Efficacy of Intracameral Mydriatics in Pediatric Lens Surgery

Pediatric lens disease is rare, and presents in many varieties, which makes surgical treatment challenging [1]. Maximal pupillary dilatation is preferable and facilitates the surgery.

This is currently obtained by instillation of eye drops such as tropicamide, cyclopentolate, or phenylephrine. Tropicamide prevents a contraction of the sphincter pupillae by an anticholinergic effect on the M4 muscarinic receptors [2]. Phenylephrine has a synergistic action via an alpha-1-agonistic pathway and contracts the iris dilator muscle [3]. Phenylephrine can increase the blood pressure and cause reflex bradycardia through vasoconstriction, whereas tropicamide can cause anticholinergic side effects [2, 3]. Absorption by the conjunctiva and the nasolacrimal mucosa can increase the risk of side effects [4, 5]. Additional drops can be required to ensure proper dilatation, and result in an increase of the administered dose [4]. This overexposure increases the risk of adverse effects [3]. Administering eye drops is not always straightforward and might cause stress to both the child and their parents. When a child is not properly dilated at the start of surgery, it is difficult to determine whether this is due to insufficient contact time or a limited effect on the target tissue. There is few data on the use of mydriatics during pediatric lens surgery, as most of the available data concerning mydriatic eye drops in children comes from the screening programs for retinopathy of prematurity [2, 6, 7].

A preparation for injection into the anterior chamber of the eye has been approved for adult use (Mydrane^®, Laboratoires Théa, France) [8, 9]. It contains a combination of tropicamide (0.04 mg/0.2 mL), phenylephrine (0.62 mg/0.2 mL), and lidocaine (2 mg/0.2 mL). The local administration allows for a lower dosage of the active substances than in eye drops.

Three prospective studies were recently published on the use of intracameral mydriatics in a pediatric population [10‒13]. The first study only included children who responded well to preoperative mydriatics and reported 75% of patients reaching a minimal pupil diameter of 6 mm in a paper reporting preliminary results [10]. The full results of this trial were recently published, with 78 eyes out of 75 patients reaching sufficient mydriasis (93.5% success rate) [11]. Another study by the same group compared the use of topical mydriatics to intracameral injection of mydriatics. They reported a slightly smaller pupil size after intracameral injection which was nonetheless sufficiently large to proceed with surgery [12]. A second study group reported sufficient pupillary dilatation to perform the anterior capsulorhexis in 98.1% of patients [13]. No side effects related to the intracameral mydriatic occurred in either study.

With this study, we aimed to evaluate the efficacy and ocular side effects with the use of ICM in a consecutive series of children undergoing lens surgery, without excluding children who might be expected to respond poorly, such as children with persistent fetal vasculature (PFV), lens luxation, or microphthalmos. This study is part of a larger study (the MyKid study), where we additionally evaluated the occurrence of systemic side effects and serum levels of tropicamide and phenylephrine. Those results will be reported elsewhere by the members of this team.

We started a single-center prospective cohort trial to evaluate the efficacy and safety of intracameral Mydrane in a pediatric population. We aimed to include 40 consecutive patients between 8 weeks and 17 years of age who were planned for lens surgery, including cataract, PFV or lens subluxation. Both unilateral cataracts and bilateral cataracts were included.

Patients with a history of diabetes, uveitis, eye trauma, or corneal disease were excluded from the trial. Children with cardiac conduction disorders, congestive heart failure, bradycardia, a reported allergy for one of the components of Mydrane, impaired respiratory function or impaired renal function (creatinine clearance below 10 mL/min) were excluded from the trial as well.

Girls post puberty underwent a pregnancy test to confirm the absence of pregnancy at the time of informed consent. Additional information and support regarding the testing were provided at this time.

The inclusion period of the study was from October 2020 to November 2021. Forty patients were included, and 53 eyes were operated following the study protocol.

For this part of the study, we randomized one eye of the bilateral cases to reach a study population of 40 eyes of 40 different patients (Table 1). Written informed consent was obtained from the parents or legal guardians of every patient, as well as from the patients themselves when aged 6 years or older.

This study protocol was reviewed and approved by the Ethical Committee Universitair Ziekenhuis Antwerpen and Universiteit Antwerpen (UZA-UA) under the approval number EDGE001169. The study was also registered in the EUCT-Database under the number 2023-504173-21-00, and thus additionally reviewed and approved by an independent Belgian Ethical Committee. The study was performed following the principles of the Declaration of Helsinki.

