View 2: The case for primary vitrectomy

The term “primary vitrectomy for rhegmatogenous retinal detachment” implies that pars plana vitrectomy is the first surgical intervention in the treatment of this disease. In the literature, however, its definition is frequently widened: it often includes patients with rhegmatogenous retinal detachment (RRD) who have undergone either cryotherapy or photocoagulation for retinal breaks or small detachments before vitrectomy, although, strictly speaking, pars plana primary vitrectomy (PPPV) is not the first surgical intervention in these cases. The following assessment is based on the analysis of 25 publications of primary vitrectomy listed in Table 1.

Table 1.

Reports of primary vitrectomy for the treatment of rhegmatogenous retinal detachment

Author	Year	Buckle	No	Primary success	Final success	PVR	≥0.33	≥0.4	≥0.5
Escoffery17	1985	0%	29	79%	79%	7%		76%
van Effenterre18	1987	0%	60	86%	92%	0%		76%
Hakin19	1993	79%	124	64%	82%		34%
Gartry13	1993	65%	114	74%	92%	8%	35%
Girard20	1995	81%	103	74%	85%	16%			40%
Höing21	1995	0%	32	78%	94%	19%			44%
Bartz-Schmidt5	1996	100%	33	94%	100%	3%		79%
Heimann22	1996	0%	53	64%	92%	6%		41%
Yang23	1997	100%	9	89%	100%	0%		0%
El-Asrar10	1997	100%	22	100%	100%	0%		32%
Desai24	1997	100%	10	100%	100%	0%		70%
Hoerauf25	1997	83%	37	87%		8%
Sharma7	1998	100%	21	90%	90%	10%		19%
Brazitikos26	1999	36%	14	100%	100%	0%		69%
Campo27	1999	0%	275	88%	96%	6%		69%
Devenyi6	1999	100%	94	100%	100%	0%
Newman28	1999	48%	25	84%	96%	8%			48%
Oshima29	1999	51%	63	92%	100%	0%
Brazitikos30	2000		103	93%	97%	4%
Brazitikos30	2000	0%	20	85%	95%	10%
Gastaud31	2000	58%	19	84%	100%	0%
Pournaras32	2000	100%	23	92%	100%	4%		65%
Speicher33	2000	0%	78	94%	96%	5%
Miki8	2000		87	92%	100%	1%
Tanner34	2001	0%	9	89%	100%	0%		67%

PVR = proliferative vitreoretinopathy, VA = visual acuity.

ADVANTAGES OF PRIMARY VITRECTOMY

The removal of opacities in the vitreous and of capsular remnants or synechia are clear advantages of PPPV. The better intraoperative control of PPPV is supported by the high rates of intraoperative reattachment, even in very difficult cases, achieved by internal drainage and endotamponade. This is further emphasised by the low rate of intraoperative complications, the most frequently observed being iatrogenic breaks (6% of cases in series reporting this complication), and that of lens damage in phakic eyes in 3%. Surgeons nowadays have become more familiar with this technique compared to scleral buckling surgery, as the indications for vitrectomy and the total number of cases outside retinal detachment surgery have increased immensely during the past decade. Further, endoillumination, indentation, a higher magnification, wide angle viewing systems, the removal of opacities, membranes, and the unfolding of detached retina with perfluorocarbon liquids enable a better view of the pathological anatomy and an improvement in the identification of previously unseen breaks, as documented in 77 of 87 cases of previously unseen breaks. The problems associated with external drainage, such as choroidal haemorrhage, retinal incarceration, and retinal perforation, as well as that of scleral perforation during suturing of the exoplant (if PPPV is not combined with additional scleral buckling surgery) are avoided.

In the postoperative period, major drawbacks of scleral buckling surgery are avoided: only minor changes of refraction occur compared to those following scleral buckling surgery. Choroidal detachments, summarised by Ambati and Arroyo to occur in 23–44% of eyes following scleral buckling surgery,² have been reported in only three patients following PPPV. Infections, intrusions, and extrusions of episcleral buckling material complicating scleral buckling surgery are precluded completely (if no additional scleral buckling is performed) and have not been reported in the literature reviewed above. Postoperative imbalance of extraocular muscles, leading to long term diplopia occurring in 5–25% following scleral buckling surgery,³ have not been mentioned in the series summarised in Table 1.

