Calcification of intraocular lenses (IOLs) is an infrequently reported complication of cataract treatment. Bucher and coauthors1 describe an IOL that calcified 1 day after surgery. Olson et al.2 report several cases of intraoperative calcification of IOLs. In both reports, calcification occurred soon after implantation. Recently, Yu and Shek3 described a delayed form of IOL calcification in 3 patients with the same model of hydrogel IOL. Others have reported similar findings.4
Many aspects of this potentially severe complication remain unknown, including its etiology, clinical features, pathogenesis, prognosis, and treatment. This paper clarifies the clinical features of the phenomenon based on observation of a large case series.
Patients and methods
Forty-four patients with a diagnosis of delayed IOL calcification were examined between November 1999 and April 2000 at Queen Mary Hospital or Tung Wah Hospital, Hong Kong. The diagnosis of IOL calcification was based on characteristic clinical features of confluent, granular, white deposits on the IOL surface. The presence of calcium was confirmed in 3 eyes by examining the explanted IOLs.3 All enrolled patients were examined by an ophthalmologist experienced in recognizing the condition. Patients found to have other causes of media opacity, such as lens cell growth5 or posterior capsule opacification (PCO), were excluded from the study unless they had coexisting IOL calcification.
Data collected included patient age and sex and the eye involved. Detailed medical, ophthalmic, drug, and allergy histories were obtained. The patients' charts were reviewed for preexisting ocular pathology and best corrected visual acuity (BCVA) before IOL implantation. Operative details on the date of surgery, type of anesthesia, IOL model and power, IOL serial number, surgical method, viscoelastic material and other intraoperative agents used, additional procedures, and complications were retrieved. Postoperative data included medications used, BCVA 3 months after surgery, and the date of the IOL calcification diagnosis. Three patients had IOL explantation before the study. Their BCVA before explantation was recorded.
All patients had a detailed ophthalmic examination including measurement of BCVA, intraocular pressure (IOP), slitlamp evaluation after pupil dilation, and fundus evaluation. Visual acuity was measured in Snellen (decimal) notation. For calculation purposes, log10 values were generated (−1 log10 = 1 logMAR unit).
The severity of IOL calcification was graded according to the appearance of the IOL and clarity of the fundal view as follows: 1 = forceps marks without generalized calcification (Figure 1); 2 = mild generalized calcification with a clear view of the posterior capsule and retina (Figure 2); 3 = moderate generalized calcification, posterior capsule and retina visible but with loss of detail (Figure 3); 4 = advanced generalized calcification with no retinal view (Figure 4).
A paired t test was used to study the difference between postoperative visual acuity and current visual acuity and between preoperative and postoperative visual acuity. Stepwise regression analysis was used to examine the difference in visual acuity loss among patients with diabetes mellitus, hypertension, and/or ischemic heart disease. Multiple regression analysis was used to study the effects of age, sex, left or right eye, acetylcholine use, diabetes mellitus, hypertension, ischemic heart disease, time to diagnosis, disease duration, surgeon, and anesthesia on visual acuity loss. The Spearman rank correlation coefficient was used to study the relation between the calcification grade and visual acuity loss. The effect of diabetes mellitus, hypertension, and ischemic heart disease on calcification grades was analyzed by the Wilcoxon rank-sum test with exact P values. The SPSS version 10.0 and SAS version 8.0 were used for all analyses, and a 5% level was used in all significance tests.
Forty-six eyes of 44 patients (19 men, 25 women) had features typical of IOL calcification. One eye only was involved in 42 patients, and both eyes were affected in 2 patients. The mean age of affected patients was 73 years (range 59 to 95 years). All except 1 patient (eye 22) were ethnic Chinese. Most patients had other preexisting medical illness. Diabetes mellitus was noted in 23 eyes (50%) and systemic hypertension in 23 eyes (50%) (Table 1).
Preoperative BCVA ranged from 0.01 to 0.3 (geometric mean 0.05). Intraocular lens implantation was performed by 9 surgeons between March 1997 and April 1999 at Queen Mary Hospital and Tung Wah Hospital. Phacoemulsification was done using the Alcon Legacy or Alcon Ten Thousand unit. Eighty-seven percent of eyes had local anesthesia, 7% had topical anesthesia without intracameral lignocaine, 4% had topical anesthesia with intracameral lignocaine, and 2% had general anesthesia. Fortified balanced salt solution (BSS Plus®) and sodium hyaluronate 3.0%-chondroitin sulfate 4.0% (Viscoat®) were used in all eyes. Intracameral acetylcholine was used in 41% of eyes. Two surgeries were combined with trabeculectomy and antimetabolites, and 1 surgery was combined with posterior vitrectomy and endolaser. All surgeries were uneventful except in 1 eye in which the posterior capsule ruptured. Dexamethasone and neomycin eyedrops were used after surgery in all patients. The postoperative recovery was uneventful in all cases.