After disinfecting and draping the eye following standard protocol, we made a temporal corneo-limbal incision between 1.1 and 2.65 mm wide, depending on the age of the patient. 0.1 mL of intracameral Mydrane was then injected into the anterior chamber. A cohesive ophthalmic viscoelastic device (OVD) (Healon GV Pro, Johnson and Johnson, USA) was injected 30 s later. If the pupil was not sufficiently dilated to proceed with the surgery, we first tried to enhance pupillary dilatation by iris stretching. If the pupil size remained insufficient, we either placed 4 iris retractors in smaller eyes, or used a pupil dilating device (Malyugin ring, MST, USA) in larger eyes. In young children and lens subluxation cases we then made a 1.1 mm side port. A caliper ring (Morcher type 4L, Morcher GmbH, Germany) was inserted into the anterior chamber and positioned on the anterior capsule to indicate size and centration of the capsulorhexis. When deemed necessary, the capsule was stained with trypan blue (Vision Blue, Dorc, The Netherlands). After making an anterior continuous curvilinear capsulorhexis, we removed the crystalline lens by either bimanual irrigation/aspiration in the children <1 year or coaxial irrigation/aspiration in older children. We then filled the anterior chamber with cohesive OVD, making sure not to fill the capsular bag but positioning the remainder of the anterior capsule close to the posterior capsule.

We made a small break in the posterior capsule by dragging the capsule sideways with the tip of a 25G-needle. We then injected light chain cohesive OVD (Healon Pro, Johnson and Johnson, USA) behind the posterior capsule with a 41 G subretinal needle to push the anterior hyaloid backwards and separate it from the posterior capsule. As soon as the delineation of the ligament of Wieger became visible, a posterior continuous curvilinear capsulorhexis was made with a microforceps.

We ensured whether the space of Berger was sufficiently developed and large enough to accommodate the intraocular lens (IOL) by anterior segment optical coherence tomography (Zeiss Callisto, Carl Zeiss AG, Germany) [14]. We only performed an anterior vitrectomy in of an abnormal or increased adhesion of the anterior hyaloid to the posterior capsule.

When using small incisions, we then enlarged the temporal incision to 2.65 mm. The IOL (Morcher type 89A or 89D, Morcher GmbH, Germany) was folded into an IOL-injector (Accuject, Medicel, Switzerland) filled with OVD (Healon Pro, Johnson and Johnson, USA). The maximal diameter of the Morcher type 89A is 7.5 mm, the 89D type has a maximal diameter of 6.5 mm.

The IOL was then injected into the anterior chamber and both capsulorhexis blades were positioned in the groove surrounding the lens optic between both haptics using OVD (Healon Pro, Johnson and Johnson, USA) (bag-in-the-lens technique) [15]. Surgery was then concluded by injection of 0.5 mL of 20% diluted carbachol (Miostat, Alcon, USA) into the anterior chamber, aspiration of remaining OVD and closure of all incisions using either Nylon 10/0 (in children under 1 year of age) or Vicryl 10/0 in older children.

All the surgeries were performed by two surgeons (L.V.O. and M.J.T.) and were recorded. We had missing videos for 4 cases due to limited disk space on the microscope (n = 1) or unsuccessful recording (n = 3). The video registration was incomplete for 4 cases, due to either late start of the video recording (n = 3) or premature ending of the video (n = 1).

We defined four time points on which we measured the pupillary diameter: T0 immediately before incision; T1 before starting the anterior capsulorhexis (or before mechanical iris dilatation in case of insufficient dilatation); T2 before injection of the lens and T3 before injection of carbachol. At T1, the intraocular caliper ring (Morcher 4L, with a standard outer diameter of 4.8 mm and corresponding inner diameter of 4.5 mm) was used as a guide to scale the pupil size. By scaling the images of T0, T2, and T3 based on the T1 images using image processing software (Adobe Illustrator, Adobe, USA), we were able to accurately measure the pupil size at all four defined stages in the video recording. The pupil size was measured in both the horizontal and vertical axis and the average of both was used as the pupil diameter. An example of the scaling at the various predefined time points can be seen in Figure 1. In case additional manipulations (iris stretching) or instruments (iris retraction hooks, pupil expansion rings (Malyugin ring, MST, USA) were used, we did not perform the measurements at T2 and T3, as they would not reflect the efficacy of the Mydrane. In these cases, we defined T1 as the pupillary size immediately before the mechanical pupillary dilatation. We defined “duration of mydriasis” as the time elapsed between injection of Mydrane at the beginning of the surgery and the injection of carbachol at the end, thus reflecting the duration of the surgery.