PRESUMED ADVANTAGES OF PPPV

One of the major reasons for the increasing use of PPPV is the expectation that better anatomical and functional results are achieved with this method in more complicated forms of RRD. The combined primary success rate of the studies of PPPV reviewed was 85%. Compared to the primary success rates of larger, consecutive studies of scleral buckling surgery (75–91%) summarised by Wilkinson,⁴ this further encourages the use of PPPV, particularly as more complex situations of RRD are operated on with PPPV. PPPV is thought to be particularly successful in pseudophakic/aphakic patients, as demonstrated by the high primary success rates of 91% that seems to be significantly superior to those of scleral buckling.^4,5 In these situations, a more thorough removal of the peripheral vitreous is possible. Further, unclear hole situations and small retinal breaks are more common; and postoperative cataract formation is not a concern. Moreover, the overall final success rates of 95% with PPPV in more complex cases of RRD seem to be exceptionally good compared to the majority of reports of scleral buckling surgery (range 88–97%), and in 98% of pseudophakic and aphakic patients compared to 80–96%.⁴ One of the presumed advantages of PPPV might be that redetachments following PPPV are “easier” cases compared to failures after scleral buckling surgery. The former are often caused by a single missed/new break, and a repetition of the internal tamponade with treatment of the new break is frequently sufficient to treat the redetachment. However, for methodological reasons (for example, differences of inclusion criteria, operating techniques, follow up period, or measurement of visual acuity in various series of scleral buckling surgery and PPPV), the data provided to date are inadequate for a sound comparison of anatomical or functional results of PPPV and scleral buckling surgery. The same holds true for postoperative proliferative vitreoretinopathy (PVR) formation and macular pucker. Although various hypotheses about why PPPV should result in a lowering of postoperative PVR have been proposed (for example, removal of the vitreous with its chemotactic and mitogenic stimuli, and the “washout” of RPE cells out of the subretinal space and vitreous cavity^6,7), there still are considerable rates of postoperative PVR in 6% of all patients following PPPV. Furthermore, macular pucker was seen in 9% of patients in studies in which this complication was investigated. Finally, the supremacy of functional results following PPPV has yet to be proved, although the calculated percentage of 63% of patients with a visual acuity of 0.4 or better in series of PPPV summarised in Table 1 compares very favourably with the 39–56% of successful cases only following scleral buckling surgery.⁴ Again, the functional results following PPPV in pseudophakic/aphakic patients seem to be even more superior to scleral buckling surgery when comparing the published results. However, the need for appropriate data is underlined by two recent retrospective studies which compared PPPV to scleral buckling in more complex situations of RRD and another series including patients with flap tears only.^8,9 Neither study could demonstrate a significant advantage of PPPV over scleral buckling concerning anatomical and functional success.

DISADVANTAGES OF PPPV

The major intraoperative complication, iatrogenic breaks (although supposedly not significantly influencing the outcome of the surgery), will probably not be completely eliminated even if the greatest intraoperative care is taken. The same holds true for intraoperative damage to the lens and the postoperative increase in nuclear cataract, which was established in more than a third of phakic patients. These complications not only cause a decrease in visual acuity and a myopic shift of postoperative refraction but will cause loss of accommodation in young patients and, sooner or later, lead to the necessity of additional surgery in most patients. If similar success rates can be achieved with both surgical methods, this has to be seen not only from the patient’s point of view but also against the background of the management of surgical and financial resources. The same reasoning has to be applied regarding the costs of the procedure of PPPV itself. These are significantly higher compared to scleral buckling surgery alone, although some authors argue that PPPV is cheaper than scleral buckling surgery in the long run, because of a higher anatomical success rate and lower number of reoperations following PPPV.⁶

Analysis of the literature would further suggest that a greater number of postoperative breaks can be found following PPPV compared to scleral buckling surgery. No definite distinction can be made if a break, which is detected postoperatively, has developed de novo or has just been missed before or during the surgery. However, if one postulates that more breaks are identified during PPPV compared to scleral buckling surgery, there is no other explanation for the high rate of postoperative breaks compared to series of conventional surgery than that these breaks developed after the initial surgery. Possible mechanisms for the development of new breaks following PPPV are accidental touching of the retina during surgery that will later result in a retinal tear; new tangential forces from scar formation, especially in the region of the sclerotomies; contraction forces of the remaining vitreous cortex; formation and contraction of an epiretinal membrane; and/or continuing PVD after PPPV.^8,10,11 Finally, it is worth mentioning that four series of PPPV have identified a total of 27 patients with a long lasting rise in the intraocular pressure following the procedure.

