Clinical and Experimental Vision and Eye Research

Show Contents

Spectrum of presentation in primary congenital glaucomaand its relation to the early outcome
Spectrum of presentation in primary congenital glaucoma
and its relation to the early outcome
Sushmita Kaushik, Savleen Kaur, Indu Bala Dhiman, Adit Gupta, Srishti Raj, Surinder Singh Pandav
Department of Ophthalmology, Advanced Eye Centre, Postgraduate Institute of Medical Education and Research, Chandigarh, Punjab, India
Address for correspondence: Sushmita Kaushik,Advanced Eye Centre, PostgraduateInstitute of Medical Education and Research,Chandigarh - 160 012, Punjab, India.
Phone: 91-172-2756111. Fax: 91-172-2747837.
Received: 05-07-2018;
Accepted 12-07-2018
doi: 10.15713/ins.clever.3
Aim: This study aims to study the presentation pattern of primary congenital glaucoma(PCG) in Northern India and analyze the relationship to its outcome if any.
Methods: A total of 90 patients (157 eyes) with PCG presenting between January2003 and December 2011, with at least 1-year follow-up after surgery were included inthe study. Inclusion criteria were increased corneal diameter (>12.0 mm), intraocularpressure (IOP) >18 mmHg, and/or presence of Haab's striae with or without opticdisc changes. Demographic data, whether the patient was self-reported or doctorreferred, presenting IOP, corneal diameter, corneal clarity, and optic nerve damagewhere visualized, were recorded. Outcome of treatment was determined depending onIOP control and requirement of drugs. Differences in presenting feature between selfreportedand doctor-referred patients, and association between presenting features andoutcome was analyzed.
Results: Majority (67%) were self-reported. 102 eyes had hazy corneas; 45 eyes (28.7%)presented with corneal opacity. Doctor-referred patients presented earlier (P ≤ 0.001),had significantly lower IOP (P = 0.009), and smaller corneal diameters (P = 0.049) thanself-reported. Good outcome was correlated with lower IOP (P = 0.005) and smallercornea (P < 0.001) at presentation. Corneal opacity at presentation was stronglycorrelated to poor outcome (P < 0.001). Optic disc cupping at presentation had nobearing on the final outcome.
Conclusions: Corneal involvement appears to occur early in this cohort of PCG.Corneal examination under torchlight by paramedical personnel, general physicians,and pediatricians might lead to an early referral before vision-threatening complicationsdevelop. However, early corneal involvement may indicate more severe disease and islikely to have a poor prognosis regardless of the time of referral.
Keywords: Outcome, presentation, primary congenital glaucoma
How to cite this article: Kaushik S, Kaur S, Dhiman IB,Gupta A, Raj S, Pandav SS. Spectrum of presentation in primarycongenital glaucoma and its relation to the early outcome. CliExp Vis Eye Res J 2018;1(1):9-13.


Primary congenital glaucoma (PCG) is a relatively uncommoncondition but results in a lifetime of irreversible blindness if leftuntreated. There is a relatively narrow treatment window, which,if missed, is likely to lead to severe visual disability.

For any disease to be detected in time, it is important forthe signs and symptoms to be recognized. Unlike glaucoma inadults, which is notoriously difficult to detect since there are nosymptoms, glaucoma in children does present with symptomsand signs which can be detected by pediatricians and evenparents, who are usually the first contacts of these children. PCGclassically presents with a triad of photophobia, epiphora, andblepharospasm. Many children in India present with cornealedema initially, without buphthalmos or any of the classicalsigns.[1,2] Conversely, many children may have had symptomsfor a considerable period of time, before presentation to anophthalmologist simply because the disease was not thought of,hence, delaying diagnosis.

In India, there are few centers treating PCG and many childrenpresent too late for any meaningful treatment. Ours is the onlytertiary care referral center treating PCG for five large states inNorthern India, with a combined population of approximately84 million persons.[3] This gives us the opportunity to manage alarge number of children with congenital glaucoma. We noticedthat many of our children had severe corneal involvementvery early in the disease, and this presentation seemed to beassociated with poor outcome. Many infants presented withadvanced disease when the parents noticed "something wrong."This study aimed to characterize the presentation patterns ofPCG in this population, and if the outcome varied dependingon how the child presents, and whether pediatricians referringchildren early mattered in the final outcome.