Considering a minimal increase of 4 mm in pupil size compared to the pupil size pre-administration of Mydrane, 34 patients were deemed necessary to be included to have 80% power to detect a significant increase in pupil size at a two-sided significance level of 0.05. We hereby anticipated a standard deviation of the change in pupil size of at most 8 mm, and thus a minimum effect size of 0.5. To compensate for possible drop-out of the study, the sample size was increased to 40 participants.

Patient characteristics and intervention outcomes (successful dilatation, duration of mydriasis, iris prolapse) were summarized as number and percentages for categorical variables and as mean with standard deviation and median with range for continuous variables. Subgroups were compared by Mann-Whitney U test for continuous variables and chi-square test or Fisher’s exact test for categorical variables.

The evolution of the pupil size was studied in a linear mixed effects model, with time as a categorical fixed effect (4 time points: before the first corneal incision (T0) – before anterior capsulorhexis (T1) before IOL insertion (T2) – before Carbachol injection (T3)). A random intercept per patient was added to account for the dependence among observations of the same patient. Patients with insufficient dilatation with Mydrane alone were excluded from T2 onwards. In case of a significant effect of time, post hoc pairwise comparisons were performed to assess differences between the time points. Bonferroni-holm corrections were used to correct for multiple testing.

The impact of baseline patient characteristics on the evolution of the pupil size was studied by adding these as fixed effects and in interaction with time to the mixed effects model. The following characteristics were considered: age, gender, presence of PFV, type of cataract and corneal diameter. To avoid overfitting, each of these characteristics was studied in a separate model.

The relation between duration of mydriasis and pupil diameter was studied by Spearman rho correlation coefficient. All statistical analyses were performed in R version 4.1.3. p values below 0.05 were considered significant.

We included 40 eyes of 40 patients: 20 girls and 20 boys between 2 months and 16 years old (median 16.4 months). Further patient characteristics can be found in Table 1.

We did not need any additional manipulations to obtain sufficient pupillary dilatation to perform the surgery in 30 patients (75% [95% CI: 59–87%]). Three eyes were left aphakic because of insufficient capsule support for IOL implantation. The power of the implanted IOL varied between 16.5 and 31 Diopters (median 29 D). We did see a certain degree of pupillary dilatation in the remaining 10 patients (25%), indicating a partial response, but the pupil remained too small to continue the surgery without additional surgical maneuvers. In this partial response group, the mean pupil size at T0 was 0.99 mm (SD 0.30), which increased to a mean of 3.54 mm (SD 1.04) at T1 (p = 0.031). We stained the anterior capsule with trypan blue in 19 patients (47.5%).

We found a significant difference in pupil size before Mydrane injection (T0) between the success group (mean pupil size = 1.46 mm, SD 0.39) and the partial success group (mean pupil size = 0.99 mm, SD 0.30) (p = 0.005). None of the other patient characteristics significantly differed between these groups (Table 2).

The duration of the surgery as reflected by the “duration of mydriasis” was longer for the partial success group compared to the success group: 63.6 min (SD 19.8) versus 46.0 min (SD 21.7), p = 0.029 (Table 2). This is further emphasized by a significant relation between the “duration of mydriasis” and the pupil size at T1 (Spearman correlation rho = −0.50, p = 0.002), with smaller pupil sizes resulting in longer surgery duration.

There was a higher incidence of iris prolapse in the partial success group (40%) when compared to the success group (10%) with borderline no significance (p = 0.052). The patients with iris prolapse were significantly (p = 0.008) younger (median age 3.5 months [range 1.9–5.3]) than those without iris prolapse (median age 26.8 months [range 1.8–192]) and had a smaller corneal diameter (mean 10.3 mm [SD 0.57] with iris prolapse versus mean 11.1 mm [SD 1.06] without iris prolapse, p = 0.020) (Table 3). The maximum age at which we recorded an iris prolapse was 5.3 months.

There was a significant increase in pupillary diameter between T0 and T1 (diameter_T0 = 1.37 mm [SD 0.41] to diameter_T1 = 5.27 mm [SD 1.20] p < 0.001), with a further significant increase between T1 and T2 (diameter_T2 = 5.95 mm (SD 0.99) p < 0.001). There was no significant evolution in diameter between T2 and T3 (diameter_T3 = 5.95 mm [SD 0.95] p = 0.97) (Fig. 2).

