Introduction
Acanthamoeba keratitis (AK) is an uncommon but sight-threatening corneal infection caused by several amoebas of the genus Acanthamoeba. In developed countries, contact lens use is considered as the predominant risk factor for AK [1‐4], whereas in China, the most common risk factor identified was ocular trauma, followed by exposure to contaminated water [5‐7]. In consequence, the majority of AK patients were farmers, but they had poor awareness of medical treatment, which easily delayed the illness. Moreover, few topical antiamoebic agents are available because of their poor corneal penetration and topical toxicity. Doctors must therefore remove the pathogens via corneal transplantation. Penetrating keratoplasty (PKP) is a full-thickness transplant procedure, performed to replace the damaged or diseased cornea with healthy corneal tissue. The donor corneal tissue used in PKP procedures is required to be fresh, but even so, postoperative endothelial immune rejection may still occur. For another widely-performed transplant procedure, lamellar keratoplasty, higher rates of postoperative recurrence have been noted and the residual recipient stroma may have an adverse influence on visual outcomes. To get better surgical results, ophthalmologists have made efforts to improve surgical skills. Big-bubble technique (BB-DALK) has been frequently performed in deep anterior lamellar keratoplasty procedures, after which the postoperative recurrence rate is relatively lower while the visual outcomes are better owing to the removal of the corneal stroma down to Descemet’s membrane (DM) [8, 9]. However, few studies have compared the visual outcomes and complications in patients with AK after BB-DALK or PKP. In order to address this, we compared the postoperative visual acuity, graft survival, Acanthamoeba recurrence, and immune rejection in patients with medically unresponsive AK after BB-DALK or PKP.
Patients and methods
Patients
This retrospective study was approved by the Institutional Review Board of Eye Hospital of Shandong First Medical University. This study was conducted in compliance with the tenets of the Declaration of Helsinki.
Anzeige
A consecutive series of 51 eyes (51 patients) was diagnosed with AK at the Eye Hospital of Shandong First Medical University from August 2013 to August 2020. AK was diagnosed as 1 or more of the following: (1) positive corneal scraping; (2) positive confocal microscopy; (3) positive culture on corneal scraping; and (4) identification of cysts or trophozoites on biopsy specimens. Patient data included age, gender, profession, predisposing risk factors, clinical presentation, and surgical treatment were collected. The clinical presentation was divided into three stages according to the method reported by Robaei et al. [10, 11].
Medical and surgical treatment
Following a diagnosis of AK, the patients were administered topically chlorhexidine 0.02% every 30 min, polyhexamethylene biguanide (PHMB) 0.02% every 30 min, and metronidazole every 1 h. Therapeutic keratoplasty was considered if the corneal inflammatory reaction was not controlled after 2 weeks of antiamoebic drug therapy. The clinical presentations included: corneal stromal edema and infiltrate gradually worsened; corneal ulcers did not heal, but had a tendency to expand, and even with a risk of perforation; and hypopyon gradually increased. DALK surgery was considered in patients with infection or infiltrates greater than 4/5ths of the corneal thickness, but not involved the whole layer, as observed by slit-lamp microscopy, anterior segment optical coherence tomography (As-OCT), and after DM was exposed. For patients detected with endothelial plaque during BB-DALK, or diagnosed with perforation, PK was performed instead.
Surgical procedures
All surgeries were performed by a single surgeon (H.G.). In patients with combined hypopyon, a lateral incision was first made at the corneal limbus, and the anterior aqueous humor was injected into the anterior chamber to flush the hypopyon away. The size of the Hessburg-Barron vacuum trephine (Katena, Denville, New Jersey, USA) was selected according to the scope of the corneal ulcer; its diameter was 1.5 mm greater than that of the corneal ring infiltrate. Approximately 300 μm of corneal thickness was dissected using a corneal trephine. After a large bubble was injected into the posterior stroma, the diseased stroma was cut off using a 45° blade to bare the DM. After DM was exposed, the recipient bed was assessed under the surgical microscope. If the recipient bed was clear, and no endothelial plaques were visible, DALK was continued. In case of infiltration of the residual corneal stoma, or endothelial plaques in DM, PK was conducted instead [12, 13]. A corneal graft with a diameter of 0.25 mm more than the recipient bed was sutured to the recipient. After surgery, the diseased corneal tissue was sent for fungal pathogen culture and pathological examination.