Clinical and Experimental Vision and Eye Research, January-June, Vol 1, 2018 9

Kaushik, et al. PCG presentation related to outcome

Materials and Methods

This was a retrospective review of patients with PCG presentingto a tertiary care center in North India and registered in theelectronic database between January 2005 and December 2013,and completing 1-year follow-up after surgery. The study wasapproved by the Institute Ethics Committee (Vide ApprovalNo 8906/PG-2Trg/2011/6279) and adhered to the tenets ofthe Declaration of Helsinki. Informed consent had been takenby the parents or legal guardians of all children.

All patients had to have been diagnosed PCG by any oneof three glaucoma specialists (SSP, SK, and SR) based oncharacteristic clinical features including size of the globe, cornealfeatures, intraocular pressure (IOP), and disc evaluation wherepossible with examination under anesthesia when required. Allchildren identified as PCG were included for the analysis. Theinclusion criteria were as follows:
  1. Increased corneal diameter (>12.0 mm) along with raisedIOP (>18 mmHg) and/or presence of Haab's striae;
  2. Optic disc changes (where examination was possible);
  3. History of epiphora, photophobia, and sudden cornealclouding was considered corroborating factors.

The following information was recorded at presentation foreach case: Place of origin, birth history, family history, laterality,age of onset, and sex, whether the child was referred by a doctor orself-reported. The following ocular features at presentation wererecorded: Visual acuity (whenever possible), IOP by applanationtonometry (Goldmann/Perkins), corneal diameter, and cornealclarity including the presence of corneal edema, corneal opacityor scarring, and disc evaluation for glaucomatous damage.Most of these features were recorded from findings noted atthe first examination under anesthesia (inhalational anesthesiausing sevoflurane) under the operating microscope. IOP wasmeasured using the Perkins tonometer (Haag-Streit model,Clement Clark, Essex, England) as soon as the child was sedatedto negate the effects of general anesthesia as far as possible. Thehorizontal corneal diameters were measured using Castroviejo'scalipers. Corneal clarity was determined by observing iris detailsthrough the cornea under the microscope. Corneal status wasevaluated as follows: Presence or absence of corneal edema,presence or absence of Haab's striae, and presence or absenceof corneal scarring leading to corneal opacity. The severity ofcorneal edema was graded according to the visibility of the iristhrough the edematous cornea. Mild haze: Cornea hazy butiris details clearly visible; moderate Haze: Iris and pupil visiblethrough the hazy cornea but no iris details visible; severe edema:Iris not visible through edematous cornea.

The number and type of surgical procedures and/ormedications required to control IOP were recorded. IOP wasregarded as satisfactory if < 16 mmHg in patients examinedunder anesthesia, or < 21 mmHg in children old enough to beexamined with the slit lamp. The outcome was graded as goodif the IOP was brought under control without drugs, satisfactoryif up to two drugs was required to control IOP, and poor if theIOP was refractory to treatment needing three or more drugs forIOP control, or there was the development of hypotony or othersight-threatening complications.

The outcome was correlated to the factors listed above to seeif the mode of presentation had any bearing on the outcome oftreatment.

Statistical analysis

Statistical analysis was done using the IBM SPSS Statistics 19®program. Descriptive statistics were computed for all variables.Mann-Whitney U-test was used to analyze differences inpresenting features such as IOP, corneal diameter, corneal clarity,and optic nerve damage where visualized between self-reportedand doctor referral patients. Spearman's correlation was used toanalyze association between outcome and presenting features.Linear regression was used to analyze if one could predictoutcome from presenting signs.


Data of 90 patients who fulfilled the inclusion criteria wereanalyzed [Table 1]. There were 34 female and 56 male patientswho presented at a mean age of 4.36 ± 4.3 months and 4.5 ±6.7 months, respectively (P = 0.91). The majority were selfreferrals(67%) though boys tended to be referred by a doctormore often than girls (35% vs. 26%) though the difference wasnot significant. Self-reported patients presented significantlylater than those referred by a doctor (5.64 ± 6.82 months vs.1.9 ± 1.3 months, respectively; P < 0.001). Tearing and largeappearing eyes were the most common cause of referral by adoctor, while corneal opacity in one or the other eyes was themost common reason for the referral in self-reported cases. 67children (74%) presented with bilateral disease, which wassimilar in both boys and girls. Seven children had a history of asibling or a parent with primary glaucoma, with equal incidencein boys and girls (four had siblings with PCG and three had afather with POAG). Most patients (86%) came from the statesof Punjab (62%), Haryana (14%), and Himachal Pradesh (10%),reflecting the geographical proximity to our hospital.

