologica 2012

Intraocular pressure changes following 20G pars-plana vitrectomy Carsten Framme,1 Susanne Klotz,2 Ute E.K. Wolf-Schnurrbusch,1 Peter Wiedemann2 and Sebastian Wolf1 1

Department of Ophthalmology, Inselspital, University of Bern, Freiburgstrasse, Bern, Switzerland 2 Clinic and Polyclinic for Ophthalmology, University Hospital of Leipzig, Liebigstr. 10-14, Leipzig, Deutschland

ABSTRACT. Purpose: To observe the excursions of short-term intraocular pressure (IOP) after 20-G pars-plana vitrectomy (ppV). Material and methods: In a prospective study, 851 patients (age: mean 63 ± 15 years) underwent unilateral ppV for various vitreoretinal diseases using different endotamponades [Balanced Salt Solution (BSS) 33.1%, Air 7.2%, SF6 33.6%, silicon oil 5000 cst 26.1%]. Intraocular pressure was measured in all patients before and at 3, 6, 24 and 48 hr after surgery. Survival analysis was performed to determine the cumulative hazard of IOP changes depending on endotamponade and time point after ppV (Log-Rang - Mantel Cox; p < 0.0001). Results: At baseline, IOP ranged from 0 to 50 mmHg (mean IOP: 15.3 ± 5.3 mmHg). Mean IOP after surgery revealed a slight elevation (3 hr: 16.5 ± 11.0 mmHg; 6 hr 16.9 ± 9.8 mmHg; 24 hr 19.7 ± 8.0 mmHg; 48 hr 17.3 ± 6.2 mmHg; range: 0–64 mmHg). Silicon oil filling revealed highest mean values at already 3 hr after surgery (21.8 mmHg). Also, BSS filling showed a peak after 3 hr; however, mean values were lower. Equivalent high IOP values as for silicon oil tamponade were found for gas filling; however, maximal peak was reached after 24 hr but not after 3 hr post-treatment. The cumulative hazard in all patients to reach IOP ‡ 30 mmHg after 24 hr was 23.9%; (IOP ‡ 40 mmHg = 8.2%). Herein, oil filling revealed highest risk at all time points after surgery. The risk of suffering from IOP < 5 mmHg lasting longer than 6 hr was only 1.2% after 20 G vitrectomy. Conclusion: Intraocular pressure measurements after ppV are important to prevent unintentional high IOP, especially within the early phase (3 hr posttreatment) in eyes with silicon oil filling. Gas filling leads to prolonged IOP increase (24 hr post-treatment). Long-lasting hypotony (‡6 hr) is very rare after 20G vitrectomy. Key words: 20 gauge – hypotony – intraocular pressure – IOP increase – pars plana vitrectomy

Acta Ophthalmol. 2012: 90: 744–749 ª 2011 The Authors Acta Ophthalmologica ª 2011 Acta Ophthalmologica Scandinavica Foundation

doi: 10.1111/j.1755-3768.2011.02251.x

744

Introduction Elevated intraocular pressure (IOP) is a potential risk factor for optic nerve damage. Such elevated IOP has been reported in a considerable amount of eyes after eye surgeries of various techniques (Tranos et al. 2004). Because the consequences of chronic glaucoma leading to visual field loss are very well known, the clinical significance of acutely raised IOP as after surgery is less certain but frequently reported (Podolsky & Ritch 1981; Desai et al. 1997; Tranos et al. 2004). It is confirmed that there is a retinal autoregulatory capability at IOP increases, which might be able to reduce the perfusion pressure, and that this autoregulation protects against transient rises of the IOP (Alm and Bill, 1972,; Tranos et al. 2004). Early experiments in primates by Geijer & Bill (1979) could show that the blood flow in the optic disc and the retina was equivalently sensitive to IOP increase and that effective autoregulation was present for 5–7 hr at perfusion pressures above 50 mmHg. However, extrapolation of such results to human conditions should be carried out carefully, because perfusion pressure is much higher than in primates (Tranos et al. 2004). Doppler flowmetry could confirm efficient autoregulation of the optic nerve head also in humans

