Indian Journal of Ophthalmology

: 1995  |  Volume : 43  |  Issue : 4  |  Page : 185--190

Capsulorhexis: Its safe limits

Abhay Vasavada, Jagruti Desai 
 From Iladevi Cataract & Intraocular Lens Research Centre, Raghudeep Eye Clinic, Ahmedabad, India

Correspondence Address:
Abhay Vasavada
M.S., Iladevi Cataract & Intraocular Lens Research Centre, Raghudeep Eye Clinic, Gurukul Road, Memnagar, Ahmedabad 380 052


We undertook this study to determine the safe limits of capsulorhexis during nucleus expression in 40 eyes of patients undergoing extracapsular cataract extraction (ECCE) with a posterior chamber intraocular lens (PC IOL) implantation and in 30 cadaver eyes. In group I (patient eyes), capsulorhexis of 4.5 to 7.5 mm was performed and the nucleus was expressed by hydrodissection. The nuclei measured 4.5 to 9.0 mm. One relaxing incision at 12 o«SQ»clock position had to be placed in 9 patients. In group II (cadaver eyes), continuous curvilinear capsulotomies of 4.0, 4.5, 5.0, 5.5, 6.0 and 6.5 mm were made in 5 eyes each. No relaxing incisions were placed. In both the groups, nuclei of all sizes could be safely delivered through intact capsulotomies measuring 5.5 mm or more. In two patient eyes, posterior capsule rupture occurred with rhexis measuring 4.5 and 5.0 mm, respectively. In the cadaver eyes, intracapsular extraction occurred in 4 eyes with rhexis measuring 5.0 mm or less. We conclude that a rhexis less than 5.5 mm is not safe for nucleus delivery during ECCE.

How to cite this article:
Vasavada A, Desai J. Capsulorhexis: Its safe limits.Indian J Ophthalmol 1995;43:185-190

How to cite this URL:
Vasavada A, Desai J. Capsulorhexis: Its safe limits. Indian J Ophthalmol [serial online] 1995 [cited 2023 Feb 2 ];43:185-190
Available from:

Full Text

Capsulorhexis is the only form of capsulotomy that leaves a 'true' capsular bag with all its advantages.[1][2][3][4]However, as the strong continuous margin of the rhexis does not tear, the mechanics of nucleus removal through a rhexis are different. The nucleus has to be expressed by fluid pressure within the bag without putting any stress on the zonules.[5] Hydrodissection or injection of fluid underneath the capsule breaks the adhesions between the cortex and the capsule.[6],[7] Hydrodelineation or fluid injection between the hard nucleus and the epinucleus can also facilitate nucleus removal through a small capsulotomy. The margin of the rhexis has an elasticity which allows safe expression of a nucleus larger than the opening.[8][9][10][11][12] However, beyond certain limits, disproportion between the size of the rhexis and the nucleus can lead to complications.[13][14][15][16] To define the safe limits of the size of capsulorhexis, we undertook a controlled study in patient as well as cadaver eyes.


This prospective study was carried out in two groups. Group I consisted of 40 eyes of patients undergoing ECCE with in-the-bag PC IOL implantation and Group II consisted of 30 cadaver eyes.

In group I, the following patients were excluded from the study: patients with any other eye diseases like iritis, pseudoexfoliation or glaucoma; those in whom the pupil did not dilate or in whom a complete rhexis was not done; and patients with systemic diseases like diabetes, hypertension, and ischaemic heart disease.

 Surgical Procedure

Group I: A posterior corneal section was used. After making a groove the anterior chamber was entered through a 3-mm wide incision to the right of the midline. The anterior chamber was filled with 2% methylcellulose. A capsulorhexis was performed with a 26-gauge needle cystitome. The size of the rhexis was measured with a caliper through the cornea. The intended diameter of the rhexis was within the range of 4.5 to 7.0 mm and in the individual case, was decided by the surgeon's judgement of the size and hardness of the nucleus.

Hydrodissection was then performed with a 27-gauge cannula fitted to a 5-ml syringe filled with balanced salt solution (BSS). The cannula was introduced into the anterior chamber and just underneath the capsular rim at 7 o'clock position. The surgeon's hand was steadied and BSS injected. A fluid wave could be seen going behind the nucleus. Injection was continued until one part of the equator of the nucleus was forced out of the rhexis.

A relaxing incision was made in cases in which part of the nucleus was not forced out of the rhexis. This was done by making multiple punctures with the cystitome at the margin of the rhexis at 12 o'clock position.

