Face-down positioning or posturing after macular hole surgery

Macular holes cause severe impairment of sight. With the aim of improving the outcome of surgery for macular holes, particularly larger macular holes (those measuring over 400 μm), a variable period of face-down positioning may be advised. This review is an update of a Cochrane Review published in 2011.

To evaluate the effect of postoperative face-down positioning on the outcome of surgery for macular hole.

We searched the Cochrane Central Register of Controlled Trials (CENTRAL; 2022, Issue 5), which contains the Cochrane Eyes and Vision Trials Register, Ovid MEDLINE, Ovid Embase, the ISRCTN registry, ClinicalTrials.gov, and the WHO International Clinical Trials Registry Platform. There were no date or language restrictions in the electronic searches for trials. We last searched the electronic databases on 25 May 2022.

We included randomised controlled trials (RCTs) in which postoperative face-down positioning was compared to no face-down positioning following surgery for macular holes. The primary outcome of interest was closure of the macular hole. Other outcomes of interest included visual outcomes, quality of life outcomes, and the occurrence of adverse events. Pairs of review authors independently selected studies for inclusion, extracted data, assessed risk of bias, and evaluated the certainty of evidence using GRADE.

We used standard methodological procedures expected by Cochrane. We analysed dichotomous data as risk ratios (RRs), and continuous data as mean differences (MDs), with 95% confidence intervals (CI). The unit of analysis was eyes.

We included eight studies allocating a total of 709 eyes (699 participants). There was heterogeneity in study design, including the control group treatment (from no positioning to strict maintenance of other 'face-forward' postures) and surgical procedures (with or without inner limiting membrane peeling, with or without cataract surgery). There were also different durations of positioning, with two studies using 3 days, two studies using 5 days, and three studies using 10 days of face-down positioning. Whilst the overall risk of bias was low, all included studies were judged to be at high or unclear risk of bias due to absence of assessment of adherence to the 'prescribed' intervention of face-down positioning or posturing. The primary outcome of successful anatomical hole closure at one to six months following surgery was reported in 95 of every 100 eyes of participants advised to position face-down for at least three days after surgery, and in 85 of every 100 eyes of participants not advised to position face-down (RR 1.05, 95% CI 0.99 to 1.12, 709 eyes, 8 studies, I² = 44%). Amongst the 327 eyes of participants with macular holes of at least 400 μm, hole closure was noted in 94 of every 100 eyes of participants advised to position face-down, and in 84 of every 100 eyes of participants not advised to position face-down (RR 1.08, 95% CI 0.93 to 1.26, 5 studies, I² = 62%). Amongst the 129 eyes of participants with macular holes of less than 400 μm, hole closure was noted in 100 of every 100 eyes of participants advised to position face-down, and in 96 of every 100 eyes of participants not advised to position face-down (RR 1.03, CI 0.97 to 1.11, 4 studies, I² = 0%). The certainty of the evidence was low, downgraded for imprecision (CIs including no effect) and study design limitations (with different durations of face-down posturing used in the absence of a dose-response gradient, and limitations in measuring the exposure). Meta-analysis of visual acuity data was challenging given the use of different definitions of postoperative visual outcome across studies. Three studies reported findings by gain in Early Treatment Diabetic Retinopathy Study (ETDRS) letters (MD 2.04, 95% CI -0.01 to 4.09, very low-certainty evidence). Meta-analyses of quality of life data were not possible because of inconsistency in outcome metrics across studies. One study reported no difference between groups in quality of life, as reported on a validated quality of life metric scale (the National Eye Institute Visual Function Questionnaire - 25 (NEI VFQ-25), between face-down positioning for five days and non-face-down positioning (median NEI VFQ-25 score was 89 (interquartile range (IQR) 76 to 94) in the face-down group versus 87 (IQR 73 to 93) in the non-face-down group (adjusted mean difference on a logistic scale 0.02, 95% CI -0.03 to 0.07, P = 0.41)). Two studies reported increased ease of positioning and less pain in non-face-down positioning groups on non-validated 0-to-10-point visual analogue scores. On an ease-of-positioning score running from 0 (very difficult) to 10 (very easy), there were consistent reports of the discomfort associated with face-down positioning: the median participant-reported ease-of-positioning score was 6 (IQR 4 to 8) in those undergoing 5 days of face-down positioning versus 9 (IQR 7 to 10) in the comparator group (P = 0.01). On a pain score with 0 being pain-free and 10 being in severe pain, mean pain score was 6.52 ± 2.48 in the face-down positioning group versus 2.53 ± 2.6 in the non-face-down positioning group. The adverse event of postoperative nerve compression occurred in less than 1 in every 100 (3 per 1000) participants advised to position face-down, and 0 in every 100 participants not advised to position face-down (699 participants, 8 studies, moderate-certainty evidence).

We identified eight RCTs evaluating face-down positioning following surgery for macular hole. The included studies were not all directly comparable due to differences in the surgical techniques used and the durations of postoperative positioning advised. Low-certainty evidence suggests that face-down positioning may have little or no effect on macular hole closure after surgery. Face-down positioning is a low-risk intervention, with serious adverse events affecting fewer than 1 in 300 people. We suggest that any future trials focus on patients with larger macular holes, with interventions and outcome measures used in previous trials (i.e. with inner limiting membrane peeling, positioning durations of three to five days, and validated quality of life metrics) to allow future meta-analyses to determine any effect with greater precision and confidence.