Acta Ophthalmologica 2012

(Pillunat et al. 1997). Tissue damage caused by increased IOP seems to correlate with the relative perfusion pressure but not with the absolute level of IOP (Radius & Anderson 1981). Derived from such a significant amount of experimental studies in animals, there is finally sufficient evidence for possible IOP-induced damage of the optic nerve, especially if regarding the doppler-flowmetric findings of Pillunat et al. (1997) showing that the vessel autoregulation might be locally absent in certain individuals. This could also explain why there are a noteworthy amount of patients suffering from optic disc atrophy of ‘unknown origin’, especially after pars-plana vitrectomy (ppV) with silicon oil filling (La Cour et al. 2010). However, despite the fact that shortterm IOP rise is frequently not thought to be associated with permanent optic nerve damage, even in a short-term postoperative period several retinal surgeons might remember cases of permanent reduced visual acuity or blindness without any iatrogenic intraoperative damage to retina or optic disc but because of unrecognized highly elevated IOP during the first night after surgery. Such findings could underline the described absence of necessary vessel autoregulation in certain individuals (Pillunat et al. 1997) leading to permanent visual loss even after short-term but significant IOP increase after vitrectomy using gas or silicon oil. Regarding outpatient vitrectomy procedures, especially without IOP measurement possibilities during the night, it was our aim to evaluate the potential risk of suffering from elevated IOP in a large number of patients undergoing vitrectomies for different retinal or vitreal pathologies. Therefore, we performed a study to evaluate the IOP response after 20-G ppV as well as to document the incidence, timing and severity of IOP increase and hypotony in the immediate postoperative period.

Material and Methods Patients who underwent 20-gauge pars plana vitrectomy were recruited for this prospective study over a time period of 12 months. Inclusion criteria were age older than 18 years and the willingness to participate in the study. Only one eye of each patient and only

the first vitrectomy during the recruitment phase was included into the study. Exclusion criteria were the participation in another study. All patients underwent standard 20-gauge ppV using a three-port technique. Depending on the intraoperative situation, additional procedures were performed during ppV. These included encircling band, pan retinal endolaser photocoagulation, kryo- or laserretinopexy, retinektomy, and endotamponades. Endotaponades included BSS, air, SF6 gas and silicone oil (5000 cSt). Operations ended up with normal or low IOP. If Silicone oil was used as endotaponade, silicon oil was directly exchanged against perfluordecaline in all eyes that needed silicone oil endotamponade. Silicon oil instillation was performed in aphakic patients. The surgical technique was considered a complete filling of the anterior chamber with silicon oil and combined Ando iridectomy. Intraocular pressure was then measured by Schioetz method aiming to 2 mmHg before the end of operation. Because of the low IOP at the end of the procedure, water replaces anterior oil filling within 1 day, thus preventing oil overfill. In cases of silicon oil removal, a three-time air-water exchange was performed to sufficiently remove all silicon remnants best as possible. In phakic patients, Schioetz measurement was aimed to be 15 mmHg. Patients with macular hole surgery were asked to keep face-down positioning. Assessment of IOP took place 1 day before surgery (baseline) and at 3, 6, 24 and 48 hr after surgery using the standard slit-lamp adapted applanatoric Goldmann measurement technique. Eyes with increased IOP > 30 mmHg at any postoperative assessment were treated with topical antiglaucomatous medication or systemic acetazolamid to lower IOP. Therefore, mean IOP values later than 3 hr after surgery could be influenced towards lower IOPs. Therefore, the statistical analysis was primarily limited to a survival analysis. The survival analysis was performed to determine the cumulative hazard of IOP changes depending on endotamponade and time point after ppV using GraphPadPrism 5 software (GraphPad software Inc., La Jolla, CA, USA). The specific statistical analysis tool used herein was the Log-Rang (Mantel Cox) – test.

The research followed the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board. Informed consent was obtained from each subject after the explanation of the nature and possible risks of our study.

Results This prospective study included 851 eyes of 851 patients (466 women and 385 men) with a mean age of 63 ± 15 years (range 18–95 years). All patients underwent vitrectomy for the following underlying pathologies: rhegmatogenous retinal detachment (n = 205), proliferative vitreoretinopathy (PVR) retinal detachment (n = 126), diabetic retinopathy (n = 182), macular hole or macular pucker (n = 180); other pathologies as, e.g. trauma or endophthalmitis (n = 36), and finally surgery for silicon oil removal (n = 122). Additional procedures were an encircling band in 209 eyes, panretinal endolaser photocoagulation in 225 eyes, and kryo- or laserretinopexy in 206 eyes. At the end of the operation, 282 eyes were filled with BSS, 61 eyes with air, 286 eyes with SF6 gas, and 222 eyes with silicone oil (5000 cSt). At baseline, IOP ranged from 0 to 50 mmHg (mean IOP: 15.3 ± 5.3 mmHg). A total of 11 eyes had IOP < 5 mmHg at baseline. Diagnosis of these 11 eyes included endophthalmitis, open globe trauma and retinal detachment. All eyes with increase IOP > 30 mmHg at baseline (n = 20) had secondary glaucoma after silicon oil endotamponade for previous complicated retinal detachment. Regarding the mean IOP after surgery, a slight elevation (3 hr: 16.5 ± 11.0 mmHg; 6 hr 16.9 ± 9.8 mmHg; 24 hr 19.7 ± 8.0 mmHg; 48 hr 17.3 ± 6.2 mmHg) was observable (Fig. 1). Thus, the mean IOP was within normal IOP range at all time points. However, the range for the IOP measurements was large (0–64 mmHg) at all time points. If considering the different types of endotamponade used at the end of vitrectomy, highest IOP was found for silicon oil instillation reaching a mean value of 21.8 mmHg (baseline 14.7 mmHg) already only 3 hr after operation (Fig. 2). This IOP was relatively stable over the time course of 48 hr ending with 20.6 mmHg. Also