Table 1: Patient demography
Spectrum of presentation in primary congenital glaucomaand its relation to the early outcome

10 Clinical and Experimental Vision and Eye Research, January-June, Vol 1, 2018

PCG presentation related to outcome Kaushik, et al.

157 eyes of these 90 children were analyzed for ocular findingsand outcome. Presenting features are given in Table 2. Mosteyes (104) presented with a hazy appearance of cornea with orwithout excessive tearing (66.2%). Six children had been advisedsac massage for congenital dacryocystitis elsewhere. 45 eyes(28.7%) presented with corneal opacity noticed by the parents,and in another 34 eyes (21%), corneal opacity was discoveredon EUA. Mean IOP at presentation was 18.9 ± 7.7 mmHg,and mean corneal diameter at presentation was 12.9 ± 1.9 mm.Corneal edema was detected in 43 (27.4%) eyes, Haab's striae in73 (46.5%) eyes, and 79 (50.3%) eyes had a corneal opacity onEUA. There was no view of the disc in 21 eyes. Of the rest, 60%had cup-disc ratio < 0.5, and 23% had advanced dice cupping(>0.8). The mean cup-disc ratio at presentation was 0.42 ± 0.28.

The pattern of referral of patients (self-reported and doctorreferredpatients) was separately analyzed. Those childrenwho were referred by a doctor presented earlier (P ≤ 0.001)and had significantly lower IOP (P = 0.009) and had smallercorneal diameters (P = 0.049) and clearer corneas than thosewho reported by themselves [Table 3]. 74% of doctor-referredpatients required only one surgical procedure compared to58% of self-reported patients (P = 0.03). Poor outcome wasobserved in 31% of self-reported patients compared to 24% ofdoctor referred, though the difference did not reach statisticalsignificance. Significantly, 64% of self-reported patients wereunder the care of some physician for non-ocular complaints orvaccinations but were not referred to an ophthalmologist.

Table 2: Presenting ocular findings
Spectrum of presentation in primary congenital glaucomaand its relation to the early outcome

Procedures required to control IOP are summarized inTable 4. 102 eyes (65%) could be controlled with one surgicalprocedure, 40 eyes (25.4%) required two procedures, 13 eyes(8.2%) underwent three surgeries, and two eyes required foursurgeries to control the IOP. The most commonly performedprimary procedure was trabeculotomy in approximately2/3rd of eyes (104 eyes), followed by combined trabeculotomywith trabeculectomy in 28 eyes (18%), trabeculectomy withmitomycin C in 24 eyes (15%), and one child presented witha painful blind eye and underwent diode laser transscleralcyclophotocoagulation. 23 eyes needed an Ahmed glaucomavalve, which was the second procedure in nine eyes, thirdprocedure in 12 eyes, and was implanted as the fourth procedurein two eyes. Two patients underwent optical penetratingkeratoplasty after glaucoma surgery. Three patients developeda cataract after trabeculectomy and required intraocular lensimplantation. Two patients had intractable glaucoma withpoor visual prognosis and underwent diode CPC as the secondprocedure after trabeculectomy. The number of proceduresrequired for IOP control correlated strongly with the IOP atpresentation (P = 0.003).

111 eyes (70.7%) had satisfactory control of IOP with orwithout one or two drugs after surgery (good outcome), of which58 eyes (37%) required no drugs. 46 eyes (29.3%) requiredmore than two drugs after surgery for satisfactory IOP control.Good outcome was correlated with lower IOP at presentation(P = 0.005), lower age at presentation (P = 0.034), smallercorneal diameter at presentation (P < 0.001), and visibilityof Haab's striae during EUA (P = 0.024). The presence ofcorneal opacity at presentation was strongly correlated to pooroutcome (P < 0.001) by linear regression analysis, lower IOP(P = 0.032), and absence of corneal opacity (P = 0.007) werepredictive of good outcome(Figure 1). Optic disc cupping atpresentation did not have any bearing on the final outcome. Themean follow-up was 22.67 ± 8.2 months (range 15-26 months).

