Article Text

Download PDFPDF
Outcomes of keratoplasty in the mucopolysaccharidoses: an international perspective
  1. Kaitlyn L Ohden1,
  2. Susanne Pitz2,
  3. Jane Ashworth3,
  4. Augusto Magalhães4,
  5. Diane R Marinho5,
  6. Päivi Lindahl6,
  7. Kristina Teär Fahnehjelm7,
  8. C Gail Summers8
  1. 1Medical School, University of Minnesota, Minneapolis, Minnesota, USA
  2. 2Department of Ophthalmology, University Medical Center, Johnannes Gutenberg-University, Mainz, Germany
  3. 3Department of Ophthalmology, Manchester Royal Eye Hospital, Manchester, UK
  4. 4Department of Ophthalmology, S. João Hospital, Porto, Portugal
  5. 5Department of Ophthalmology, Hospital de Clínicas de Porto Alegre, Porto Alegre, RS, Brazil
  6. 6Helsinki University Eye Hospital, Helsinki, Finland
  7. 7Department of Clinical Neuroscience and Rehabilitation, Institute of Neuroscience and Physiology, University of Gothenburg St. Erik Eye Hospital, Stockholm, Sweden
  8. 8Departments of Ophthalmology & Visual Neurosciences and Pediatrics, University of Minnesota, Minneapolis, Minnesota, USA
  1. Correspondence to Dr C Gail Summers, Departments of Ophthalmology & Visual Neurosciences and Pediatrics, University of Minnesota, 701 25th Ave. S., #300, Minneapolis, MN 55454, USA; summe001{at}


Objective To describe visual outcomes after penetrating keratoplasty and deep anterior lamellar keratoplasty in patients with mucopolysaccharidoses.

Methods This is a retrospective review of keratoplasty in consecutive patients from Brazil, England, Finland, Germany, Portugal, Sweden and the USA. All patients had corneal clouding due to mucopolysaccharidoses. Preoperative and postoperative visual outcome and ocular comorbidities were identified. Success was arbitrarily defined as any improvement in visual acuity or best-corrected visual acuity better than logarithm of the minimum angle of resolution 0.30 (20/40). Statistical analysis included only data from first operated eyes in the 16 patients who underwent bilateral keratoplasty.

Results Forty-eight eyes from 32 patients with mucopolysaccharidoses I, IV or VI are reported. Mean follow-up was 70 months (range: 5–186). Penetrating keratoplasty was performed in 45 eyes and deep anterior lamellar keratoplasty in 3 eyes. At last follow-up, a successful visual outcome for penetrating keratoplasty in first operated/only operated eyes was found in 63%. Rejection episodes occurred in 23% of grafts; however, a clear graft was recorded at last follow-up in 94%. Ocular pathway comorbidities were identified in 63% of eyes transplanted.

Conclusions Clear corneal grafts can be obtained for patients with corneal clouding due to mucopolysaccharidosis with improvement in visual acuity in the majority.

  • Cornea
  • Child health (paediatrics)
  • Genetics
  • Treatment Surgery
  • Vision

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Mucopolysaccharidoses (MPS) are a group of inherited disorders caused by deficient lysosomal enzymes required to metabolise glycosaminoglycans (GAGs). These disorders are classified according to the enzyme defect, and those with corneal clouding include MPS IH (Hurler), MPS IS (Scheie), MPS IH/S (Hurler/Sheie), MPS IV (Morquio) and MPS VI (Maroteaux-Lamy).1 Persons affected by MPS appear normal at birth; however, over time, GAGs accumulate in cells, causing systemic damage. Affected patients may exhibit a wide range of clinical findings, including coarse facial features, short stature, organomegaly, skeletal dysplasia, joint disease, cardiac and pulmonary disease, and ophthalmic pathology.1 Neurological decline and cognitive impairment may occur in some types.

Corneal clouding due to GAG deposition remains an important cause of photophobia and vision loss for patients with MPS.1 ,2 While advances are being made in haematopoietic stem cell transplant (HSCT) and enzyme replacement therapy (ERT) to prevent continued deposition of GAGs, visual disability is common.2–4 Penetrating keratoplasty (PK) has been shown to be a viable option to correct severe corneal clouding; however, case studies are small. Here, we report a large international collaboration of data on keratoplasty for corneal clouding in MPS types I, IV and VI. Our primary end points were visual outcome and graft survival after keratoplasty, with additional data collected on associated retinal dystrophy, optic neuropathy, ocular hypertension/glaucoma and episodes of graft rejection.


