course and outcome of accidental sodium hydroxide ocular injury

dc[avmtum

attcep©

7

Course and Outcome of Accidental Sodium HydroxideOcular Injury

NAMRATA SHARMA, DIGVIJAY SINGH, AMIT SOBTI, PRAKASHCHAND AGARWAL,
THIRUMURTHY VELPANDIAN, JEEWAN S. TITIYAL, AND SUPRIYO GHOSE
ab

● PURPOSE: To evaluate the course and outcome ofpatients with accidental ocular alkali burns.● DESIGN: Prospective, interventional case series.● METHODS: Study of a cohort of 16 patients (31 eyes)who sustained concomitant accidental sodium hydrox-ide ocular burns and received appropriate treatment ata tertiary care eye hospital in India. The patients werefollowed up for 1 year, and parameters includingbest-corrected visual acuity, epithelial defect area,conjunctival and limbal involvement, and injury-re-lated complications were evaluated.● RESULTS: Severe sodium hydroxide exposure of a meanuration of 12 � 2.5 minutes and delay in specialist eyeare caused moderate to severe injury (grade II, 19%n � 6]; grade III, 19% [n � 6]; grade IV, 10% [n � 3];nd grade VI, 52% [n � 16]). Median best-correctedisual acuity at presentation was 1.0 logarithm of theinimal angle of resolution (logMAR) units (range, 0.3

o 1.9 logMAR units), and at 1 year, it was 1.0 logMARnits (range, 0 to 1.9 logMAR units; P � .121). Theedian initial epithelial defect was 100 mm2 (range, 18

to 121 mm2), which healed in all eyes by 3.5 months.Initial median limbal involvement was 12 clock hours(range, 3 to 12 clock hours), resulting in a residual limbalstem cell deficiency of 6 clock hours (range, 0 to 12 clockhours) at 1 year. Most common complications wereglaucoma and cataract. Corneal ulcers developed in 2eyes, and keratolimbal graft was performed in 1 patient.Grade VI injuries had significantly worse outcome thanthe lower-grade injuries.● CONCLUSIONS: The course and outcome of ocularlkali burns depends on effective first aid (including ahorough eyewash), age, initial grade of injury, responseo treatment, prevention of secondary infection, andontrol of glaucoma. Despite appropriate treatment, theseyes responded poorly and carried a guarded visualrognosis. (Am J Ophthalmol 2012;154:740–749.

2012 by Elsevier Inc. All rights reserved.)

Accepted for publication April 24, 2012.From the Dr Rajendra Prasad Centre for Ophthalmic Sciences, All

India Institute of Medical Sciences, Ansari Nagar, New Delhi, India(N.S., D.S., A.S., P.A., T.V., J.S.T., S.G.).

Inquiries to Namrata Sharma, Cornea and Refractive Surgical Services,Dr. Rajendra Prasad Centre for Ophthalmic Sciences, All India Instituteof Medical Sciences, Ansari Nagar, New Delhi, India 110029; e-mail:
[email protected]
© 2012 BY ELSEVIER INC. A40

C HEMICAL INJURIES ARE OPHTHALMIC EMERGENCIES

that have the potential to cause significant mor-bidity and loss of vision.1,2 Injuries with alkaline

gents are slightly more common and produce more severeurns than acids.1–4 The exposure is predominantly indus-

trial or occupational and is more common among youngmales.

Herein we describe a unique cohort of 16 patients whosustained simultaneous ocular and skin burns with a highlyalkaline industrial grade sodium hydroxide (NaOH) solu-tion after a traffic accident in rural India. We studiedfactors determining the severity of injury, response totreatment, and final outcome over a follow-up period of 12months.

METHODS

SIXTEEN PATIENTS WERE ENROLLED FROM THE EYE CASU-

alty service of our hospital, where they sought treatmentfor ocular and skin burns of varying severity. The patientshad previously received care at a local hospital near theaccident site, where they received an eyewash with normalsaline and were prescribed preservative-free lubricant dropsand topical antibiotics (chloramphenicol 0.5% 4 timesdaily). At our center, the tear film showed an alkaline pH.Repeat eyewash of the ocular surface extending into theconjunctival fornices (after double eversion of the lids) wasperformed with normal saline until a neutral pH of 7.0 wasachieved. The eyewash samples and conjunctival swabswere collected for chemical analysis.