CONTROVERSIES ABOUT THE CURRENT USE OF PRIMARY VITRECTOMY

PPPV has gained a tremendous popularity in recent years and in some centres in the United Kingdom, PPPV is the method of choice in up to 63% of all patients with RRD.¹² The major advantage of PPPV that has been established to date and the main reason for its current popularity is that it lowers the intraoperative complication rate and enables better control in more challenging situations of RRD.¹³ By changing the operative method to PPPV in such situations, the surgeon avoids the hazards of scleral perforation and external drainage; bad visualisation of the retinal periphery; a very soft eyeball during surgery; “fish-mouthing” in awkward breaks; insufficient elimination of vitreous traction on the break; or very difficult placement of scleral buckles in complicated break arrangements. During surgery, it is possible to work in a pressure stable environment, to clear media opacities, have an enhanced view to look and search for peripheral breaks, and to reattach the retina in almost every case, no matter how complicated the preoperative anatomical situation might have been. In “unclear hole situations,” it is unnecessary to search for breaks for several hours or days, as proposed, because it is highly likely that these breaks will be detected during the operation. The patient will leave the operating table with a reattached retina supported by an internal tamponade and with the belief that all breaks have been identified and treated. This leads to more confidence for the surgeon compared to the worries one might have because of either patent breaks with residual detachment, which should hopefully be absorbed during the following the early postoperative days; “blind” circular encirclements which, with a bit of luck, may or may not work; or extensive external cryotherapy. These advantages of PPPV are increasingly often willingly “traded in” for additional cataract surgery in phakic patients, which nowadays does not present a major surgical problem for the vitrectomised eye, and is even more easily accepted in pseudophakic eyes.

Scleral buckling surgery might be a straightforward, highly successful procedure in simple cases with good visualisation of the retinal situation, and it is doubtful that PPPV will gain much popularity in this field, even in pseudophakic cases. However, in more complicated situations of RRD, the surgical procedure of scleral buckling surgery very quickly becomes much more challenging and the expressions of the “art” of scleral buckling surgery and the “surgical nerve” required are sometimes used.¹⁴ If PPPV simplifies the treatment of more complicated retinal detachments and enables surgeons to manage more complex situations of RRD at an earlier stage, the question is whether it would be justified to expose training surgeons and patients to the longer learning curve of mastering scleral buckling in complicated situations of RRD.

In addition, the very high success rates reported for non-drainage procedures¹⁵ are not comparable to other reports of RRD surgery, as a different selection of patients and dissimilar definition of primary success and redetachment is used in reports of non-drainage surgery. With the “minimal surgery” approach or conventional scleral buckling surgery, even most recent series¹⁶ fail to reproduce the high success rates of selected reports of non-drainage scleral buckling surgery.^14,15 Regarding the cost of the surgery, market forces and the policy of cutting down on expenses in the medical sector will result in favouring the cheaper one of two methods if similar results are achieved. Further, despite the increasing numbers of vitrectomies for other indications, RRD will remain one of the major indications for vitreoretinal surgery, as no measures can be undertaken to prevent the disease as yet and the absolute numbers of patients with RRD is rising owing to the increasing age of the population and number of cataract surgeries performed.

The techniques of scleral buckling surgery have been established for decades and no further enhancements have been introduced in recent years. This is reflected by the fact that the results published in recent series of scleral buckling surgery¹⁶ are not superior to the results achieved 20 years earlier.⁴ In contrast, the techniques of PPPV are constantly refined and with the prospects of a medical vitreolysis, and the uncomplicated installation of as yet hypothetical drugs that might lower postoperative PVR, additional advantages of PPPV are to be expected in the future. As indicated by numerous authors previously, however, only a prospective, multicentre randomised trial will be able to provide sufficient data to define the exact role of PPPV in the treatment of RRD. Such a trial, the “Scleral Buckling versus Primary Vitrectomy in Rhegmatogenous Retinal Detachment Study (SPR Study),” is currently under way in 25 centres in Europe.¹