Three months after IOL implantation, BCVA ranged from 0.1 to 1.0 (geometric mean 0.4), with 24 eyes (52%) having an acuity of 0.5 or better, 18 (39%) between 0.2 to 0.4, and 4 (9%) 0.1 or worse. Almost all patients had a significant improvement in visual acuity (mean 0.85 log10 ± 0.40 [SD]) (P < .001). A neodymium:YAG (Nd:YAG) laser capsulotomy was performed in 1 eye 18 months after surgery (Table 1).
Intraocular lens calcification occurred a mean of 12.5 ± 4.7 months after surgery (range 4 to 26 months) (Table 1). Seventy-four percent of patients were diagnosed between 8 and 15 months after surgery. All cases of calcification occurred in eyes with the Hydroview hydrogel IOL (Bausch & Lomb Surgical).
The patient examination was conducted a mean of 19 months after surgery (range 8 to 36 months). The BCVA at the examinations ranged from 0.01 to 0.8 (geometric mean 0.13), with 11 eyes (24%) having an acuity of 0.5 or better, 14 (30%) between 0.4 and 0.2, and 21 (46%) 0.1 or worse. Acuity was significantly lower than at 3 months after implantation (mean deterioration 0.45 ± 0.45 log10) (P < .001).
Most patients had a decrease in visual acuity after IOL calcification. Seven eyes (15%) had no drop in visual acuity. Seven eyes (15%) had a decrease of less than 0.17 log10 unit (about 1 Snellen line), 12 (26%) of 0.17 to less than 0.34 log10 (1 to 2 Snellen lines), 7 (15%) of 0.34 to less than 0.50 log10 (2 to 3 Snellen lines), 4 (9%) of 0.67 to less than 0.84 log10 (4 to 5 Snellen lines), and 9 (20%) of 0.85 log10 (5 Snellen lines) or more. The maximum loss in visual acuity was 1.65 log10 (about 10 Snellen lines). The mean reduction in vision was 0.45 log10 (about 2.8 Snellen lines).
A stepwise regression analysis of visual loss in patients revealed a significant visual loss regardless of the presence or absence of diabetes mellitus, hypertension, or ischemic heart disease (Table 2). Estimated vision loss was 0.20 log10 (P < .022) in patients with no diabetic mellitus or ischemic heart disease. Visual loss was more advanced in patients with diabetes mellitus or ischemic heart disease (Table 2).
Multiple regression analysis confirmed the effect of diabetes mellitus on visual loss (P = .016; Table 3). On average, diabetic patients lost 0.33 log10 units (2 Snellen lines) more of visual acuity than nondiabetic patients. The effect of ischemic heart disease on visual loss was also significant (P = .003; Table 3). Affected patients lost an average of 0.48 log10 units (3 Snellen lines) more than unaffected patients. Hypertension had no additional effect on visual loss (Table 3).
Eyes that exhibited calcification earlier after surgery had more advanced visual loss (P = .001; Table 3 and Figure 5). Eyes with calcification 1 month earlier lost, on average, an additional 0.04 log10 unit of acuity. In addition, eyes operated on by certain surgeons had more severe visual loss (P = .024; Table 3). Age, sex, left or right eye, acetylcholine use, disease duration, and type of anesthesia had no significant effect on the amount of visual loss (Table 3).
Under slitlamp examination, 43 eyes (93%) had generalized opacification of the IOL. Opacification was granular, white, and homogeneous. It affected the entire anterior and posterior surfaces but spared the inside of the IOL. Three eyes (7%) developed similar opacification that was localized to 2 marks. These marks were paired and linear and were observed on the anterior IOL surface only (Figure 1). It appeared as though the marks corresponded to the area of the IOL grasped by the holding forceps during IOL implantation (forceps marks).