Postoperative treatment
The patients were treated with combination topical antiamoebic agents continually. 0.1% tacrolimus eye drops were prescribed to all patients 4 times a day for 2 months. No systemic or local glucocorticoid medication was applied within 2 months after surgery, and glucocorticoid eye drops were gradually applied after 2 months based on the scanning by confocal laser scanning microscopy. If no Acanthamoeba cysts were observed in the graft and recipient bed, 0.02% fluorometholone eye drops were first given twice daily, and the site was re-examined by confocal laser scanning microscopy after 2 weeks of application. Tobramycin dexamethasone eye ointment was applied once every night if no Acanthamoeba recurrence was observed. Subsequently, all the patients were observed monthly for 6 months. The dosage of glucocorticoid eye drops was adjusted if the patients’ condition remained stable and there were no signs of Acanthamoeba recurrence. Tacrolimus was discontinued after glucocorticoid addition in patients with BB-DALK.
Anzeige
The best-corrected visual acuity (BCVA), intraocular pressure, postoperative refractive outcomes, graft survival, endothelial cell density, perioperative and postoperative complications were evaluated at 1 and 3 years after surgery. The BCVA was recorded by the Snellen chart and converted logarithm of minimal angle of resolution (LogMAR) units for statistical analysis according to the methods of Moussa et al. [14].
In addition, corneal buttons obtained during repeated keratoplasty from Acanthamoeba recurrence patients were stained with hematoxylin-eosin (HE) stain to observe distribution of Acanthamoeba cysts.
Statistical analysis
Statistical analyses were performed using SPSS 24.0 (SPSS, Chicago, Illinois, USA). A P-value of < 0.05 was considered statistically significant. Baseline characteristics are presented as Mean ± SD or Median (IQ1,IQ3) for continuous variables and n(%) for categorical variables. The main demographics, preoperative differences and postoperative refractive status between groups were compared with the Wilcoxon test, t test, chi-square test or Fisher’s exact test. Survival analysis was performed using Kaplan-Meier analysis method and log-rank test.
Results
Patient information
The patients were divided into two groups in accordance with their surgical treatment: 27 eyes in the BB-DALK group and 24 eyes in the PKP group. The mean follow-up period of BB-DALK group was 26.78 ± 21.11 months (range, 0–78 months), whereas 27.29 ± 21.23 months (range, 0–79 months) in the PKP group, the comparisons showed no significant differences (t=-0.087, P = 0.931). The demographics and preoperative characteristics of the two groups were comparable, and the intergroup comparisons showed no significant differences (Table 1).