We reviewed collective data on 32 patients with MPS types IH (Hurler), IS (Scheie), IH/S (Hurler-Scheie), IV (Morquio) and VI (Maroteaux-Lamy) who underwent corneal transplantation in at least one eye. These individuals represent consecutive keratoplasties in patients with MPS who are from Brazil, England, Finland, Germany, Portugal, Sweden and the USA. Sixteen of these patients had bilateral transplants. For statistical purposes, only the first operated eye was included for patients who had bilateral transplants, resulting in a total of 32 eyes for this report. We also report data on three deep anterior lamellar keratoplasties (DALK). The Institutional Review Board at the University of Minnesota approved the study, which was conducted in accordance with the tenets of the Declaration of Helsinki.

The following data were collected: type of MPS, ophthalmic and medical history, best-corrected monocular visual acuity (BCVA) before and after keratoplasty (converted to logarithm of the minimum angle of resolution (logMAR) equivalent), age at keratoplasty, type of keratoplasty, duration of follow-up, complications and comorbidities. The BCVA end point was chosen to be the last recorded ophthalmological evaluation. Visual acuity results were analysed with respect to MPS type, type of procedure, age at keratoplasty and complications. Visual acuity for count fingers (CFs) was assigned to be 2.00 logMAR and hand motion (HM) was assigned to be 3.00 logMAR.5 Light perception (LP) and no light perception (NLP) were not assigned logMAR values as they represent simple detection of stimulus.5 Thus, mean logMAR BCVA reported here does not include LP and NLP.

Final outcome was described according to function and structure. We have arbitrarily defined functional success as any improvement in visual acuity or a final visual acuity of logMAR 0.30 (20/40) or better if BCVA was not recorded (NR) prior to keratoplasty. Structural keratoplasty success was defined as a clear graft, regardless of whether rejection episodes occurred. We defined paediatric patients as those 18 years or younger at time of transplant.


A total of 48 eyes from 32 patients were identified. In patients with bilateral keratoplasties, only the first operated eye was included for statistical purposes, yielding a total of 32 penetrating keratoplasties in the primary analysis. Raw data for all 48 eyes are provided in online supplementary file 1. Data for second operated eyes are included only in the last two paragraphs of results. Some of the cases presented herein have been previously reported.6 Full-thickness keratoplasties were performed in 13 children and 19 adults. Mean age at keratoplasty was 23 years (range: 8–64 years). Mean duration of follow-up was 71 months (range: 5–186 months, table 1). Three eyes also underwent cataract removal and intraocular lens (IOL) placement and two had cyclophotocoagulation during the same surgery or within the follow-up period. No patient developed endophthalmitis. DALK was performed in three additional eyes (one child and one adult). We were unable to obtain information regarding prior to or concurrent therapy with HSCT or ERT on all patients.

Table 1

Patient demographics

Mean pre-keratoplasty BCVA was 1.5 (20/600), LP in 4 and NR in 5; mean post-keratoplasty was 0.79 (20/120) and LP in 4. All types of MPS showed improvement or stabilisation of mean BCVA. Patients with MPS IH, IH/S and VI were younger at keratoplasty and had more associated ophthalmic findings compared with MPS IS and IV. A total of 63% (20 of 32) transplants had a successful visual outcome. Of the 12 eyes without improvement in BCVA, visual acuity before and after keratoplasty was the same in seven eyes, NR in two and worse in three at last follow-up (table 2). Of the four eyes with preoperative visual acuity of HM, only one attained post-transplant BCVA better than HM. Of four eyes (all paediatric) that had BCVA of LP before transplant, two attained BCVA better than LP after surgery.

Table 2

Keratoplasty outcomes

Thirteen surgeries were performed on paediatric patients, the youngest at age 8 years. Several of the paediatric surgeries exhibited the most dramatic improvements in BCVA, including two that went from LP to 0.48 (20/80) and 0.60 (20/60) BCVA. However, two of three transplants with BCVA deterioration were also paediatric. Also, 3 of 13 (23%) paediatric transplants experienced at least one rejection episode, which was similar to the 4/19 rejection rate of adults (22%).