745

Acta Ophthalmologica 2012

Fig. 1. Intraocular pressure in all eyes within the first 48 hr after 20G-pars-plana vitrectomy. Highest mean pressure was observed 24 hr after treatment.

Fig. 3. Cumulative hazard for intraocular pressure (IOP) ‡30 mmHg in all patients. After 24 hr, 22% of all patients are in risk of increased IOP.

Fig. 2. Postoperative intraocular pressure (IOP) depending on endotemponade. Intraocular pressure increase after silicon oil tamponade increased within 3 hr post-treatment and presented as being stable whereas IOP increase after gas filling showed a peak at 24 hr post-treatment.

Fig. 4. Cumulative hazard for intraocular pressure (IOP) ‡30 mmHg in dependence of endotamponade. Silicon oil after 3 hr and gas filling after 24 hr post-treatment showed highest risk for increased IOP.

BSS filling showed highest peak 3 hr after operation but at a slightly lower level (20.0 mmHg) and after 48 hr normal IOP values were already reached (15.1 mmHg). Regarding gas filling, IOP first decreased after 3 hr from 14.6 to 12.3 mmHg but then reached a maximum peak with 22.2 mmHg delayed after 24 hr even above those values after silicon oil filling. Patients undergoing oil removal showed highest IOPs at baseline (mean: 18.6 mmHg) dropping to 10.7 mmHg after 6 hr and highest values after 24 hr (mean: 14.7 mmHg) but always within normal range of IOP (Fig. 2). The cumulative hazard in all patients to reach IOP ‡ 30 mmHg was 10.9% after 3 hr, 15.6% after 6 hr, 23.9% after 24 hr and finally 26.0% after 48 hr postsurgery (Fig. 3). Extracting these results to the different endotamponades used, for silicon oil tamponade this cumulative hazard was 24.3% already after 3 hr increasing continuously to 44.6% after 48 hr (Fig. 4). The risk for gas-filled eyes to exceed IOP ‡ 30 mmHg was low

within the first 6 hr after surgery (5.9%), but reached values of 23.1– 24.5% at time points 24–48 hr. Eyes receiving air filling showed lowest risks of suffering from IOP ‡ 30 mmHg (11.5% after 48 hr); (Fig. 4). BSS filling after ppV revealed such a risk in between the values of silicon oil filling and gas filling for the first 6 hr (up to 14.2%) and finally a risk of values in between gas- and air-filling (stabilization at about 16.0%). The cumulative hazard in all patients to reach IOP ‡ 40 mmHg was 3.8% after 3 hr, 5.5% after 6 hr, 8.2% after 24 hr and finally 8.7% after 48 hr postsurgery (Fig. 5). Again extracting these results to the different endotamponades used, for silicon oil tamponade, this cumulative hazard was highest with 8.1% already after 3 hr increasing continuously to 15.8% after 48 hr (Fig. 6). The risk for gasfilled eyes to exceed IOP ‡ 40 mmHg was very low within the first 6 hr after surgery (0.7%) and reached highest values of 5.9% at time points 24– 48 hr. Eyes receiving air filling again showed lowest risks of suffering from

746

Fig. 5. Cumulative hazard for intraocular pressure (IOP) ‡40 mmHg in all patients. After 24 hr, 8% of all patients are at risk of increased IOP.

Fig. 6. Cumulative hazard for intraocular pressure (IOP) ‡40 mmHg in dependence of endotamponade. Silicon oil filling revealed highest risk for this increase of IOP (15% of all patients after 48 hr post-treatment).

IOP ‡ 40 mmHg (3.3% within the first 48 hr post-treatment); (Fig. 6). For such a risk, interestingly, BSS filling revealed a higher risk than airand gas-filling (up to 7.1% 24 hr after surgery). The cumulative hazard for hypotony