The accident victims were treated as in-patients using astandard treatment protocol. The protocol included topi-cal moxifloxacin hydrochloride 0.5% 4 times daily (Viga-mox; Alcon, Inc, Fort Worth, Texas, USA), topicalprednisolone acetate 1% 4 times daily (Predforte; Aller-gan, Irvine, California, USA), preservative-free lubricantdrops on an hourly basis (hydroxypropyl methylcellulose2%; Ocular Pharmacy, Dr Rajendra Prasad Centre forOphthalmic Sciences, Delhi, India), topical sodium citrate10% every 4 hours (Ocular Pharmacy, Dr Rajendra PrasadCenter for Ophthalmic Sciences), and topical sodiumascorbate 10% every 4 hours (Ocular Pharmacy, Dr Ra-jendra Prasad Centre for Ophthalmic Sciences). Oral
vitamin C, 1 g daily, was given for 1 month to all patients
LL RIGHTS RESERVED. 0002-9394/$36.00http://dx.doi.org/10.1016/j.ajo.2012.04.018

mailto:[email protected]

http://dx.doi.org/10.1016/j.ajo.2012.04.018

(T. Limcee; Sarabhai Piramal Pharma, Vadodara, India),and 250 mg oral acetazolamide 4 times daily (T. Diamox;Wyeth Limited, Mumbai, India) was used to lower theintraocular pressure, as needed (topical antiglaucoma med-ications were avoided because of the risk of epithelialtoxicity). Oral ciprofloxacin 500 mg twice daily was givenfor 7 to 10 days (T. Ciprobid; Cadila Healthcare, Gujarat,India) in 8 patients with extensive burns. Skin burns weretreated in consultation with the surgical burn specialistusing silver sulfadiazine 1% w/w, chlorhexidine gluconate0.2% w/w, and regular paraffin gauze dressings. Oral ulcersand oral, aural, and nasal burns were managed in consul-

FIGURE 1. Representative clinical photographs of eyes with vainjury, showing good recovery and favorable prognosis. (Top roepithelial defect and ciliary congestion. (Top row middle) Sixnearly clear cornea with no significant sequelae. (Second rowprognosis. (Second row left) Eye with grade III chemical injury(Second row middle) Six-month follow-up. (Second row right)involvement. (Third row) Grade IV chemical injury showing pgrade IV chemical injury at presentation. Note the epithelial,Six-month follow-up. (Third row right) One-year follow-up. Norow) Grade VI chemical injury showing dismal recovery and uninjury at presentation. Note the involvement of the entire ocfollow-up. Note the persistence of inflammation. (Bottom row riwith 360-degree pannus formation.

tation with the dental and otorhinolaryngology experts.

COURSE AND OUTCOME OF ACVOL. 154, NO. 4

The treatment was adjusted according to individualresponse. The prominent modifications included taperingtopical steroids and then discontinuing in all cases within2 weeks, adding fortified topical antibiotics in cases suspi-cious for infectious keratitis, increasing the frequency oftopical lubricants or addition of gel preparations in eyeswith eyelid abnormalities, adding oral doxycycline 100 mgtwice daily in 4 patients with progressive corneal melting,removing trichiatic lashes, as needed, and making anynecessary changes after surgical interventions.

The injury was classified according to the ocular chem-ical burns classification created by Dua and associates.5

grades of ocular chemical injury. (Top row) Grade II chemicalft) Eye with grade II chemical injury at presentation. Note theth follow-up. (Top row right) One-year follow-up. Note theade III chemical injury showing good recovery and favorableresentation. Note the epithelial defect and limbal involvement-year follow-up. Note the corneal opacity at the site of limbalecovery and unfavorable prognosis. (Third row left) Eye withal, limbal, and conjunctival involvement. (Third row middle)

e leukomatous corneal opacity with pannus formation. (Bottomrable prognosis. (Bottom row left) Eye with grade VI chemicalsurface and corneal stroma. (Bottom row middle) Six-monthOne-year follow-up. Note the large vascularized corneal opacity

ryingw le-mon) Grat pOneoor rstromte thfavoularght)

The classification scheme involves grading ocular surface

UTE ALKALI OCULAR INJURY 741

burns based on limbal and conjunctival involvement andincludes an analog scale. Accordingly, the ocular parame-

TABLE 1. Course and Outcome of Best-Corrected Vis

Serial

Number

Age

(years)

Grade (Dua

Classification)

At

Presentation

2-Week

Follow-up

1 25 VI 1.08 0.6

2 25 VI 1.18 1.9

3 20 VI 1.9 1.78

4 20 VI 1.9 1.78

5 32 III 0.48 0

6 32 IV 0.6 0.48

7 45 III 1.48 1.9

8 45 IV 1.3 1.9

9 15 III 0.6 0.3

10 15 VI 1.9 1.48

11 20 II 1 0.16

12 20 VI 1.9 1.48

13 25 II 0.6 0.16

14 25 II 0.6 0.16

15 28 VI 1 0.3

16 28 VI 1 0.48

17 55 VI 1.9 1.78

18 55 VI 1.78 1.48

19 18 VI 1.9 1.3

20 22 II 0.48 0.3

21 22 II 0.48 0.16

22 22 III 0.3 0.16

23 22 III 0.3 0.16

24 35 VI 1.48 1.3

25 35 VI 1.48 1.48

26 35 IV 0.6 0.6

27 35 VI 1.8 1.78

28 25 VI 1.3 1.3

29 25 VI 1 1.3

30 25 II 1 0.3

31 25 III 0.78 0.16

logMAR � logarithm of the minimal angle of resolution.