Forty-four eyes (96%) had forceps marks, 3 without generalized IOL opacification and 41 in association with generalized IOL opacity (Figure 6). There were no forceps marks in 2 eyes (4%) (Figure 4). The forceps marks were conspicuous because they had a different opacification density from that of the adjacent surface. Most forceps marks were less opaque than the surrounding area (Figure 6). However, in a few eyes, the marks appeared more opaque (Figure 7).
The severity of calcification varied among eyes. Three eyes (7%) had grade 1 calcification, 17 eyes (37%) had grade 3, and 5 eyes (11%) had grade 4. Spearman correlation analysis showed that the higher the calcification grade, the more severe the visual acuity loss (r = 0.456, P = .001). The mean loss was 0.17 log10 (1 Snellen line) in eyes with grade 1 calcification, 0.28 log10 (1.7 Snellen lines) in eyes with grade 2, 0.50 log10 (3.0 Snellen lines) in eyes with grade 3, and 1.27 log10 (7.6 Snellen lines) in eyes with grade 4.
There was no difference in the calcification grades in patients with and without diabetes mellitus (P = .123), hypertension (P = .262), and ischemic heart disease (P = 1.000).
After our report of delayed IOL calcification in 3 patients with a Hydroview IOL,3 we discovered more eyes with similar clinical features. All cases were in eyes with a Hydroview IOL, a hydrogel lens made from a copolymer of HEMA (2-hydroxyethyl methacrylate) and HOHEXMA (6-hydroxyhexyl methacrylate). This IOL has been reported to give good results after implantation (H. Seward, MD, “Worldwide Results of the Storz Hydroview IOL,” presented at the Symposium on Cataract, IOL and Refractive Surgery, Seattle, Washington, USA, June 1996). Calcification of the IOL was unknown before our observations in late 1998. Similar reports followed from other countries, with the phenomenon affecting the same IOL model.3 Information from the IOL manufacturer showed that more than 400 000 Hydroview IOLs had been sold to 3500 centers worldwide. Only 31 centers reported the phenomenon (“Calcium Deposits on Hydroview IOLs May Come from Silicone Packing,” Ocular Surgery News, March 2001).
The presence of calcium on the IOL was confirmed by analysis of the explanted lenses. Yu and Shek3 detected calcium and phosphorus on explanted lenses with elemental analysis and identified hydroxyapatite by x-ray diffraction. Apple and coauthors4 demonstrated the presence of calcium on the explanted lenses by staining with alizarin red.
Although there are reports of early-onset IOL calcification affecting lenses of HEMA or silicone,1,2 delayed-onset calcification has been reported only on Hydroview IOLs. From March 1997 to April 1999, 2100 IOL implantation surgeries were done at the 2 hospitals in our study. Hydroview IOLs were used in 498 surgeries, with poly(methyl methacrylate), acrylic, or silicone lenses used in the rest. Forty-six calcified Hydroview IOLs were detected, and no other models were involved. Thus, Hydroview appears to be an important factor in IOL calcification. The chance of calcification of a Hydroview IOL at our center was calculated to be 9% with a 95% confidence interval.
Hydrogel comprises a large family of polymers. Certain hydrogels are very active in promoting calcification when immersed in solutions containing calcium phosphate.6–8 Calcification also affects hydrogel contact lenses.9 Winter and Simpson10 report the calcification of synthetic sponge made from hydrogel (polyHEMA) in young pigs. The calcified IOL reported by Bucher and coauthors1 was a hydrogel (polyHEMA) lens. We believe that the calcium affinity of the HEMA/HOHEXMA copolymer is an important factor in IOL calcification. The role of IOL material in calcification must be investigated further.
In the current study, IOL calcification was noted 4 to 26 months after cataract surgery. Delayed onset was the uniform finding, with calcification tending to occur approximately 1 year after implantation. No new cases were seen after 2 years. However, we used few Hydroview IOLs in early 1997, with the number increasing substantially in mid-1998. This might explain the apparent surge in cases of calcification in mid-1999. Whether there was a genuine peak occurrence must be clarified by further study. Also, our longest follow-up was about 3 years, and it is premature to conclude that unaffected IOLs will remain free of calcification after 2 years.
All our patients except 1 were ethnic Chinese, and most had a coexisting medical illness, notably diabetes mellitus and hypertension. Whether these are risk factors for IOL calcification is uncertain as we did not have a control group for comparison. It would be worthwhile to examine this closer in future investigations.