Age had a significant impact on the pupil size: children of 6 months or younger had a significantly smaller pupil at baseline (T0) than children older than 6 months (diameter_T0,<6m = 1.16 (SD 0.29) versus diameter_T0,>6m = 1.49 (SD 0.43), p = 0.031). This difference further increases over the time points T1, T2 and T3 (Fig. 2, p < 0.001).

We also noted a significant relation between gender and evolution of the pupillary dilatation. Girls had a larger increase in pupillary diameter than boys at any of the time points (+0.51 mm at T1 [p = 0.068], +0.66 mm at T2 [p = 0.005] and +0.63 mm at T3 [p = 0.011] (Fig. 2)). It should be noted that the mean age of the boys was lower (10.53 months [range 1.8–152.6]) than that of the girls (23.8 months [range 1.8–191.5]), albeit not significant (p = 0.49). Half (50%) of the boys were younger than 6 months, compared to 30% for the girls (p = 0.33). Due to the low numbers in each group (n = 20), the power of this study is insufficient to detect a difference here. However, there was no difference between both genders in pupil diameter at baseline (T0: boys median 1.29 mm [range 0.75–3.14] vs. girls median 1.43 mm [0.59–1.92], p = 0.094), nor was there a difference in corneal diameter (boys median 11.25 mm [range 9–12.4], girls median 11.0 mm [range 8.25–12.7], p = 0.64), which does point toward a certain clinical relevance.

We divided the group in five categories based on type of lens disease: anterior, fetal nuclear, posterior cataracts, presence of PFV and lens luxation. For analysis, we grouped these into two categories: one with a normal vitreolenticular interface (grouping the anterior, fetal nuclear and lens luxation cases) and another with an abnormal vitreolenticular interface (consisting of posterior cataracts and PFV cases) [16]. We could not demonstrate a significant difference in pupil size or effect over time between these groups (Fig. 2).

For horizontal corneal diameter, we found a significant relation to the increase in pupillary diameter. The greater the corneal diameter, the greater the increase at T1, T2, and T3 (p < 0.001). Even at baseline there is a larger pupil diameter in eyes with larger horizontal corneal diameters (estimate [se] of the baseline difference based on mixed effects model 0.17 [0.081] per mm difference in corneal diameter, p = 0.039). These findings remain significant after correction for the age of the patients. For visualization of the differences in baseline pupil size and evolution over time, the horizontal corneal diameter is split in 4 more or less equally sized groups (≤10, 10 < x ≤ 11, 11 < x ≤ 12, >12 mm) (Fig. 2).

Regarding axial length, we found a significant correlation to the increase in pupillary diameter. The greater the axial length, the more pronounced the increase is at T1, T2, and T3 (p < 0.001). As for corneal diameter, this is probably influenced by age because this is strongly correlated with axial length (Spearman correlation ρ = 0.79, p < 0.001). After correction for age (younger or older than 6 months), a significant effect of axial length on the evolution of the pupillary diameter remains. However, because of the small number of participants in this study, these results should be interpreted with care and should be confirmed by future studies.

Twenty-two patients had different shades of blue eyes, 18 had variations of brown eyes, bearing in mind that the young children might not have had their final iris pigmentation at the time of surgery. We did not find a significant correlation between iris color and pupil size.

We found a 75% success (95% CI, 59–87%) in our cohort obtaining sufficient pupillary dilatation to perform lens surgery without the help of mechanically dilating devices. The dilatation lasted for the entire duration of the surgery in all these cases. In the partial success group, the surgery lasted longer, reflecting the higher complexity of these eyes, including the time needed for additional maneuvers to dilate the pupil. In all the cases where additional surgical steps or dilatation tools were needed, there was a limited but clinically insufficient effect of the Mydrane injections. We did not administer a second dose of Mydrane in any of the operated eyes and did not encounter any adverse events.

Mydrane has advantages over other alternatives to preoperatively obtain pupillary dilatation, such as epinephrine in the irrigating solution or as an intracameral injection. First, it combines two active substances with a different mode of action. Second, there is no need for dilution. Finally, there is maximal control of the total dose, contrary to epinephrine in the irrigation solution.

We found that the group with successful dilatation had a significantly larger pupil at baseline than the non-success group (Table 2). The pupillary diameter at baseline might thus be a good prognostic parameter for the effect of Mydrane. We also saw an increased incidence of iris prolapse in the partial success group (40% vs. 10%) (Table 2) but in this small sample, no significance could be shown. The age of the children in the partial response group was lower, so this effect is possibly due to age rather than response: it could be a consequence of the less mature irises in this age group, making them more prone to iris prolapse.