A total of seven grafts (22%) in patients of all ages had at least one rejection episode, with a mean visual outcome for these eyes of 0.37 (20/50) and two LP, compared with non-rejected grafts with BCVA outcome of 0.61 (20/80) and two LP. Rejection was most common in MPS VI (4/8 eyes had at least one rejection episode). No grafts were rejected in patients with MPS IH or MPS IV. One graft in a patient with MPS IS showed recurrence of corneal deposits after 43 months of follow-up. Overall, 2/32 full-thickness grafts failed, giving an overall graft success rate of 94% for graft survival. None of the three DALK grafts experienced rejection episodes.

Sixty-three per cent of eyes transplanted had at least one ocular pathway comorbidity. These included glaucoma/ocular hypertension (n=15), optic nerve atrophy or oedema (n=6), retinal degeneration (n=4), visual field constriction (n=1), exotropia (n=1) and decreased amplitude or increased latency on electroretinogram (ERG) (n=3). Retinopathy was diagnosed on the basis of symptoms (visual field loss and nyctalopia), signs (pigmentary changes) and/or ERG abnormalities.

Sixteen cases had bilateral corneal transplants concurrently or subsequently (see online supplementary file 2). Three of the bilateral cases were done concurrently for reasons related to general anaesthesia risk. All bilateral cases had similar BCVA before and after transplant in the first operated and fellow eyes. Five first operated eyes experienced rejection, and three of the respective second eyes also had episodes of rejection.

Three eyes had DALK performed rather than full-thickness keratoplasty (PK). All had excellent BCVA outcomes with mean final BCVA of 0.2 (20/30). Of note, one patient with MPS IS had a PK in the first operated eye and later had DALK in the fellow eye. The final BCVA for each eye was 0.10 (20/25).


MPS are relatively rare disorders, and there is a paucity of data regarding treatment options for corneal clouding in MPS. This international collaboration permitted evaluation of 32 full-thickness keratoplasties in patients with MPS. Of these, 20 (63%) showed improved BCVA or postoperative BCVA of at least 0.30 (20/40) at last follow-up (mean 71 months). Overall graft survival was 94% (30/32). In addition, we report three DALK cases with favourable outcomes in patients with MPS.

Patients with the most phenotypically severe forms of the MPS present with facial dysmorphism and respiratory disease in early life, and require screening for associated ocular findings. The milder forms of MPS have more variable phenotypes, in which ocular pathology may be the first clue to the diagnosis.1 ,7–10 Ophthalmic findings in MPS include pseudoexophthalmos, shallow orbits, ptosis, corneal opacification, glaucoma, papilledema, optic atrophy and retinal degeneration.1 Progressive corneal clouding due to GAG accumulation is a significant cause of visual disability and may preclude accurate assessment of the fundus. Intraocular pressure may also be difficult to accurately assess due to increased corneal hysteresis.11 Hurler (MPS I), Morquio (MPS IV) and Maroteaux-Lamy (MPS VI) syndromes are frequently associated with severe, progressive corneal clouding with visual acuity deterioration and marked photosensitivity.12

While there are advances in HSCT and ERT for MPS, no current treatment halts the progression of corneal clouding. Keratoplasty can be a beneficial intervention for patients with corneal opacification and should be considered if vision is impaired or view of the fundus precludes sufficient evaluation of the optic nerve and retina. Current literature shows that primary full-thickness graft survival for all conditions ranges from 62% to 86%.13 ,14 There are many reports of keratoplasty for MPS that have resulted in clear grafts and favourable visual outcomes.6 ,15 ,16 One small study showed that the majority of corneal grafts for MPS I (all types) remain clear, with a few documented cases of graft failure.6 There are some reports of recurrent corneal opacification following keratoplasty in patients with MPS IV,17 while others show excellent outcomes.6 Results for keratoplasty in patients with MPS VI are mixed,15 ,16 ,18 with one report of a patient who had previously been treated with HSCT maintaining a clear corneal graft for 13 years.4 Most reported cases of keratoplasty for MPS experienced some benefit, with no reported endophthalmitis and few complications.

Our data support previous studies showing that the majority of patients with MPS who undergo transplant for corneal clouding experience benefit. The patients least likely to obtain improvement in visual acuity from keratoplasty had preoperative BCVA of HM or LP, although there were some successes when preoperative BCVA was poor. Of the three patients who experienced worse BCVA after transplant, deterioration was mild in relation to starting BCVA. Some patients experienced an improvement in BCVA after transplant, but vision later deteriorated due to ocular comorbidities.