Table 1
Preoperative characteristics of patients in BB-DALK group and PKP group
Characteristics | BB-DALK group | PKP group | P |
|---|---|---|---|
No. Eyes | 27 | 24 | |
Age, y (Mean ± SD) | 44.56 ± 15.5 | 48.79 ± 15.7 | 0.339 |
Gender n(%) | 0.220 | ||
Male Female | 17(63.0) 10(37.0) | 11(45.8) 13(54.2) | |
Profession n(%) | 0.494 | ||
Farmer Worker Professional technicians Student Other | 7(25.9) 1(3.7) 11(40.7) 3(11.1) 5(18.5) | 7(29.2) 2(8.3) 5(20.8) 2(8.3) 8(33.3) | |
Onset of symptoms to keratoplasty, d Median (Range) | 43(18–96) | 78.5(20–313) | 0.011* |
Causes n (%) | 0.991 | ||
Contact lens wear Contaminated water Plant trauma Foreign body Unknown reason | 3(11.1) 9(33.3) 6(22.2) 5(18.5) 4(14.8) | 2(8.3) 7((29.2) 5(20.8) 6(25) 4(16.7) | |
Treatment prior to diagnosis of AK n(%) | 0.902 | ||
Fungal keratitis HSV keratitis Topical corticosteroid use | 2(7.4) 11(40.7) 12(44.4) | 2((8.3) 6(25.0) 7(29.2) | |
Preoperative LogMAR BCVA | |||
Mean (SD) | 1.95 ± 0.61 | 2.93 ± 0.39 | 0.039* |
Clinical presentations at time of keratoplasty n(%) | 0.024 | ||
Ring infiltrate Endothelial plaques Hypopyon Uveitis Glaucoma | 8(29.6) 2(7.4) 4(14.8) 2(7.4) 1(3.7) | 17(70.8) 18(75) 12(50) 18(75) 9(37.5) | |
Disease stage at presentation n(%) | |||
Stage 2 Stage 3 | 19(70.4) 8(29.6) | 7(29.2) 17((70.8) | 0.003* |
Preoperative clinical presentations
Sudden pain with rubbing, photophobia, tearing, and decreased visual acuity in the keratoplasty was 43.0 days (range, 18–96 days) in the BB-DALK group, whereas 78.5 days (range, 20–313 days) in the PKP group (P = 0.011).
The preoperative BCVA (logMAR) was 2.93 ± 0.39 in the PKP group (Fig. 1), which showed statistically significant worse than that in the BB-DALK group (Fig. 2), which was 1.95 ± 0.61 (P = 0.039).
Fig. 1
The preoperative and postoperative BCVA (logMAR) of PKP group
×
![]()
Anzeige
Fig. 2
The preoperative and postoperative BCVA (logMAR) of BB-DALK group
×
![]()
In our study, the disease stages at presentation in the two groups were stage 2 and stage 3, with the percentage of stage 2 higher in the BB-DALK group, whereas the percentage of stage 3 higher in the PKP group (P = 0.003). The clinical presentations (Fig. 3) were mainly corneal ulcers and ring infiltrate, and the ring infiltrate, endothelial plaque, hypopyon, uveitis, and glaucoma were more common in the PKP group (P = 0.024, Table 1).
Acanthamoeba cysts were found in 46 cases (90.2%) on examination of corneal scrapings and 31 cases (60.8%) examined by confocal microscopy. A total of 38 specimens (74.5%) had positive culture results for cysts.
Fig. 3
A, Preoperative slit-lamp photograph showed a 40-year-old male patient with AK. B, A clear graft was observed at one year after BB-DALK. C, Preoperative slit-lamp photograph showed a 58-year-old female patient with AK. D, A clear graft was observed at one year after PKP
×
![]()
Anzeige
Postoperative VA and refractive outcomes
At 1 year after surgery, the BCVA (logMAR) was 0.71 ± 0.59 in the BB-DALK group (Fig. 2), which showed no differences from that in the PKP group (Fig. 1), which was 0.79 ± 0.66 (P = 0.144). The curvature was 44.51 ± 2.94D and 45.62 ± 2.97D in the BB-DALK and PKP groups, respectively (P = 0.214). The astigmatism and equivalent spherical radius in the two groups showed no statistically significant differences (P > 0.05, Table 2).
At 3 years after surgery, the BCVA (logMAR) was 0.71 ± 0.64 in the BB-DALK group (Fig. 2), which showed statistically significant better than that in the PKP group (Fig. 1), which was 0.93 ± 0.76 (P = 0.010). The curvature was 44.99 ± 2.42D and 46.03 ± 2.38D in the BB-DALK and PKP groups, respectively (P = 0.861). The astigmatism and equivalent spherical radius in the two groups showed no statistically significant differences (P > 0.05, Table 2).