We observed a significant increase in pupillary size between T0 and T1, with a further significant increase between T1 and T2 but not anymore between T2 and T3 (Fig. 2). This agrees with the results in the adult population [8], taking the longer duration of the pediatric cataract surgery as compared to adult surgery into account. Even in the partial responder group, there was a significant increase in pupil size between T0 and T1. This indicates that there is a certain action of Mydrane in all children, but that there are factors limiting the effect in the partial success group. This is a consequence of the complexity of pediatric lens surgery, which more often presents with abnormalities such as PFV, interface abnormalities and abnormal biometrical parameters such as microphthalmia [16].

We noticed a larger baseline pupil size and a greater increase in pupil diameter in children over 6 months old when compared to children under 6 months. This can be due to the iris in the younger age group still being immature and less responsive to the mydriatics. Furthermore, these eyes often have a smaller anterior chamber, which limits the maximally obtainable pupil size. This is reflected by our finding of a positive relation between the increase in pupillary diameter after Mydrane administration and the horizontal corneal diameter (Fig. 2). When the anterior chamber is very small, it will be difficult to obtain a sufficient pharmacological mydriasis.

Furthermore, we found a significantly larger increase in pupil size in girls compared to boys while the pupil size at baseline was not significantly different. Because of the rarity of pediatric cataract and the presumed greater safety of intracameral Mydrane over topical mydriatics, we have not included a control group in this study. Thus, we cannot compare the efficacy of intracameral Mydrane to topical mydriatics. However, we do consider a 75% success rate in pupillary dilatation a very good result because of the high rate of ocular comorbidities limiting pupil dilatation in this patient population.

Our results are in line with two recently published studies that demonstrated a success rate between 75 and 98.1% [10‒13]. Contrary to these studies, we did not exclude associated ocular pathology nor insufficient preoperative response to mydriatics, as we wanted our cohort to reflect the true efficacy of ICM in any kind of pediatric lens surgery. We also used a ring caliper in the anterior chamber which enabled an accurate measurement of the pupil size. This is an important difference to the aforementioned studies, where the pupil size was measured by referring to the corneal diameter, which can result in some variation in the true pupillary diameter considering the curvature of the cornea.

Intracameral mydriatics are currently used off-label in pediatric cases. Because pediatric cataract is a rare disease, it is valuable to have results from several trials. The fact that our results is in line with two previously performed studies increases the safety profile of Mydrane in the pediatric population.

We conclude that using ICM instead of mydriatic eye drops can improve the experience of both the child and their parents, without compromising the quality of the surgery. Furthermore, it eliminates the stress for the surgical team if the child is not properly dilated at the beginning of surgery and it shortens the time needed for preoperative preparation as well as the time of narcosis.

Intracameral injection of 0.1 mL of Mydrane, a fixed combination of tropicamide (0.04 mg/0.2 mL), phenylephrine (0.62 mg/0.2 mL) and lidocaine (2 mg/0.2 mL), can be safely used in children and shows a good and sustained efficacy. Smaller pupils at baseline, younger age, male sex and small horizontal corneal diameter were related to a poorer response to Mydrane. The use of intracameral mydriatics decreases the need for preoperative eye drops and as a result diminishes stress for both the children and their parents.

The authors would like to acknowledge Tibo Garroy for his work on the determination of the levels of tropicamide and phenylephrine in the serum of the patients in this study.

This study protocol was reviewed and approved by the Ethical Committee Universitair Ziekenhuis Antwerpen and Universiteit Antwerpen (UZA-UA) under the Approval No. EDGE001169. The study was also registered in the EUCT-Database under the number 2023-504173-21-00, and thus additionally reviewed and approved by an independent Belgian Ethical Committee. The study was performed following the principles of the Declaration of Helsinki. Written informed consent was obtained from the parents or legal guardians of every patient, as well as from the patients themselves when aged 6 years or older.

The authors have no conflict of interest to declare with regard to Mydrane. M.-J.B.R.T. has an intellectual property on the bag-in-the-lens and ring caliper used in this paper.

No funding or financial support was received for this study.

Planning of the work: L.V.O., K.W., H.H., W.A., A.L.N.N., A.C., V.S., and M.-J.B.R.T. Conduct of the work: L.V.O., H.H., W.A., C.D., I.D.B., C.S., A.L.N.N., A.C., V.S., and M.-J.B.R.T. Statistical analysis: K.W. Reporting of the work: L.V.O., K.W., H.H., W.A., C.D., A.L.N.N., A.C., V.S., and M.-J.B.R.T.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.