Most of our paediatric cases had MPS IH and 8/13 had additional significant ocular comorbidities. The challenges associated with adequate examination of severely affected patients may have a marked effect on measured visual acuity or intraocular pressure. Nonetheless, the improvement in BCVA in the paediatric patients was the most dramatic. The success of these cases may be in part due to the mean older age of the paediatric patients (12.4 years) as very young paediatric patients undergoing keratoplasty frequently experience rejection and poor outcome.19

Our keratoplasty graft success rate of 94% is higher than reported keratoplasty graft success for all conditions. We postulate that this is due to a younger recipient population (mean age at keratoplasty was 23 years) or that persons with MPS are more receptive to corneal grafts than those with other diseases requiring keratoplasty. The majority of rejection episodes reported were controlled medically, with no effect on final BCVA outcome. No rejection episodes occurred in patients with MPS IH and MPS IV, while MPS VI may be associated with a higher risk of graft rejection.6 ,16 Finally, our data on bilateral transplants suggest that graft outcomes of the first transplant for MPS may be predictive of second-eye outcomes.

While PK is the most widely used intervention for corneal opacification, DALK may be another option as endothelial function in the MPS may be preserved. Careful consideration of DALK versus full-thickness grafts for MPS must be made on a case-by-case basis as many patients with MPS may be ineligible for DALK due to excessive GAG accumulation in the stroma or if corneal clouding obscures assessment of endothelial cell status.18 Others have reported success with DALK in MPS.18 ,20 ,21 In our patient with full-thickness PK in one eye and DALK in the second eye, the BCVA outcome for each eye was identical. Our three cases with DALK had excellent visual outcomes and did not experience any rejection. Less severe cases are typically selected for DALK, possibly representing selection bias. Our results support suggestions that DALK may be preferred over PK with respect to outcome, rate of rejection and corticosteroid-associated pathology;22 however, larger sample sizes are needed to demonstrate superiority.

We acknowledge that there are multiple limitations to this report, due in part to the fact that MPS is a relatively rare disease with very recent advances in diagnosis and treatment options. There are the usual limitations associated with a small retrospective study, as well as non-standardised procedures, variable follow-up times and use of two end points only at latest follow-up (visual acuity and graft clarity). The worst visual acuities (patients with BCVA of LP) could not be included in statistical calculations to obtain mean BCVA; therefore, the mean improvement in VA may actually be greater than reported. Additionally, acuities for HM and CF were assigned numerical values, which may or may not be accurate. Multiple potential visual confounders make it difficult to isolate the effects of keratoplasty on visual outcome. We suspect many of the poor BCVA outcomes were due, in part, to optic neuropathy and retinopathy. Despite correction improving BCVA, post-keratoplasty refraction may not have been tolerated well by all patients in daily life. It is also possible that patients with fewer systemic or ocular comorbidities were chosen for transplantation,6 leading to selection bias. Particularly severe forms of MPS are frequently associated with mobility problems, marked intellectual and behavioural disabilities, and anaesthetic risks, making adequate examination difficult.8 In a few of our patients, data were incomplete.

Given the noted limitations of this study, a report on visual outcome for a large cohort of patients with MPS who underwent keratoplasty and were followed for almost six years, on average, is unique. This international collaboration provides additional information on bilateral transplant and DALK, which may be useful in determining treatment for future patients. Due to the frequently shortened lifespan and other difficulties facing these patients, we suggest keratoplasty for appropriate patients with vision-limiting corneal opacity as this will likely improve quality of life for patients. Adverse outcomes are due to rejection and the associated ocular abnormalities of optic atrophy, retinal degeneration and glaucoma. Our data show that clear grafts and stable to improved vision can be obtained for patients with corneal clouding due to MPS, although multiple ocular comorbidities may affect the final visual outcome.



  • Contributors All authors contributed substantially to the conception and design of this paper. CGS was responsible for data acquisition. SP, JA, AM, DRM, PL, KTF and CGS submitted data for interpretation. CGS and KLO were responsible for primary literature searches as well as data organisation and interpretation. KLO and CGS drafted the paper. All authors had access to the data, contributed to revising the manuscript and approved the final version to be submitted. All authors accept responsibility for the integrity of the work.

  • Funding Unrestricted grant to the Department of Ophthalmology at the University of Minnesota from Research to Prevent Blindness, Inc., New York, New York, USA.

  • Competing interests None declared.

  • Ethics approval The Institutional Review Board at the University of Minnesota.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Both supplementary files will be available online to anyone with access to the article.

Linked Articles

  • At a glance
    Keith Barton James Chodosh Jost Jonas