Table 2
Postoperative refractive status of the BB-DALK group and PKP group
1 year after surgery | 3 years after surgery | |||||
|---|---|---|---|---|---|---|
BB-DALK group | PKP group | P value | BB-DALK group | PKPgroup | P value | |
BCVA (logMAR) | 0.71 ± 0.59 | 0.79 ± 0.66 | 0.144 | 0.71 ± 0.64 | 0.93 ± 0.76 | 0.010* |
Corneal Curvature | 44.51 ± 2.94 | 45.62 ± 2.97 | 0.214 | 44.99 ± 2.42 | 46.03 ± 2.38 | 0.861 |
Astigmatism | -3.06 ± 3.88 | -4.08 ± 2.15 | 0.315 | -3.51 ± 3.34 | -4.83 ± 1.75 | 0.091 |
Equivalent Spherical Radius | -1.72 ± 2.74 | -2.70 ± 2.21 | 0.569 | -1.83 ± 2.89 | -2.87 ± 3.03 | 0.179 |
Endothelial Cell Count | 1985 ± 112 | 1895 ± 232 | 0.088 | 1899 ± 125 | 1608 ± 231 | 0.032* |
Graft survival
The graft survival rate was 95.5% and 88.9% at 1 year after surgery in the BB-DALK and PKP group, respectively. Kaplan-Meier analysis and log rank tests showed no significant differences between the two groups (P = 0.467).
Anzeige
The graft survival rate was 89.5% and 61.1% at 3 years after surgery in the BB-DALK and PKP group, respectively. Kaplan-Meier analysis and log rank tests showed statistically significant differences between the two groups (P = 0.046, Fig. 4).
Fig. 4
Kaplan-Meier curve of graft survival rate for two groups at 1 and 3 years after surgery
×
![]()
Loss of corneal endothelial cells
The endothelial cell density was 1985 ± 112cells per square millimeter at 1 year after surgery in the BB-DALK group, with no statistically significant differences compared with that in the PKP group, which was 1895 ± 232cells per square millimeter (P = 0.088, Table 2).
The endothelial cell density was 1899 ± 125cells per square millimeter at 3 years after surgery in the BB-DALK group, with statistically significant differences compared with that in the PKP group, which was 1608 ± 231 cells per square millimeter (P = 0.032, Table 2).
Acanthamoeba recurrence after keratoplasty
The diagnostic methods for the detection of Acanthamoeba recurrence were according to the reference of Zhang et al. [7]. 5 eyes (9.8%) including 3 eyes (11.1%) in the BB-DALK group and 2 eyes (8.3%) in the PKP group experienced Acanthamoeba recurrence after keratoplasty. The rates of recurrence showed no difference between the two groups (P = 1.000).
2 eyes experienced Acanthamoeba recurrence in the graft-host junction in the PKP group. The recipient bed was stained with HE at the recurrence site, which revealed scattered Acanthamoeba cysts at the cut edge of the recipient bed. This indicated that the trephined range during corneal transplantation did not cover all the Acanthamoeba infected stroma, which led to the postoperative recurrence (Fig. 5).
Fig. 5
A, The HE stain of Acanthamoeba recurrent recipient bed in the PKP group. B, The Acanthamoeba cysts (black arrow) were detected at the cut edge of the recipient bed
×
![]()
3 eyes (11.1%) experienced Acanthamoeba recurrence in the recipient bed in the BB-DALK group. The HE stain of the corneal buttons obtained during repeated keratoplasty revealed scattered distribution of Acanthamoeba cysts at the graft-recipient bed junction, indicating that the depth of dissection was not sufficient to remove all the Acanthamoeba infected stroma during corneal transplantation (Fig. 6).