The part of the equator of the nucleus which was out of the capsular bag was then rotated and the nucleus was prolapsed into the anterior chamber. The corneal incision was enlarged, titrating it against the size of the nucleus. The nucleus was expressed by injection of fresh viscoelastic material into the anterior chamber or by pressure on the cornea at 6 o'clock position. The diameter of the reduced size of the nucleus was measured.

Group II: Group II consisted of 30 cadaver eyes, in which, the corneal endothelium was considered nonviable for keratoplasty. The eyes were dissected about 48 to 72 hours after enucleation. The eyeball was stabilised by a roll of gauze and with a clamp. An incision was made just inside the limbus and the corneal button was excised. The iris was removed by cutting it at the root. Rhexis was performed as in the patient eyes with a cystitome.

Continuous curvilinear capsulotomies of 4.0, 4.5, 5.0, 5.5, 6.0 and 6.5 mm were made in 5 eyes each. Unlike in group I eyes, the rhexis was measured directly with a caliper. Hydrodissection was performed in a similar manner until the nucleus tilted so that one part of the equator was out of the rhexis. The nucleus was then rotated out of the capsular bag. In cases in which the nucleus was too big and did not tilt out of the rhexis, relaxing incisions were not made; but, fluid injection was continued until intracapsular extraction of the lens occurred. The nucleus size was measured in all cases.


In Group I of 40 patient eyes the patient age ranged from 36 to 80 years (mean, 60.4 years). There were 24 males and 16 females. The distribution of the type of cataract is shown in [Table:1].

The size of the rhexis ranged from 4.5 to 7.5 mm [Table:2]. Eight patients had a rhexis size of 5.0 mm or less and 32 patients had a rhexis size of 5.5 mm or more.

The size of the nuclei ranged from 4.5 to 9.0 mm (average, 7.82 mm). In 9 patients, nuclei were of 9.0 mm diameter. Interestingly, in one patient (case no. 16, [Figure:1], such a large nucleus was expressed through a 5.5 mm rhexis.

[Figure:1] shows the size of the rhexis and the corresponding nucleus size in each case. The disparity between the size of the nucleus and the rhexis was calculated in each case as the disparity ratio. This was defined as the diameter of the nucleus in millimetre by the diameter of the rhexis in millimetre. In 31 patients the nuclei could be delivered through an intact rhexis, while in 9 patients one relaxing incision had to be placed. The disparity ratio ranged from 0.9 to 1.64 in the 31 cases with an intact rhexis.

Relaxing incisions were judged necessary in 9 (22.4%) patients [Table:2]. The incidence of "one relaxing incision" among rhexis of < 5.5 mm and > 5.5 mm was compared statistically using chi-square test with Yates' correction. The difference was highly significant (p = 0.0106) which justifies the cut off point of 5.5 mm as a safe lower limit of rhexis. Of these nine patients with relaxing incisions, two patients (case nos. 4 and 8) with rhexis measuring 4.5 and 5.0 mm, respectively developed zonular dialysis and posterior capsule rupture during nucleus expression. The nucleus size in these two cases was 9.0 mm. The disparity ratio in these two cases was 2.0 and 1.8, respectively [Figure:1]. The incidence of posterior capsule rupture among rhexis of < 5.5 mm and > 5.5 mm were compared statistically using chi-square test with Yates' correction. The difference was found to be statistically significant (p < 0.05).

In group II of 30 cadaver eyes, the age at death of the donors ranged from 55 to 80 years (average, 63.2 years).

The six groups of eyes with rhexis diameters of 4.0, 4.5, 5.0, 5.5, 6.0 and 6.5 mm and the corresponding nucleus diameters are shown in [Figure:2]. No relaxing incision was given to any of the cadaver eyes. In 26 eyes the nucleus could be safely delivered through an intact rhexis, while in 4 eyes intracapsular extraction resulted. The disparity ratio in the 26 cases ranged from 0.9 to 1.63 [Figure:2], while in 4 eyes where intracapsular extraction resulted, it ranged from 1.8 to 2.0.

The nucleus size ranged from 4.0 to 9.0 mm (average, 7.4 mm). Large nuclei measuring 9.0 mm were present in 3 eyes. Of these, in 2 eyes (case nos. 20 and 30, [Figure:2], the nuclei could be expressed through capsulotomies of 5.5 and 6.0 mm, respectively. In the remaining eye (case no. 15, [Figure:2], delivery of the nucleus was attempted through a rhexis of 5.0 mm. Total zonular dialysis occurred, resulting in intracapsular extraction. In another 3 cases, intracapsular extraction occurred. These 3 eyes had a rhexis size less than 5.0 mm. (case nos. 5,9,10; [Figure:2].