TABLE 2. Subgroup Analysis of Best-Corrected VisualAcuity by Grade of Sodium Hydroxide Ocular Injury at

Presentation and after Complete Healing

Grade (Dua

Classification)

Median (Range) Best-Corrected Visual

Acuity (logMAR units)

Significance of

Change between

Presentation and

Final Visit

(P Value)At Presentation At 1 Year

II 0.6 (0.48 to 1) 0.08 (0 to 0.16) .03

III 0.54 (0.3 to 1.48) 0.0 (0 to 1.9) .11

IV 0.6 (0.6 to 1.3) 1.9 (0.48 to 1.9) .29

VI 1.63 (1 to 1.9) 1.9 (0.3 to 1.9) .50

logMAR � logarithm of the minimal angle of resolution.

ters evaluated at the primary visit included best-corrected

AMERICAN JOURNAL OF742

visual acuity (BCVA), ocular congestion, conjunctivalchemosis, size of epithelial defect, degree of limbalinvolvement (in clock hours), degree of conjunctivalinvolvement (as a percentage), anterior chamber reac-tion, intraocular pressure, and evidence of any second-ary bacterial infection.

The patients were closely followed up for a period of1 year. At months 3, 6, 9, and 12, the above parameterswere recorded, along with documentation of any com-plications, such as limbal stem cell deficiency, sym-blepharon formation, pseudopterygia, lid abnormalities,glaucoma, or cataract. Therapeutic keratoplasty, amni-otic membrane transplantation, keratolimbal graft, andcultured limbal stem cell transplantation were under-taken, if required.

The data were entered and analyzed using SPSS softwareversion 11.5 (SPSS, Inc, Chicago, Illinois, USA) and

cuity in Patients with Sodium Hydroxide Ocular Injury

Best-Corrected Visual Acuity (logMAR Score)

1-Month

Follow-up

3-Month

Follow-up

6-Month

Follow-up

9-Month

Follow-up

1-Year

Follow-up

0.6 1.3 1.9 1.9 1.9

1.9 1.9 1.9 1.9 1.9

1.78 1.9 1.9 1.9 1.9

1.78 1.9 1.9 1.9 1.9

0 0 0 0 0

0.48 0.48 0.48 0.48 0.48

1.9 1.9 1.9 1.9 1.9

1.9 1.9 1.9 1.9 1.9

0.16 0.16 0 0 0

1.48 1.78 1.08 1 0.6

0.16 0.3 0 0 0

1.48 1.3 1.18 1.18 1

0.16 0 0 0 0

0.16 0 0 0 0

0.3 1.9 1.18 0.6 0.48

0.48 1.9 1.9 1.3 1.9

1.78 1.9 1.9 1.9 1.9

1.48 1.9 1.9 1.9 1.9

1.18 0.48 0.6 0.6 0.3

0.3 0.16 0.16 0.16 0.16

0.16 0.16 0.16 0.16 0.16

0.16 0 0 0 0

0.16 0.16 0 0 0

1.3 1.9 1.9 1.9 1.9

1.48 1.9 1.9 1.9 1.9

1.9 1.18 1.9 1.9 1.9

1.9 1.48 1.9 1.9 1.9

1.9 1.9 1.9 1.9 1.9

1.9 1.9 1.9 1.9 1.9

0.3 0.3 0.3 0.3 0.16

0.16 0 0 0 0

ual A

Stata software version 8.0 (StataCorp LP, College Station,

OPHTHALMOLOGY OCTOBER 2012

0

Texas, USA) following the appropriate protocol. TheFriedman and Wilcoxon rank-sum (Mann–Whitney) testswere applied to assess the BCVA, epithelial defect area,and limbal involvement. Patients with limbal involvementwere categorized into 2 groups: 6 clock hours or fewer andmore than 6 clock hours. The groups were analyzed usingthe Wilcoxon signed-rank test.

RESULTS

THE PATIENTS WERE ALL MALES WITH A MEAN AGE OF 28 �

10.28 years (range, 15 to 55 years). The median grade ofinjury was grade III (range, grades II through VI) inpatients younger than 25 years (n � 10) and grade VI(range, grades III through VI) in patients older than 25years (n � 6; P � .14). The mean duration of exposure tothe offending agent was 12 � 2.54 minutes, and the timefrom exposure to initial treatment at the first hospital was

TABLE 3. Course and Resolution of Epithelial De

Serial Number Age (years)

Grade

(Dua Classification) At Presentation 2-

1 25 VI 100

2 25 VI 100

3 20 VI 100

4 20 VI 100

5 32 III 56

6 32 IV 120

7 45 III 121

8 45 IV 121

9 15 III 40

10 15 VI 121

11 20 II 64

12 20 VI 121

13 25 II 17.5

14 25 II 39.2

15 28 VI 121

16 28 VI 121

17 55 VI 121

18 55 VI 121

19 18 VI 64

20 22 II 48

21 22 II 36

22 22 III 68

23 22 III 22.5

24 35 VI 121

25 35 VI 121

26 35 IV 121

27 35 VI 121

28 25 VI 121

29 25 VI 121

30 25 II 30.25

31 25 III 48.75

2 hours. The mean time to presentation at our center was


1.17 days. The patients had varying degrees of skin andmucosal burns, but systemically were stable. One patient,who was older than 45 years of age, had diabetes mellitusand another patient had hypertension. Both of thesepatients had good disease control on oral medications.