Fig. 6
A, The HE stain of Acanthamoeba recurrent corneal buttons in the BB-DALK group. B, The Acanthamoeba cyst (black arrow) was detected at the graft-recipient bed junction
×
![]()
Intraoperative and postoperative complications
Intraoperative microperforation of the DM occurred in 4 eyes during stromal dissection in the BB-DALK group. These four patients still received DALK with an air bubble injected into the anterior chamber. No cases of hyphema, iris damage, and lens dislocation were observed during the surgery in the PKP group.
In the PKP group, immune rejection was observed in 5 eyes, the rate of rejection episodes was 20.8%. In addition, elevation of intraocular pressure was detected in 6 eyes preoperatively. The intraocular pressure returned to normal after antiglaucoma medication in 4 eyes within 1 month after surgery. The other two patients refused to undertake trabecular surgery, the intraocular pressure was controlled at 22-28mmHg.
Finally, complicated cataract was observed in 3 eyes.
In the BB-DALK group, no immune rejection was detected during the follow-up. However, elevation of intraocular pressure was detected in 1 eye preoperatively, which was controlled with local medical therapy within 1 month after surgery.
Discussion
AK is an uncommon but sight-threatening corneal infection. Prognosis is often poor because of a significant delay in diagnosis and a lack of effective medical treatment [15, 16]. Regarding the timing of corneal transplantation, studies in developed countries recommend a continuous medical treatment for more than 3 months and optical keratoplasty (OKP) when corneal inflammation subsides and stromal scar forms [11]. This can reduce the recurrence of infection and the need for repeated corneal transplantation. However, AK patients in China generally experience chronic disease, associated with a high degree of infection. These conditions are often combined with intraocular infections, such as hypopyon and uveitis [5‐7]. In this study, all AK patients were in stages 2 and 3, and the clinical presentations of corneal ulcers and ring infiltrate were more common in the PKP group (P = 0.024). An important feature of the ring infiltrate was its blurred boundary, which did not allow clear distinction of the extent of Acanthamoeba infection. Tu et al. showed that the presence of ring infiltrate at diagnosis was independently associated with worse visual outcomes [17]. However, it is difficult to control such a serious infection with clinically applied antiamoebic drugs. As the infection gradually worsens to involve the limbus cornea or the whole layer leading to perforation, and a large-diameter PKP needs to be adopted, the risk of postoperative complications, such as non-healing of the graft epithelium and immune rejection, increases [18]. Therefore, corneal transplantation is an ideal treatment when there is no possibility of remission after treatment of medically refractory Acanthamoeba keratitis.
PKP and DALK are the current mainstream surgical approaches. In our study, we chose the appropriate surgical approach mainly based on the depth of the corneal ulcer. If the infection did not involve the DM, DALK was chosen, and when the infection involved the whole layer, PKP was selected [19]. We compared the differences in postoperative visual acuity, refractive status, and graft survival between the two surgical approaches. In the BB-DALK group, the postoperative BCVA (logMAR) was maintained at 0.7 at 1 and 3 years after surgery. Except for intraocular pressure elevation detected in one eye, no immune rejection or complicated cataracts were detected during the follow-up. Compared with the BB-DALK group, the PKP group demonstrated a significantly lower BCVA because of severe preoperative corneal infection and anterior chamber inflammation inevitably. In addition, immune rejection and elevated intraocular pressure were observed in 5 and 6 eyes, respectively, affecting the endothelial cells and graft survival. Three years postoperatively, the endothelial cell count and the graft survival rate of the PKP group were lower than those of the BB-DALK group, and these two indicators in patients with Acanthamoeba keratitis were also lower than those in patients with other corneal diseases including keratoconus [19], corneal dystrophy [20] and fungal keratitis [9, 13]. Hence, if the whole corneal layer is not affected by Acanthamoeba infection, BB-DALK is recommended, as this can reduce the risk of postoperative complications, such as immune rejection and secondary glaucoma, and the visual acuity and graft survival rate can be maintained [21, 22].