The major difficulty during manual expression of the nucleus from the capsular bag into the anterior chamber is the resistance offered by a very strong capsulorhexis margin. It is important to have the smallest size of the capsulorhexis that is compatible with safe nucleus delivery.

Using capsulorhexis with ECCE, several authors have reported complications like difficulty in nucleus expression,[13] intracapsular extraction of the lens,[14]zonular rupture[15] and extension of relaxing incision causing posterior capsule rupture.[16]

Various studies have tested the elasticity of the rhexis by stretching the opening of the rhexis with a micrometer screw gauge or a 2-lens manipulator hooks. The stretching capacity has been reported as 206% (125 to 300%) increase in area in pig eyes[8] and 223% (109 to 336%) increase in area and a mean increase in diameter from 5.2 to 13.5 mm in human cadaver eyes.[9] Two other studies[10,11] on eyebank eyes have demonstrated that the circumference of the intact capsulorhexis could be enlarged by an additional 62% (range, 15 to 100%) before a radial tear occurred. Another study[12] on human cadaver eyes states that a rhexis of between 4.0 and 5.0 mm can be stretched to deliver almost any size of nucleus. This, however, is not the case in clinical practice. Hence, in this study we attempted to determine the consequences of a disproportion between the diameter of the rhexis and the nucleus. We tried to reproduce the clinical situation in cadaver eyes and nucleus expression was done in the same way as in the patient eyes. We paid careful attention to keep the procedure identical in both the groups.

In group I, out of 8 eyes with a small rhexis opening, 2 developed posterior capsule rupture despite our placing one relaxing incision. Both the eyes had a nucleus size of 9.0 mm. In 32 eyes with rhexis diameter 5.5 mm or more, none developed posterior capsule rupture. Seven patients had nuclei of 9.0 mm diameter. The difference in the incidence of posterior capsule rupture among the two groups was statistically significant (p < 0.05).

It is interesting to note that out of the 8 eyes with small rhexis, relaxing incisions had to be placed in 5 (62.5%). In the remaining 3 eyes, with the help of hydrodelineation, the size of the nucleus was reduced and delivered safely. On the other hand, in 32 patients with rhexis larger than 5.5 mm, only 4 (12.5%) needed a relaxing incision. The difference in the incidence of relaxing incisions among the two groups was statistically significant (p = 0.0106).

We determined 5.5 mm to be a safe lower limit for rhexis size because the largest diameter nucleus in our study could be delivered through it. In order to statistically determine the value which would prevent both a relaxing incision and a posterior capsule rupture. regression analysis based on disparity ratio and size of capsulorhexis was performed. The correlation coefficient between rhexis and disparity ratio was -0.4166 (n = 40, p < 0.01). The data of patients with one relaxing incision was subjected to further scrutiny. The range of disparity ratio in such patients was 1.2 to 2.0. The coefficient of correlation between the size of rhexis and disparity ratio in such patients is r = -0.6848 (n = 9, p < 0.1). Although the correlation coefficient was not significant statistically, it was fairly high indicating negative correlation between the size of the rhexis and disparity ratio. The predicted value of the least disparity ratio, i.e., 1.2 using the regression formula comes to 5.70. This shows that 5.70 would be safe lower limit to prevent a relaxing incision and a posterior capsule rupture and agrees well with our limit of 5.50.

We attempted to quantify the relation between the size of the rhexis and the nucleus. The ratio of the diameter of the nucleus over the diameter of the rhexis is a fair index of disparity at the opening. The higher the ratio, more are the chances of complications. In the current study, when the integrity of the opening could be maintained without the need for a relaxing incision, the ratio ranged from 0.9 to 1.64. In simple words, the rhexis opening can stretch to allow the expression of a nucleus about one-and-a-half times its size. In the two eyes in which posterior capsule rupture occurred, the ratio was 2.0 and 1.8, respectively.

A study[17] of 210 patients using a similar technique of hydroexpression of the nucleus has reported zonular dialysis during nucleus delivery in 2 (0.95%) patients when using capsulorhexis of 5.5 to 6.0 mm diameter. Another study[6] reported safe nucleus delivery in 26 patients with 5.0 to 7.0 mm rhexis diameter. The authors found that the increase in the area of the capsulorhexis to allow the passage of the nucleus was between 11 and 145%.

In group II, intracapsular extraction resulted in 4 eyes. All the 4 eyes (case nos. 5, 9, 10, 15; [Figure:2] had a rhexis of 5.0 mm or less. The size of the nucleus was between 8.0 and 9.0 mm. The hydrodelineation failed to reduce the size of the nucleus and therefore the disparity ratio for the four cases was 2.0, 1.8, 1.9 and 1.8, respectively.