Patients with diabetes had more severe visual loss than nondiabetic patients. Ischemic heart disease also exacerbated visual loss. However, neither disease was associated with a higher calcification grade. This indicates that their effect on visual loss was not caused by more advanced IOL calcification. Excessive visual loss in diabetic patients could be explained by complications of the disease. The reason for additional visual loss in ischemic heart disease is unclear.
Three surgeons in particular had a higher incidence of more severe visual loss. However, they operated on 1 or 2 eyes only, so this result should be interpreted with caution.
Bucher and coauthors1 attribute IOL calcification to the use of a phosphate-containing solution during surgery. We also investigated the effect of the intraoperative agents. Our analysis showed that the use of intraoperative acetylcholine did not exacerbate visual loss or IOL calcification. We used BSS Plus and Viscoat in all patients and were unable to evaluate their influence because of the lack of a control group. BSS Plus contains calcium chloride dihydrate (3.85 mg in 500 mL) and dibasic sodium phosphate (0.433 mg in 500 mL). Viscoat contains sodium dihydrogen phosphate hydrate (0.45 mg/mL) and disodium hydrogen phosphate (2 mg/mL). Further study is necessary to clarify their role in calcification.
About half our patients had diabetes mellitus. Thus, we thought it was important to determine whether the complications of the disease rather than the IOL calcification was the cause of visual loss in these patients. Although diabetic eyes had more severe visual loss than nondiabetic eyes, eyes not affected by diabetes or ischemic heart disease also had significant visual loss. Visual loss could not be explained by diabetes alone, and another explanation had to be found. Posterior capsule opacification is another cause of reduced vision,11 and hydrogel IOLs are associated with a higher incidence of PCO.12,13 Only 1 Nd:YAG laser capsulotomy was performed in our series, showing that PCO was not a major factor. In contrast, grade 3 or 4 IOL calcification had major impact on clarity of the fundus view. We thus believe that IOL calcification was the major factor leading to visual loss in our patients.
Eighty-seven percent of patients had forceps marks on the IOL. Forceps marks have been reported, especially on acrylic IOLs.14 The marks were believed to result from mechanical scratching of the IOL surface. However, the forceps marks in our patients were not scratches. They were formed because they had a different opacification density than the surrounding area. A forceps was applied on the IOLs briefly during surgery; however, calcification was delayed for months. The marks and the time lag provide clues to the pathogenesis of the calcification. The pathogenesis, although unknown at this time, could have been modified by the interaction between the IOL and the holding forceps. The interaction was largely mechanical. The forceps might have scratched material off the IOL surface, compressing the hydrogel polymer and the water that it contained. This might have influenced the distribution of intraoperative agents, including viscoelastic materials, on the IOL surface. The forceps itself might have brought debris onto the IOL. Other interactions might have persisted long enough to affect the calcification process. It is also possible that calcification began soon after surgery but became clinically evident only several months later. The forceps marks and the time lag must be explained before a hypothesis of the calcification process can be formed.
The severity of calcification varied. At the time we examined the patients, 83% of eyes had grade 2 or 3 calcification. About one half had early calcification (grade 1 or 2), and one half had more advanced calcification (grade 3 or 4). This picture could change over time. After 1 or 2 years, the distribution might be significantly different. More study is necessary to clarify the natural history of the phenomenon.
The visual loss was more severe when the IOL calcified relatively soon after surgery. However, disease duration had no significant relation to the amount of visual loss. We believe that the calcification progressed at different rates in different patients. Certain factors might have modulated the process and accelerated it in some eyes. These must be defined by further study of such factors as the hospital environment, procedural irregularities, patient characteristics, and irregularities in IOL manufacturing and handling.
Two patients had bilateral involvement, and calcification was very advanced (grade 3 and 4) in 1 eye and mild (grade 1) in the other. The eye that had IOL implantation later developed advanced visual loss, and the eye having surgery first retained good vision. The surgical methods were similar in both eyes. The lenses implanted later came in a different type package (Surefold system) than the earlier lenses. The Surefold system comprised a bay formed by 2 plastic legs that could be used as a folding forceps. Because only 2 patients had bilateral IOL calcification, meaningful statistical analysis could not be performed. Whether there was an IOL factor involved must be investigated further.
Delayed-onset IOL calcification is a recently described entity, many aspects of which remain to be clarified. Based on observation of a large case series, we defined its salient clinical features. All affected IOLs were Hydroview lenses, and the calcification developed 4 to 26 months after cataract surgery. The apparent peak occurrence was at 1 year, but further study is needed.