The patients had different grades according to the Duaclassification. Of the patients, 6 (19%) had grade II injury,6 (19%) had grade III injury, 3 (10%) had grade IV injury,and 16 (52%) had grade VI injury (Figure 1). The medianBCVA on day 1 was 1.0 logarithm of the minimal angle ofresolution (logMAR) units (range, 0.3 to 1.9 logMARunits). Similarly, the BCVA at the end of 1 year was 1.0logMAR units (range, 0 to 1.9 logMAR units; P � .121;Table 1). On subgroup analysis, the BCVA for grade VIinjuries was significantly worse than that for lower-gradeinjuries, whereas comparison of the median BCVA ofgrade II and III injuries, grade II and IV injuries, and gradeIII and IV injuries revealed no significant differences(Table 2).

The corneal epithelium showed signs of healing with

in Patients with Sodium Hydroxide Ocular Injury

Epithelial Defect Area (mm2)

ollow-up 1-Month Follow-up 3-Month Follow-up 6-Month Follow-up

.5 5.2 0 0

10.8 0 0

.76 16.4 0 0

.51 0.8 0 0

0 0 0

.3 0 4.5 0

10.5 24 0

.5 9 9 0

2 0 0

44.1 0 0

0 0 0

23.94 0 0

0 0 0

0 0 0

25 6 0

52.5 8 0

20.79 0 0

59.25 28 0

12 0 0

0 0 0

0 0 0

0 0 0

0 0 0

81.81 4 0

.8 81.34 0 0

.5 10.5 0 0

28 0 0

100 0 0

.06 60.8 0 0

.2 0 0 0

0 0 0

fects

Week F

31

44

23

3

0

1

14

17

7

48

0

56

0

0

30

56

80

64

16

0

0

0

0

84

91

22

38

98

95

4

medical management (Table 3). Likewise, the limbal


m2s

involvement (Table 4) and conjunctival involvement(Table 5) also showed improvement. The median epithe-lial defect area at day 1 was 100 mm2 (range, 18 to 121

m2), which decreased to a median of 0 mm2 (range, 0 to8 mm2) by 3 months. Seven patients with grade VI injurytill had epithelial defects ranging from 4 to 28 mm2 at the

3-month follow-up visit. However, by week 14, all epithe-lial defects had healed. The epithelial defect area variedsubstantially according to the severity of injury (Tables6 and 7). Patients with grade IV and VI injuries hadsignificantly worse epithelial defects than patients withother grades of injury (Table 7). There was no significantdifference between the grade II and III injuries (P � .14,Mann–Whitney U test).

The median limbal involvement on day 1 was 12 clockhours (range, 3 to 12 clock hours). There was a progressivedecrease in limbal involvement with time, and the residualmedian limbal stem cell deficiency was 6 clock hours

TABLE 4. Course and Outcome of Limbal Involve

Serial

Number Age (years)

Grade

(Dua Classification)

At

Presentation

2-Week

Follow-u

1 25 VI 12 6

2 25 VI 12 7

3 20 VI 12 6

4 20 VI 12 3

5 32 III 4 2

6 32 IV 9 6

7 45 III 4 3

8 45 IV 8 3

9 15 III 6 2

10 15 VI 12 12

11 20 II 3 0

12 20 VI 12 9

13 25 II 3 0

14 25 II 3 0

15 28 VI 12 6

16 28 VI 12 7

17 55 VI 12 9

18 55 VI 12 12

19 18 VI 12 6

20 22 II 3 1

21 22 II 3 0

22 22 III 6 1

23 22 III 6 0

24 35 VI 12 9

25 35 VI 12 7

26 35 IV 9 9

27 35 VI 12 7

28 25 VI 12 9

29 25 VI 12 9

30 25 II 4 1

31 25 III 6 3

(range, 0 to 12 clock hours) at the end of 1 year (Table 4).