Acanthamoeba recurrence after AK corneal transplantation is inevitable, the recurrence rate reported in the literature ranges from 3.6 to 86% [7, 10, 15]. 5 eyes (9.8%), including 3 eyes (11.1%) in the BB-DALK group and 2 eyes (8.3%) in the PKP group experienced in our study. The rates of recurrence showed no difference between the two groups (P = 1.000). However, we consider that the reduced recurrence rate is not related to the surgical approach chosen. Thus, attention was given to the following three key points. First, trephines with an appropriate diameter were selected according to the extent of corneal ulcers during surgery (diameter of 1.5 mm greater than that of the ring infiltrate) to minimize the incidence of residual Acanthamoeba cysts.
In the PKP group, Acanthamoeba infection recurred in two eyes, particularly in the graft-host junction. Scattered Acanthamoeba cysts were observed at the cut edge of the recipient bed, indicating that the trephined range during corneal transplantation did not cover all the Acanthamoeba-infected tissues, which led to postoperative recurrence.
Second, care should be taken to determine whether the infection involves the whole corneal layer after exposure of the DM. In the BB-DALK group, HE staining revealed a scattered distribution of Acanthamoeba cysts at the graft-recipient bed junction in three eyes, indicating that the depth of dissection during BB-DALK was not sufficient to completely remove the Acanthamoeba-infected stroma. Therefore, if DM was clean and no infiltration or endothelial plaques were found, BB-DALK was performed, whereas PKP was performed instead. Accordingly, these principles were followed to guide corneal transplantation in patients with fungal corneal infections [12, 13]. The recurrence rates were lower than those reported in previous studies, such as that of Ti et al. [23]. Therefore, we consider that this method is worth applying.
Third, instead of applying topical glucocorticoid eye drops for 2 months postoperatively, immunosuppressant tacrolimus eye drops can be used as an alternative to prevent immune rejection. Glucocorticoid eye drops can reduce corneal inflammation, neovascularization, and treat scleritis, but Robaei et al. concluded that corticosteroid use before diagnosis of AK was associated with an approximately 4-fold increased odds of a poorer visual outcome [10]. Moreover, corticosteroids have been shown in animal models in vivo to induce a 4- to 10-fold increase in the number of Acanthamoeba trophozoites [24]. Hence, FK506 is considered a better choice because it suppresses the expression of T cell-mediated lymphokines and the interleukin-2 receptor and the generation of cytotoxic T cells, which effectively prevents graft rejection and maintains the transparency of the graft. Likewise, Yuan et al. demonstrated the lower rate of graft failure of PKP in 15 cases of AK was associated that FK506 was prescribed postoperatively whereas corticosteroids were avoided [25].
There is a bias in the patient selection criteria in this retrospective study, which is inevitable. Firstly, there were differences in the clinical presentations and grading of the patients in the two groups before surgery. Secondly, the patients were not grouped based on randomization principles, but on intraoperative observation of whether the infection involved the full thickness.
Corneal infection and anterior chamber inflammation are serious issues in AK patients in China. Antiamoebic drugs are often ineffective; therefore, we advocate treatment with corneal transplantation as soon as possible. If the Acanthamoeba infection does not involve the whole layer of the cornea, BB-DALK is recommended to maintain the visual acuity and survival rate. To minimize residual Acanthamoeba cysts, the trephine range used during corneal implantation had a diameter of 1.5 mm greater than that of the ring infiltrate. Another important measure for reducing the recurrence after corneal transplantation was avoiding the application of topical glucocorticoid drops within 2 months after surgery.
Acknowledgements
The authors would like to thank Ms. Tong Liu from Eye Hospital of Shandong First Medical University for her editorial and linguistic assistance.
Declarations
Ethics approval and consent to participate
All procedures performed in studies involving human participants were in accordance with the ethical standards of the ethical board committee of the Eye Hospital of Shandong First Medical University and with the 1964 Helsinki declaration and its later amendments or comparable ethical standard. All participants voluntarily agreed to participate in this study and all gave written informed consent.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.