The rhexis is magnified by about one eighth when it is measured through the cornea.[18] Thus, a rhexis of 5.5 mm in the patient and cadaver eyes is not exactly comparable. The measurements are also not comparable because of postmortem changes in elasticity. The difference in the findings in the two groups may also be due to the fact that hydrodelineation effectively reduced the nucleus size in the patient eyes. In the cadaver eyes we continued fluid injection under the anterior capsule until the nucleus prolapsed or intracapsular extraction occurred and no attempt was made to purposefully reduce the size of the nucleus.

Our study was not without limitations. The sample size in vivo with rhexis < 5.5 mm was small. With clinical experience, we realised that a small-sized rhexis could be unsafe. Another limitation of the study was that only the diameter of the nucleus was measured. The circumference and the thickness of the nucleus are important parameters that act on the rhexis margin. Measuring all these three would have been better.

Our observations suggest that a rhexis of 5.5 mm is a safe lower limit for manual nucleus expression. We recommend measuring the rhexis opening routinely during ECCE. This would help the surgeon to anticipate the disparity at the opening. Hydrodissection and delineation are of great help. One relaxing incision is not a guarantee against posterior capsule rupture. Perhaps two relaxing incisions may be safer.


We are thankful to Mr. Gautam Majmudar, the Eyebank Incharge, Indian Red Cross Society, Dholka Branch, and Professor Dr. A.N. Setalvad for carrying out the statistical analysis.


1Gimbel HV, Neuhann T. Development, advantages, and methods of the continuous circular capsulorhexis technique. J Cataract Refract Surg 16:31-37, 1990.
2Blumenthal M, Assia E, Neumann D. The round capsulorhexis and the rationale for an 11 mm diameter IOL. Eur I Implant Ref Surg 2:15-19, 1990.
3Legler UFC, Assia EI, Castaneda VE, Hoggatt JP, et al. Prospective experimental study of factors related to posterior chamber IOL decentration. J Cataract Refract Surg 18:449-455, 1992.
4Miyake K, Asakura M, Kobayashi H, et al. Effect of intraocular lens fixation on the blood-aqueous barrier. Am J Ophthalmol 98:451-455, 1984.
5Assia E, Blumenthal M, Apple DJ. Hydrodissection and viscoextraction of the nucleus in planned extracapsular cataract extraction. Eur J Implant Ref Surg 4:3-8, 1992.
6Thim K, Krag S, Corydon L. Capsulorhexis and nucleus expression. Eur J Implant Ref Surg 2:37-41, 1990.
7Fine IH. Cortical cleaving hydrodissection. J Cataract Refract Surg 18:508-512, 1992.
8Krag S, Thim K, Corydon Leif. The stretching capacity of capsulorhexis: An experimental study on animal cadaver eyes. Eur J Implant Ref Surg 2:43-45, 1990.
9Thim K, Krag S, Corydon Leif. Stretching capacity of capsulorhexis and nucleus delivery. J Cataract Refract Surg 17:27-31, 1991.
10Assia EI, Apple DJ, Tsai JC, Lim ES. The elastic properties of the lens capsule in capsulorhexis. Am J Ophthalmol 111:628-632, 1991.
11Assia EI, Apple DJ, Morgan RC, Legler UFC, et al. The relationship between the stretching capability of the anterior capsule and zonules. Invest Ophthalmol Vis Sci 32:2835-2839, 1991.
12Tana P, Belmonte J. Elasticity of the capsulorhexis and delivery of the nucleus. Eur J Implant Ref Surg 5:103-108, 1993.
13Maher JE Nucleus expression after capsulorhexis. (Letter). J Cataract Refract Surg 14:693, 1988.
14Almallah OF. Capsulorhexis complications with planned ECCE. (Letter). J Cataract Refract Surg 15:232-233, 1989.
15Mackintosh GIS. Complications of capsulorhexis using ECCE in dense cataracts. Eur J Implant Ref Surg 5:82-87, 1993.
16Lim L, Wong D, Lim ASM. Posterior capsular tears from relaxing incisions in CCC during ECCE. Asia Pacific Journal of Ophthalmology, 5:14-17, 1993.
17Pande M. Continuous curvilinear (circular) capsulorhexis and planned extracapsular cataract extraction - are they compatible? Br J Ophthalmol 77:152-157, 1993.
18Duke-Elder S. System of Ophthalmology, Vol. VII. St. Louis, CV Mosby Co., 1962, pp. 328.