Limbal involvement has been analyzed according to thegrade of injury and is shown in Table 6. Limbal involve-ment improved significantly at 1 year in grade II, III, andIV injuries. An intergroup analysis demonstrated a signif-icant difference in limbal involvement based on injuryseverity (Table 7). For further analysis, the eyes weredivided into 2 groups based on limbal involvement. Thefirst group included eyes with limbal involvement of fewerthan 6 clock hours (cases with good prognosis, accordingto the Dua classification), and the second group includedeyes with more than 6 clock hours limbal involvement(cases with guarded to poor prognosis, according to theDua classification). In the first group, an initial medianlimbal involvement of 4 clock hours (range, 3 to 6 clockhours) was noted. In this group, the final median limbalstem cell deficiency was 0 clock hours (range, 0 to 5 clockhours; P � .003). However, in the second group, the initialmedian limbal involvement was 12 clock hours (range, 8

in Patients with Sodium Hydroxide Ocular Injury

al Involvement (in Clock Hours)

Limbal Stem Cell

Deficiency

1-Month

Follow-up

3-Month

Follow-up

6-Month

Follow-up

9-Month

Follow-up

1-Year

Follow-up

6 9 9 9 9

7 12 12 12 12

6 7 12 12 12

3 8 9 9 9

0 0 0 0 0

0 0 0 0 0

3 5 5 5 5

3 5 6 6 6

1 0 0 0 1

9 9 9 9 9

0 0 0 0 0

9 9 9 9 8

0 0 0 0 0

0 0 0 0 0

6 6 6 6 7

8 7 8 8 9

9 12 12 12 12

12 12 12 12 12

6 5 5 6 7

0 0 0 0 0

0 0 0 0 0

1 1 1 1 1

0 0 0 1 1

9 9 4 12 12

7 7 6 6 4

9 9 9 9 7

7 7 7 7 12

12 12 12 12 12

12 12 12 12 12

1 1 1 3 3

2 0 0 0 0

ment

Limb

p

to 12 clock hours), and the final median limbal stem cell


Tpcl

deficiency was 9 clock hours (range, 0 to 12 clock hours;P � .0012).

The conjunctival involvement for each eye is shown inTable 5, and grade-specific involvement is depicted inTable 6. There is a significant difference in conjunctivalinvolvement for each grade of injury (Table 7). Superficialvascularization was present in 70.97% of eyes, and deepvascularization was present in 58.06% of eyes at the end of1 year. The median vascularization at the end of 1 year was6 clock hours (range, 0 to 12 clock hours). There was nosignificant correlation between the overall best-correctedvisual acuity, limbal involvement, or conjunctivalinvolvement.

The Kruskal-Wallis equality of populations rank testshowed a significant difference between grade VI and theother grades with regard to the BCVA, limbal involve-ment, conjunctival involvement, and epithelial defect area

TABLE 5. Conjunctival Involvement at Presentation andOutcome in Patients with Sodium Hydroxide Ocular Injury

Serial

Number Age (years)

Grade (Dua

Classification)

Conjunctival Involvement (%)

At

Presentation

Final Outcome

(1 Year)

1 25 VI 90 Symblepharon


3 20 VI 95 Pseudopterygium

4 20 VI 95 Pyogenic granuloma

5 32 III 50 Scarring

6 32 IV 60 Scarring

7 45 III 40 Normal

8 45 IV 70 Scarring

9 15 III 40 Normal

10 15 VI 90 Scarring

11 20 II 20 Normal


13 25 II 20 Normal

14 25 II 20 Normal






20 22 II 20 Pseudopterygium

21 22 II 20 Normal

22 22 III 40 Normal

23 22 III 40 Normal



26 35 IV 70 Scarring


28 25 VI 95 Pyogenic granuloma


30 25 II 20 Normal

31 25 III 50 Normal

(P � .001).


The initial surgical procedures performed included de-bridement of necrotic conjunctival and corneal tissue in 26eyes, removal of trichiatic lashes in 19 eyes, and atarsorrhaphy in 6 eyes to promote epithelial healing.

Corneal ulcers developed in 2 eyes with coagulase-negative Staphylococcus. The ulcers did not respond totopical fortified antibiotics, and both cases underwenttherapeutic penetrating keratoplasty. However, in one eye,phthisis bulbi developed, and an anterior staphylomadeveloped in the second eye.

A keratolimbal graft was performed in 1 patient with acorneal opacity and partial limbal stem cell deficiency atthe 9-month follow-up. After surgery, the patient receivedsystemic cyclosporine 5 mg/kg body weight for 3 months,along with topical corticosteroids, antibiotics, and lubri-cants. The graft developed a persistent epithelial defectwith superinfection during the postoperative period, withresultant haziness of the graft. In another patient, anallograft was performed using cultured limbal stem celltransplantation derived from the patient’s mother. Thecells were cultivated on a denuded human amnioticmembrane, and the patient achieved a vision of 0.48logMAR units.