Table 3: Differences in presentation in self-reported patients and those referred by a doctor
Spectrum of presentation in primary congenital glaucomaand its relation to the early outcome

Clinical and Experimental Vision and Eye Research, January-June, Vol 1, 2018 11

Kaushik, et al. PCG presentation related to outcome

Table 4: Surgical procedures required to control IOP
Spectrum of presentation in primary congenital glaucomaand its relation to the early outcome
IOP: Intraocular pressure

Spectrum of presentation in primary congenital glaucomaand its relation to the early outcome
Figure 1: (a) A 7-day-old baby with PCG at presentation withacute hydrops in the left eye and corneal haze in the right eye,(b) at 6-month follow-up following combined trabeculotomy andtrabeculectomy in both eyes. Note the clear cornea in the right eyeand central scar in the left eye with peripheral clearing, (c) newbornbaby (16 h) presenting with PCG with corneal haze in both eyes,(d) follow-up at 3 years following combined trabeculotomy andtrabeculectomy in both eyes. Note the clear cornea in both eyes.


The incidence of PCG varies substantially in different ethnicgroups from 1 in 1250 births in Slovakian Roms[4] to 1:20,000in Scandinavian regions.[5] In the West, the average incidence isabout 1 in 10,000 births,[6] but appears to be higher in Asians.In Saudi Arabia, it is reported to be 1:2500,[7] while Indian datafrom Andhra Pradesh reported an incidence of 1:3300.[8] Despiteits infrequent incidence, PCG accounts for 7-20% of childhoodblindness,[9-13] this disproportionate share of pediatric blindnessmay be because diagnosis is frequently delayed, resulting in anunnecessarily poor outcome despite appropriate treatment.Once the disease has advanced due to delayed diagnosis, littlecan be done in terms of visual rehabilitation of these children.[14]

PCG has been classically reported to present with tearing,photophobia, and blepharospasm,[12-15] but the scenario inthe developing world may be very different. One study fromNigeria[16] reported seven of eight children presenting withcorneal opacity. In another study from Africa, Bowman et al.[17]reported clear cornea at presentation in only 4 of the 47 eyesstudied. In a large series from South India, Mandal et al.[1]reported clear corneas at presentation in only 10% of childrenwith developmental glaucoma presenting before 6 months ofage. Corneal edema persisted in >37% of children despite earlysurgery. In another report by the same group,[2] of 47 eyes withPCG operated within 1 month age, 46 presented with cornealedema, and one eye presented with corneal scarring. Even withsuch early treatment, one-third (32%) had persistent cornealedema after surgery.

In the BIG study, Papadoupoulous et al.[18] failed to find anassociation between IOP control and initial IOP, sex, ethnicity,time to surgery from diagnosis, corneal diameter, or the age ofdiagnosis. However, as they discuss it in their paper, the lowernumber of children (45) may have accounted for their result.

Among the 157 eyes of 90 children we studied, the averageage of presentation was 4.5 months. Only 33.75% had clearcorneas at presentation, and 45 eyes (29%) presented withcorneal opacity. The presenting IOP was 18.9 mmHg and about2/3rd of children had cup-disc ratios < 0.5. This emphasizes thatthe IOP and disc findings considered "normal" in adults may notbe valid for infants. In addition, since all these IOP measurementswere taken under general anesthesia, they are likely to have beenunderestimated. Lower IOP, smaller corneal diameter, andcorneal clarity at presentation were predictors for good outcomebut not earlier age at presentation by itself. This highlights theimportance of careful corneal examination in all infants beingevaluated for PCG and also indicates that though some babieswere referred early, if their cornea was severely affected, theirprognosis was poor.

Nevertheless, we did observe that babies referred by doctorspresented earlier, had lower IOP, better corneal features atpresentation, and required lesser number of surgical proceduresfor IOP control compared to those who were self-reported.Waiting till parents note "something wrong" in the infant's eyesmay tilt the balance toward poorer outcome than what couldhave been achieved by an early referral. An examination of thecornea under a hand light can be stressed on to pediatricians,general practitioners, and also paramedical personnel whomay be administering vaccinations, etc. This might lead to anearly referral before vision-threatening complications develop.Recognizing signs of PCG such as large eyes, tearing, and hazycorneas will allow referral of at-risk infants at a time when goodvision may be maintained, at least in those who have a reasonableprognosis of treatment.

