Prognostic Determinants
Based on the final outcome of the multiple logistic regression analysis, we considered detachment duration time should be the primary factor of the ultimate prognosis of DMD. The multiple logistic regression analysis revealed detachment duration (≤ 30 days) as the only significant independent factor for the prognosis of DMD: the longer detachment duration was, the worse the prognosis. Compared to the detachment duration, other factors do not have an equal impact on the prognosis of DMD.
DM can be well restored and perfectly healed because its nutrient fountain is aqueous humour. But If the DM layer (with endothelial cells) floats in the circulating aqueous humour for too long, corneal endothelial cells will shed, and the activity of DM will be weakened (Fig. 3). The flexibility of DM may also decrease over time. In addition, the smooth posterior corneal stroma will be scarred due to the direct impaction of the aqueous humour cycle, which also contributes to the difficulty in DM reattachment[24].
Annamalai Odayappan reported that the timing of the intervention did not influence the reattachment rate in his series of 112 patients [7]. This finding may be because their detachment duration’s interval length was not long enough to show statistical significance. In our study, most patients (29/31) experienced DMD after undergoing cataract surgeries in other hospitals; therefore, more time passed before they arrived at our department and were diagnosed with DMD. Thus, the detachment duration’s interval length was long enough (from 8 hours to 150 days). In Annamalai Odayappan’s study, the median detachment duration time was 4 days, with a reattachment rate of nearly 80%. In our study, the mean detachment duration was 38 days (the longest detachment duration time ever reported), with a 65.6% FS rate. indirectly proves that the longer the detachment duration time is, the worse endothelial function is and the worse the final prognosis.
We also found that when the detachment duration time was more than 30 days, the reattachment rate decreased significantly. If detachment duration was as long as 90 days, there was no hope to achieve FS by descemetopexy. The multiple logistic regression analysis also revealed detachment duration as the only significant independent factor for the prognosis of DMD. Therefore, we determined that detachment duration is the most important factor in the final prognosis of DMD, and we suggest an immediate intervention for DMD.
The grade of DMD also influence the final prognosis of DMD. Mild DMD can be better reset with a bubble compared to severe DMD. Large area detachment often accompanies a great distance between DM and the stroma. To reset DM to the posterior stroma, more force and pressure need to be applied. According to Z. Ismail, different detachment areas may cause different local flow rates of the aqueous humour cycle. In some mild cases, the shear force exerted by the flow on DM tends to help DM remain taut. While in serious cases such as complete DMD, it is difficult to calculate the flow rates around the detached DM. It is also difficult to determine whether the flow is still effectual [25, 26]. In addition, the larger the detachment area is, the more likely it is to create slight folds when reset treatment is adopted, which also contributes to the undesirable prognosis of DMD.
However, in the subsequent multiple logistic regression analysis, the grade of DMD was excluded, so compared to the detachment time, it has relatively little effect on the prognosis of DMD. These results are similar to those obtained by Chameen Samarawickrama in 2015[26].
In the results of our study, other factors, including the surgical reason, presence of diabetes, presence of glaucoma and age, showed no significant differences. We do not think their impact on the prognosis of DMD is as significant as we expected.
The modified descemetopexy
The most common treatments surgical intervention known as descemetopexy (intraocular injection of an air bubble), which is performed to intervene in a serious or visual axis involving DMD. The choice of material for descemetopexy has been a research focus in recent years. Many authors advocate that descemetopexy with 100% air has a lower risk for endothelial toxicity and a pupillary block. However, it may be less effective because of rapid systemic absorption. Once supra-Descemet’s fluid exists, DM redetachment would ensue immediately after air absorption. To date, the success rates of descemetopexy with 100% air have been reported range from only 80–95% [16, 27] [7, 28].
Compared with 100% sterile air, longer acting gas mixtures have better stability and retention times[15, 16], which may lead to better outcomes. However, longer acting gas mixtures had a higher risk for endothelial toxicity and a pupillary block even central retinal artery occlusion (CRAO) or branch retinal arterial occlusion (BRAO) [29]. To date, the success rates of descemetopexy with longer acting gas mixtures have been reported to range from 71.64–95%[16, 30]. There is still no consensus on the choice of materials.
In our study, our surgical method achieved ideal results and the AS rate (96.9%) was the highest ever reported. we determined that the retention time of bubbles in the AC was not the key to success; surgical methods are more important. Our newly modified descemetopexy method combines the advantages of both 100% air and long-acting gas mixtures. On the one hand, we chose 100% sterile air as the material for descemetopexy to lower the risk for endothelial toxicity. The rapid absorption rate of air can greatly shorten the time of pupillary block and reduce its risk. On the other hand, we added the “vacuum technique” to solve the problem of the presence of supra-Descemet fluid and guaranteed the seamless reattachment of DM.
we summarize the newly modified descemetopexy in four tips as follows:
(1) The “air injected incision” should be made at the limbal site where the cornea is relatively compact and opposite to DM rupture.
(2) The “puncture incision” should be made at the point of the cornea where detachment is the most obvious (avoiding the visual axis area) to release the aqueous humour interlayer[31].
(3) The “vacuum technique” was added to remove supra-Descemet’s hu mour interlayer, which may hinder the apposition of DM to the stroma completely. Remember to proceed slowly and gently with the suction step. Excessive suction can cause DM folds and endothelial damage.
(4) Finally, supplement sterile air until the air bubbles fully fill the AC and ensure the intraocular pressure is within the normal range. In this way, DM can be reattached to the corneal posterior stroma seamlessly and be fixed by both the bubble’s buoyancy and the vacuum’s attraction interlayer.
There are some limitations to this study. It is a retrospective study and has a relatively small sample size; the prognosis needed to be evaluated in a larger patient group. It cannot be ruled out that increasing the suction step may cause a slight disturbance in the aqueous humour and affect FS. Further research is needed.