Introduction

The treatment of patients with primary rhegmatogenous retinal detachment (RRD) has undergone considerable change over recent decades. In Europe, there is a definite trend towards primary pars plana vitrectomy (PPPV) as the method of first choice [4, 7, 16, 20–23, 29]. However, in most European centres, scleral buckling surgery (SBS) remains the method of choice in uncomplicated retinal situations, i.e. single breaks and/or a limited retinal detachment. In contrast, PPPV is indicated in complicated situations, i.e. vitreous haemorrhage/opacity, proliferative vitreoretinopathy (PVR), or breaks at the posterior pole [3, 5, 8–11, 25, 28, 31, 35].

Despite these clear indications, there is a large group of RRDs with “medium severity” where the choice of treatment method is not clear-cut. The “Scleral buckling versus primary vitrectomy in rhegmatogenous retinal detachment study (SPR Study)”, a prospective, randomised multicentre clinical trial [12], addresses the question of whether SBS or PPPV are associated with a better functional and anatomical outcome in treating such medium-severe RRDs.

As part of the SPR Study, recruitment lists were employed to evaluate the strength of the eligibility criteria and to record epidemiological data. In this study, we analysed the data of the recruitment lists from a 1-year period. The goals of this evaluation are to estimate the prevalence of medium-severe RRDs as defined by the SPR Study inclusion criteria and to characterise the medium-severe RRDs in terms of anatomical details.

Methods

The SPR Study inclusion criteria

Medium-severe RRD as defined by the SPR Study inclusion criteria [12] is characterised as primary RRD not treatable with a single 7.5×2.75 mm Silastic sponge. This includes aphakic/pseudophakic patients with an unclear hole situation.

Basic data provided by study centres (from questionnaire)

All patients suffering from primary RRD had to be documented prospectively by the actively participating study centres using a standardised questionnaire. The recruitment lists of a 1-year period, the year 2000, were evaluated in this analysis (termed “SPR recruitment study”). The forms contained a coloured fundus drawing and retrieved information regarding possible reasons for exclusion from the SPR Study, the choice of the surgical procedure “(to be) performed” and the hole situation. Details are given in Table 1. Patients suffering from RRD following previous intraocular surgery (except uncomplicated cataract surgery), angioproliferative retinal diseases, re-detachment or penetrating injury were excluded from this analysis.

Table 1 Questionnaire data given by the study centres
Full size table

Reviewers’ evaluation (from the drawing)

Potential inclusion into the SPR Study was evaluated by review of the fundus drawings. Lens status was taken into account regarding the eligibility of pseudo-/aphakic patients with unclear hole situation (Fig. 1).

Fig. 1
figure 1

Examples for retinal break evaluation a Break size: I=smaller than 1/2 clock hour; II=1/2 to 1 clock hour; III=more than 1 clock hour. b Location of retinal break. I=anterior; II=intermediate; III=central; IV=macular hole. c Shape of retinal break: I=round; II=horseshoe; III=cleft; IV=unusual. d Traction on the operculum

Full size image

Three experienced vitreoretinal surgeons (S.W., H.H., N.F.) independently evaluated all drawings. In cases in which the reviewers’ results differed, a majority vote (2:1) was accepted. Questionable situations and indeterminate cases were discussed and a consensus was reached. None of the additional information on the recruitment questionnaire was used to decide on potential inclusion into or exclusion from the SPR Study. The questions the reviewers had to answer are depicted in Table 2.

Table 2 Reviewer’s questions concerning the coloured fundus drawings
Full size table

Statistics and ethics

Statistical analysis was performed using SAS (the SAS System, Release 9.1.3 SP 2; SAS Institute, Cary, NC, USA) on Windows® 2000 SP 4 (Microsoft, Redmond, CA, USA). Values are given as frequency and percentage for qualitative, and mean and standard deviation for quantitative parameters respectively. Exact 95%-confidence intervals were calculated using SAS PROC FREQ (exact BINOMIAL) for the rates of medium-severe RRD.

The Ethics Committee approval of each participating centre was obtained.

Results

Basic data

The following 13 centres provided a total of 1115 cases with fundus drawings for the recruitment list in the year 2000 (in alphabetical order): Aachen, Berlin, Bremen, Cologne, Frankfurt, Freiburg, Heidelberg, Homburg, Leipzig, Liverpool, Lübeck, Regensburg, Sulzbach. The number of patients screened per centre ranged from 7 (0.6%) to 258 (23.1%) (mean 85.8, SD 73). Seventy-two of these patients (6.5%) were included in the SPR Study within the year 2000 (Fig. 2).

Fig. 2
figure 2

Flowchart of all fundus drawings of the year 2000. Eight drawings were not analysable, two drawings questionable for inclusion in the SPR Study

Full size image

Five hundred and nine (45.7%) patients were female, 597 (53.6%) were male; no information on gender was provided in 9 cases (0.8%). In 595 (53.4%) of all patients, the retina of the right eye was detached, in 520 cases (46.6%) the left eye was affected. Seven hundred and sixty patients (68.2%) were phakic, while 343 eyes (30.8%) were recorded as aphakic or pseudophakic; the lens status was missing in 12 cases (1.1%).

Myopia of more than − 7.0 dioptres was present in 143 eyes (12.8%), central retinal breaks in 49 (4.4%), PVR B/C in 140 (12.6%), other eye diseases in 248 (22.2%) and severe systemic diseases in 329 (29.5%). Ten patients (0.9%) were younger than 18 years and 2 patients (0.2%) were pregnant. Seventy-two patients (6.5%) were not willing to participate, and in 18 cases (1.6%) no study surgeon was available (multiple items could be stated positive).