Lid involvement was observed in 8 eyes (25. 81%) atthe time of presentation. Symblepharon formation wasnoted in 3 eyes (9.68%) at the 1-year follow-up. Themost common complication was secondary glaucoma in7 (22.58%) cases, followed by cataract formation, whichoccurred in 6 (19.35%) cases. Pseudopterygium devel-oped in 5 (16.13%) eyes, and scleral melting occurred in4 (12.9%) eyes. Other sequelae included adherentleucoma (n � 2, 6.4%), trichiasis (n � 2, 6.4%),pyogenic granuloma (n � 2, 6.4%), anterior staphyloma(n � 2, 6.4%), and phthisis bulbi (n � 2, 6.4%; Figure2). The distribution of cataract and glaucoma did notdiffer significantly between the various grades of injury(P � .69 and P � .09, respectively, Fisher exact test).

he development of sequelae, such as symblepharon,seudopterygium, and pyogenic granuloma, was signifi-antly more likely in eyes with grade VI burns than inower grade burns (P � .003, chi-square test).

A team from the ocular pharmacology department of ourcenter analyzed the chemical composition of the alkaliobtained from samples collected at the site of accident andfrom the eyewash fluid. The site of spillage was a driedagricultural water channel. The chemical procured wasanalyzed using the appropriate inorganic identificationtests and was found to be a concentrated solution of NaOHwith a percentage purity of 40% w/v (10 M). The liquidwas brown with a density of 1.3 to 1.4 and a pH of morethan 13.3. Test results for sulfates and chlorides werepositive, and the substance formed a black precipitate withsilver nitrate. The chemical was found to be corrosive toall tissues and caused protein solubilization and lipidsaponification. A concentrated alkaline solution such as
this is normally used in paper mills.

sit

iwioa

bpc

0.5 (4

ular in

DISCUSSION

INDUSTRIAL CHEMICAL INJURIES USUALLY ARE CAUSED BY

caustic chemicals and solvents. Although the incidence ofthese injuries is low in developed countries, it is commonin developing countries with less rigorous industrial regu-lations.6–8 Whenever possible, the offending chemicalhould be identified, because the severity of a chemicalnjury depends on the pH, volume, and inherent toxicity of

TABLE 6. Subgroup Analysis of Epithelial Defect Area and L

Grade (Dua

Classification)

Median (Range) Epithelial

Defect Area at

Presentation (mm2)

Median (Range) Limbal

(Clock Hours

At Presentation

II 37.6 (17.5 to 64) 3 (3 to 4)

III 52.4 (22.5 to 121) 6 (4 to 6)

IV 121 (120 to 121) 9 (8 to 9)

VI 121 (64 to 121) 12 (12 to 12) 1

TABLE 7. Intergroup Analysis of Various Parameters and the(Mann–W

Subgroup Comparisons

(Dua Classification)

Best-Corrected Visual

Acuity at Presentation

Final Best-Corrected

Visual Acuity

Grade II vs grade III .51 .38

Grade II vs grade IV .41 .02

Grade II vs grade VI .001 .001

Grade III vs grade IV .43 .04

Grade III vs grade VI .002 .001

Grade IV vs grade VI .04 .72

Each group represents a grade of sodium hydroxide-induced oc

FIGURE 2. Clinical photographs showing the sequelae of ocumiddle) Cicatricial entropion with trichiasis. (Top row right) Prow middle) Leucoma. (Bottom row right) Pyogenic granulom

he chemical as well as the duration of contact. It is i


mperative to analyze the nature of the offending substancehenever a large number of patients are affected concom-

tantly, such as in accidental injuries. In our series, theffending agent was identified as a concentrated solution oflkaline NaOH, commonly known as lye.

The standard of care for management of ocular chemicalurns is an urgent and thorough rinse of the eyes anderiocular skin and removal of all traces of the offendinghemical.9–11 Copious irrigation of the ocular surface,

l Involvement by Grade of Sodium Hydroxide Ocular Injury

mentP Value for Change in Limbal

Involvement (Wilcoxon

Signed-Rank Test)

Mean � Standard Deviation

Conjunctival Involvement at

Presentation (%)Year

to 3) .02 20 � 0

to 5) .04 43.35 � 4.44

to 7) .10 66.67 � 4.44

to 12) .01 93.43 � 2.14

esponding Significance Values from the Wilcoxon Rank-Sumey) Test

Significance Value

ithelial Defect

Area

Limbal Involvement

at Presentation

Final Limbal Stem

Cell Deficiency

Conjunctival

Involvement

.14 .004 .14 .001

.02 .01 .09 .006

.001 �.001 �.001 .001

.07 .01 .29 .01

.003 �.001 �.001 �.001

.78 �.001 .01 .003

jury.

hemical burns. (Top row left) Corneal perforation. (Top rowopterygium. (Bottom row left) Anterior staphyloma. (Bottom

imba

Involve

)

At 1

0 (0

1 (0

6 (0

Corrhitn

Ep

lar cseud

a.

ncluding the fornices, with double eversion of the lid is


wadbmw

ec

performed using a sterile saline or lactated ringer’s solutionuntil the pH of the ocular surface is neutralized to 7.0.12,13