12 Clinical and Experimental Vision and Eye Research, January-June, Vol 1, 2018

PCG presentation related to outcome Kaushik, et al.

It may be that delayed presentation may cause the cornea toworsen and thus compromise the outcome. However, we alsonoted that early corneal involvement may indicate more severedisease, and these infants are likely to have a poor prognosis.Recognizing the disease early would be the first steps towardearly treatment and better outcomes. Long-term evaluation ofthe visual outcome of these infants correlated to the presentationwould undoubtedly throw more light on how presentationpatterns could be a guide to prognosticating the disease.

  1. Mandal AK, Bhatia PG, Bhaskar A, Nutheti R. Longtermsurgical and visual outcomes in Indian children withdevelopmental glaucoma operated on within 6 months of birth.Ophthalmology 2004;2:283-90.
  2. Mandal AK, Gothwal VK, Bagga H, Nutheti R, Mansoori T.Outcome of surgery on infants younger than 1 month withcongenital glaucoma. Ophthalmology 2003;110:1909-15.
  3. Provisional Population Totals Paper 1 of 2011 India. RegistrarGeneral and Census Commissioner, India. New Delhi: Ministryof Home Affairs, Government of India; 2011.
  4. Gencik A. Epidemiology and genetics of primary congenitalglaucoma in Slovakia. Description of a form of primarycongenital glaucoma in gypsies with autosomal-recessiveinheritance and complete penetrance. Dev Ophthalmol1989;16:76-115.
  5. McGinnity FG, Page AB, Bryars JH. Primary congenitalglaucoma: Twentyyears experience. Ir J Med Sci 1987;156:364-5.
  6. Ho CL, Walton DS. Primary congenital glaucoma: 2004 update.J Pediatr Ophthalmol Strabismus 2004;41:271-88.
  7. Bejjani BA, Stockton DW, Lewis RA, Tomey KF, Dueker DK,Jabak M, et al. Multiple CYP1B1 mutations and incompletepenetrance in an inbred population segregating primarycongenital glaucoma suggest frequent de novo events and adominant modifier locus. Hum Mol Genet 2000;9:367-74.

  1. Dandona L, Williams JD, Williams BC, Rao GN. Populationbasedassessment of childhood blindness in southern India.Arch Ophthalmol 1998;116:545-6.
  2. Gao DW, Kubota T, Sugino K, Sugai S, Inomata H. A statisticalcomparison study of glaucoma in the third affiliated hospital ofChina medical college and Kyushu university. Nippon GankaGakkai Zasshi 1989;93:458-65.
  3. Kipp MA. Childhood glaucoma. Pediatr Clin North Am2003;50:89-104.
  4. Franks W, Taylor D. Congenital glaucoma--a preventable causeof blindness. Arch Dis Child 1989;64:649-50.
  5. Dorairaj SK, Bandrakalli P, Shetty C, R V, Misquith D, Ritch R.Childhood blindness in a rural population of southern India:prevalence and etiology. Ophthalmic Epidemiol 2008;3:176-82.
  6. Onakpoya OH, Adegbehingbe BO, Omotoye OJ, Adeoye AO.Causes of blindness in a special education school. West Afr JMed 2011;1:47-50.
  7. Tanimoto SA, Brandt JD. Options in pediatric glaucoma afterangle surgery has failed. Curr Opin Ophthalmol 2006;2:132-7.
  8. Barsoum-Homsy M, Chevrette L. Incidence and prognosisof childhood glaucoma. A study of 63 cases. Ophthalmology1986;10:1323-7.
  9. Omoti AE, Uhumwangho OM. Problems of management ofprimary congenital glaucoma in Benin City, Nigeria. NigerPostgrad Med J 2007;4:310-3.
  10. Bowman RJ, Dickerson M, Mwende J, Khaw PT. Outcomes ofgoniotomy for primary congenital glaucoma in East Africa.Ophthalmology 2011;2:236-40.
  11. Papadopoulos M, Cable N, Rahi J, Khaw PT, BIG Eye StudyInvestigators. The British infantile and childhood glaucoma(BIG) eye study. Invest Ophthalmol Vis Sci 2007;48:4100-6.

Clinical and Experimental Vision and Eye Research, January-June, Vol 1, 2018 13