Medium-severe RRD: prevalence and evaluation

The quality of the drawings sufficed for assessment in 1,107 cases (99.3%). In 8 drawings, no clear decision on SPR eligibility could be made because of unclear or non-standardised signs.

In assessing whether or not a fundus drawing fulfilled the inclusion criteria, the three reviewers agreed in 818 (73.4%) of all evaluated drawings. Two hundred and sixty-seven cases (24.0%) required 2:1 agreement, and 30 (2.7%) of those controversial cases had to be re-evaluated and put to a joint vote, thus allowing proper evaluation in 1107 patients.

Overall, a total of 312 cases (28.2%; 95% CI 25.6–30.9%) were voted to be eligible for SPR inclusion from the fundus drawing (Fig. 2). 249 (79.8%) of those 312 patients were excluded from SPR Study participation because of additional information given on the questionnaires (SPR Study exclusion criteria).

Anatomical characterisation in medium-severe RRD

Differences between medium-severe and all other RRDs are found in the frequency of PVR stage B/C, bullous RRD and the mean number of retinal breaks (Table 3).

Table 3 Retinal situation in the SPR-possible group compared to all other drawings (n=1115)
Full size table

Analysis of the hole situation revealed a total of 1864 retinal holes. The majority of those were small (less than 1/2 clock hour; n=1609, 86.3%), showed a horseshoe shape (n=1272, 68.2%) and exerted no significant traction on the flap (n=1840, 98.7%). The holes were mostly located anterior to the equator (n=1501/80.5%) and between 10 and 2 o’clock (n=1126, 60.4%).

Discussion

Since PPPV was established in 1971 by Machemer et al. [19], SBS and PPPV have been competing methods in treating retinal detachment. Numerous studies have been published recommending one of the two methods [1, 4, 13, 14, 20, 24, 26, 27, 29, 30, 32]. We could identify no consensus among several editorial comments on this issue [6, 15–18, 33, 34]. In Europe, SBS still seems to be the first choice in the majority of RRD cases, although there has been a definite trend towards PPPV in recent years [4, 7, 16, 20–23, 29]. Furthermore, there is a wide variety in the choice of operating method in comparable pre-operative situations. The SPR Study was initiated to establish an evidence-based approach to this problem and to investigate whether one of the two methods is superior in more complicated (medium-severe) RRD cases. To judge the value of the SPR Study results, one needs to know the frequency of medium-severe RRD as defined by the SPR inclusion criteria. To answer this question, we evaluated the SPR Study’s recruitment lists from a 1-year period. The main goal of this analysis is to estimate the prevalence of medium-severe RRD (SPR Study eligible) as defined by the SPR Study inclusion criteria. In addition, the detailed anatomical situation of medium-severe RRD is investigated.

Our results correspond well with the findings of other multicentre trials [2] (Table 4).

Table 4 Pre-operative findings in eyes with RRD: comparison of different studies
Full size table

Evaluation and prevalence of medium-severe RRDs

Medium-severe RRD as defined by the SPR Study criteria is a frequent finding in patients with primary RRD. Nearly one-third (28%) of all drawings of RRD cases matched the inclusion criteria. Eighty per cent of those patients could not be included in the study because of pre-existing ophthalmological (other than primary RRD) or systemic diseases. However, those exclusion criteria would not influence a surgeon’s decision outside the context of a study.

As expected, the three reviewers’ evaluations of the fundus drawings were not always consistent. One-quarter of all drawings required a 2:1 decision on SPR eligibility, and only 2.7% had to be conferred upon together. It is obvious from these data that the surgeons differ in their clinical assessment of retinal detachment. The variation in evaluating the retinal situation and the eligibility of patients may show surgeon-based bias concerning the first method of choice for surgical repair, especially for complex RRD, resulting in the exclusion of eligible patients from the SPR Study.

Characterisation of the medium-severe RRDs

In addition to the clearly defined inclusion criteria for the SPR Study, one goal of the present study was to work out details to characterise the retinal situation of eyes assessed to have medium-severe RRD and to find retinal criteria to differentiate between medium-severe and other RRDs. Overall, the mean area of detached retina we observed was higher than that found in other studies (Table 4). However, the mean area of detached retina in the well-defined retinal situation of medium-severe RRD is only a little greater than in all other eyes of the present study and as such is no valid criterion for differentiating between medium-severe and non-medium-severe RRD. In contrast, the mean number of retinal holes in the SPR-possible group was higher. SPR-possible eyes had a mean number of 2.6 holes (SD 2.4), exceeding the overall mean by more than 1 (mean 1.7, SD 1.7). These findings were not surprising, as more than the half of all eyes not assessed to have an SPR-eligible situation from the drawing were simple RRDs and presented a “single sponge possible” situation with a mean of 1.6 (SD 1.3) retinal holes.

Unclear hole situation, macular attachment and vitreous haemorrhage/opacity were comparable in SPR-eligible patients and all others. Overall, the typical medium-severe RRD has more than 2 retinal holes and a mean area of detached retina of more than 5 clock hours.

In summary, the SPR recruitment study underlines that RRD of medium severity represents a large and therefore clinically relevant subgroup of RRD patients. Additionally, the recruitment list provides a valuable overview of patients with RRD.

We therefore believe that the SPR Study results are necessary to identify the adequate therapy in the routine management of patients with RRD.