The use of prophylactic antibiotics and cycloplegics also isrecommended.10,11,13 Ocular surface healing is promoted

ith preservative-free artificial tear supplements, sodiumscorbate, sodium citrate, the controlled use of steroidsuring the initial period, and placement of a therapeuticandage contact lens.14–17 Newer ways to promote healingay include the use of autologous and cord blood serum,hich provide additional growth factors.18–20 Initially,

surgical care involves removing the embedded chemicalparticles and promoting ocular surface healing by de-briding the necrotic tissue. Amniotic membrane trans-plantation may help in promoting epithelialization.21–28

Later, limbal stem cell transplantation, penetratingkeratoplasty, and keratoprosthesis may be undertakenfor visual rehabilitation.29 –32

There are many controversies plaguing the treatment ofocular chemical injury. Opinions differ on the appropriateirrigation fluid for emergency management.33 Alkalineagents penetrate the cornea and increase the aqueous pH,subsequently damaging the intraocular structures. Al-though experimental evidence suggests that isotonic solu-tions are ineffective in neutralizing the intracameral pH,various special irrigating solutions have been tested andrecommended for rinsing the eye.34,35 Although thesespecial solutions are ideal, they may not be availableroutinely; the concept of providing an urgent wash pre-dominates, and thus, even clean water may be used.36,37

Topical steroids are known to enhance collagenolysisand corneal melting, leading to skepticism on their usein chemical injury.38 The need to reduce inflammationarly in the course of the chemical burn favors a shortourse of steroids in combination with vitamin C.38

Various studies have attempted and recommended theuse of autologous or cord serum for early epithelizationafter chemical injury.19,20 The role of an amnioticmembrane is controversial, with some studies supportingit and others reporting its failure to treat ocular burns,although it may prevent deeper infection when usedwith a tissue adhesive.21–28

Despite receiving eye irrigation at the district hospital,these patients sought treatment at our hospital with analkaline pH, probably because of an inadequate irrigationduring first aid. This suggests that there was a long period(1.17 days) in which the chemical was in contact with theocular surface. This long contact period may have contrib-uted to the severe nature of injury in most cases (grade VIinjury in 52%), which highlights the need for thoroughirrigation and immediate referral to tertiary eye carecenters.

The lower grade of injury in the younger patients may beattributed to their quicker reflexes and immediate responseat the time of accident, enabling them to minimizechemical contact time. The median BCVAs at presenta-
tion and at the 1-year follow-up were similar, indicating a

poor visual prognosis. Initial failure of epithelial regener-ation occurred, and a period of 3.5 months was required forall of the eyes to regain the epithelium. Thus, epithelialregeneration is poor in NaOH injury and may requireprolonged treatment. Treatment in the form of lubricatingdrops, amniotic membrane transplantation, and tarsor-rhaphy should be continued until all epithelial defectsheal.

Involvement of the limbus and conjunctiva are consid-ered the most important parameters for prognosis after anocular chemical injury. In our series, limbal stem celldeficiency of more than 6 clock hours was observed in 52%of the eyes. This observation can be attributed to thehighly alkaline pH of NaOH and the prolonged exposuretime (1.17 days). We also noted that an initial severe gradeof limbal and conjunctival involvement results in a sub-optimal final outcome and a higher chance of sequelae. Ahigher overall grade of injury also was associated withworse visual acuity, which does not improve significantlywith healing. Grade VI injuries had the worst prognosis,with significantly more epithelial defects, limbal involve-ment, and conjunctival involvement; longer healing time;and more complications than all of the lower-grade inju-ries. Therefore, it is prudent to prevent severe injuries withthe appropriate and urgent first aid.

The most common complication was secondary glau-coma, followed by cataract, which is consistent with thepenetrating nature of NaOH, with resultant damage to thetrabecular meshwork and lens. In addition, this findinghighlights the need to measure intraocular pressure to ruleout secondary glaucoma and to prevent any ensuingdamage to the head of the optic nerve. There was nosignificant difference in the distribution of cataract andglaucoma among the various grades of injury, but thenumber of cases was too small for a conclusive interpreta-tion. Conjunctival involvement understandably was re-lated to sequelae, such as symblepharon, pseudopterygium,and pyogenic granuloma. Complications such as scleralmelting, adherent leucoma, pseudocornea formation, in-fectious keratitis, and phthisis bulbi were observed, eventhough the patients received 1 month of inpatient therapywith satisfactory epithelialization and were under closesupervision with rigorous follow-up after discharge. Thispoor outcome highlights the severity of chemical injuriesand emphasizes the need to prevent their occurrence withproper disposal of industrial wastes.

Our case series is unique because it describes an acci-dental injury that occurred concurrently in all patients,with a chemical of the same concentration. Despite thiscommon exposure, the presenting ocular burns varied ingrade. Infection, melting, development of secondary glau-coma, and cataract formation resulted in some eyes,whereas others had good visual outcomes without signifi-cant complications.