The amount of calcification and visual loss varied greatly. Visual loss ranged from 0 to 10 Snellen lines (average 2.8 Snellen lines). A high percentage of patients had systemic disease including diabetes mellitus, hypertension, and ischemic heart disease. Significant visual loss was noted regardless of the presence or absence of these diseases. Diabetic patients tended to have a more severe visual loss, probably because of complications of the disease. Ischemic heart disease also exacerbated visual loss; however, the reason is unknown.
We introduced a grading system to describe the variable amount of calcification. Grade 1 calcification was localized to forceps marks only, while grades 2, 3, and 4 were generalized calcification with increasing severity. The system showed a correlation with the amount of visual loss. Eyes with grade 1 calcification lost about 1 Snellen line of vision and those with grade 4, more than 7 lines. A large proportion of eyes had forceps marks on the IOL. These marks, together with the delayed onset, might provide clues to the pathogenesis of calcification. Visual loss was not related to disease duration but was more severe in eyes that developed calcification soon after surgery. This suggests that the calcification progressed more rapidly in some eyes. The factors leading to accelerated progression are unknown.
We found that the intraoperative use of acetylcholine did not increase visual loss. Further study is necessary to define the significance of other intraoperative agents. The importance of IOL material in promoting calcification must also be considered as certain hydrogels are known to have a high calcium affinity. Two patients had bilateral but asymmetric involvement. The IOLs with more advanced calcification came in a new lens packaging system. This emphasizes the importance of investigating the IOL's effect on the calcification process.
1. Bucher PJM, Büchi ER, Daicker BC. Dystrophic calcification of an implanted hydroxyethylmethacrylate intraocular lens. Arch Ophthalmol 1995; 113:1431-1435
2. Olson RJ, Caldwell KD, Crandall AS, et al. Intraoperative crystallization on the intraocular lens surface. Am J Ophthalmol 1998; 126:177-184
3. Yu AKF, Shek TWH. Hydroxyapatite formation on implanted hydrogel intraocular lenses. Arch Ophthalmol 2001; 119:611-614
4. Apple DW, Werner L, Escober-Gomez M, Pandey SK. Deposits on the optical surfaces of Hydroview intraocular lenses (letter). J Cataract Refract Surg 2000; 26:796-797
5. Lenis K, Philipson B. Lens epithelial growth on the anterior surface of hydrogel IOLs; an in vivo study. Acta Ophthalmol Scand 1998; 76:184-187
6. Sugawara A, Antonucci JM, Takagi S, et al. Formation of hydroxyapatite in hydrogels from tetracalcium phosphate/dicalcium phosphate mixtures. J Nihon Univ Sch Dent 1989; 31:372-381
7. Li P, Bakker D, van Blitterswijk CA. The bone-bonding polymer Polyactive 80/20 induces hydroxycarbonate apatite formation in vitro. J Biomed Mater Res 1997; 34:79-86
8. Taguchi T, Kishida A, Akashi M. Apatite formation on/in hydrogel matrices using an alternate soaking process: II. Effect of swelling ratios of poly(vinyl alcohol) hydrogel matrices on apatite formation. J Biomater Sci Polym Ed 1999; 10:331-339
9. Levy B. Calcium deposits on glyceryl methyl methacrylate and hydroxyethyl methacrylate contact lenses. Am J Optom Physiol Opt 1984; 61:605-607
10. Winter GD, Simpson BJ. Heterotopic bone formed in a synthetic sponge in the skin of young pigs. Nature 1969; 223:88-90
11. Schaumberg DA, Dana MR, Christen WG, Glynn RJ. A systematic overview of the incidence of posterior capsule opacification. Ophthalmology 1998; 105:1213-1221
12. Hollick EJ, Spalton DJ, Ursell PG. Surface cytologic features on intraocular lenses; can increased biocompatibility have disadvantages? Arch Ophthalmol 1999; 117:872-878
13. Ravalico G, Baccara F, Lovisato A, Tognetto D. Postoperative cellular reaction on various intraocular lens materials. Ophthalmology 1997; 104:1084-1091
14. Kohnen T, Magdowski G, Koch DD. Scanning electron microscopic analysis of foldable acrylic and hydrogel intraocular lenses. J Cataract Refract Surg 1996; 22:1342-1350