In conclusion, the final visual outcome of a chemical
injury depends not only on the nature of the offending

1

1

1

1

1

chemical and the grade of injury, but also on the chemicalcontact time, patient age, the provision of prompt andappropriate first aid (specifically, a thorough wash of theocular surface into the conjunctiva until a neutral pH of7.0 is achieved), the prevention of secondary infection, thecontrol of glaucoma, and the provision of effective surgical
interventions. Grade VI injuries have a significantly poorer
of ascorbate treatment after severe experimental alkali burns

1

1

1

1

1

2

2

2

2

2

2

2


outcome than lower grade injuries, whereas there areinsignificant differences between grade II and III injuries.The standard treatment should be followed with themodification of therapy based on response and regularfollow-up. In addition, the provider should explain theguarded prognosis and the need for a prolonged therapeutic
regimen to the patient.
ALL AUTHORS HAVE COMPLETED AND SUBMITTED THE ICMJE FORM FOR DISCLOSURE OF POTENTIAL CONFLICTS OFinterest and none were reported. Involved in Design of study (N.S., A.S., D.S., S.G.); Conduct of study (A.S., D.S., T.V.); Data collection (D.S., A.S.,P.A., T.V.); Management (N.S., J.S.T., S.G.), analysis and interpretation (A.S., D.S., N.S.) of data; and Preparation, review, and approval of manuscript(N.S., D.S., J.S.T.). The manuscript conforms to the available author information and the study is in adherence with the tenets of the Declaration ofHelsinki. The study was approved by the authors’ institutional review board, the All India Institute of Medical Sciences Ethics Subcommittee, as aprospective study. Informed consent was obtained from all the subjects for treatment and participation in the study.

REFERENCES

1. Hall AH. Epidemiology of ocular chemical burn injuries. In:Shrage N, Burgher F, Blomet J, Bodson L, Gerard M, Hall A.eds. Chemical Ocular Burns: New Understanding and Treat-ments. Berlin: Springer-Verlag; 2011:9–17.

2. Spector J, Fernandez WG. Chemical thermal and biologicalocular exposures. Emerg Med Clin North Am 2008;26(1):125–136.

3. Pfister DA, Pfister RR. Acid injuries of the eye. In: KrachmerJH, Mannis MJ, Holland EJ, eds. Cornea: Fundamentals ofCornea & External Diseases. Vol. 2. St. Louis: Mosby;1997:1437–1442.

4. Pfister DA, Pfister RR. Alkali injuries of the eye. In:Krachmer JH, Mannis MJ, Holland EJ, eds. Cornea: Funda-mentals of Cornea & External Diseases. Vol. 2. St. Louis:Mosby; 1997:1443–1451.

5. Dua HS, King AJ, Joseph A. A new classification of ocularsurface burns. Br J Ophthalmol 2001;85(11):1379–1383.

6. Macdonald EC, Cauchi PA, Azuara-Blanco A, Foot B.Surveillance of severe chemical corneal injuries in the UK.Br J Ophthalmol 2009;93(9):1177–1180.

7. Saini JS, Sharma A. Ocular chemical burns-clinical anddemographic profile. Burns 1993;19(1):67–69.

8. Hong J, Qiu T, Wei A, Sun X, Xu J. Clinical characteristicsand visual outcomes of severe ocular chemical injuries inShanghai. Ophthalmology 2010;117(12):2268–2272.

9. Wagoner MD. Chemical injury of the eye- current conceptsin pathophysiology and therapy. Surv Ophthalmol 1997;41(4):275–313.

0. Gicquel JJ. Management of ocular surface chemical burns.Br J Ophthalmol 2011;95(2):159–161.

1. Hoang-Xuan T, Hannouche D. Medical treatment of ocularburns. J Fr Ophthalmol 2004;27(10):1175–1178.

2. Herr RD, White GL Jr, Bernhisel K, Mamalis N, Swan-son E. Clinical comparison of ocular irrigation fluids fol-lowing chemical injury. Am J Emerg Med 1991;9(3):228 –231.

3. Brodovsky SC, McCarty CA, Snibson G, et al. Managementof alkali burns. An 11-year retrospective review. Ophthalmol-ogy 2000;107(10):1829–1835.

4. Pfister RR, Paterson CA, Spiers JW, Hayes SA. The efficacy

depends upon the route of administration. Invest OphthalmolVis Sci 1980;19(12):1526–1529.

5. Pfister RR. Chemical injuries of the eye. Ophthalmology1983;90(10):1246–1253.

6. Brent BD, Karcioglu ZA. Effect of topical corticosteroids ongoblet-cell density in an alkali-burn model. Ann Ophthalmol1991;23(6):221–223.

7. Tsai JH, Derby E, Holland EJ, Khatana AK. Incidence andprevalence of glaucoma in severe ocular surface disease.Cornea 2006;25(5):530–532.

8. Vajpayee RB, Mukerji N, Tandon R, et al. Evaluation ofumbilical cord serum therapy for persistent corneal epithelialdefects. Br J Ophthalmol 2003;87(11):1312–1316.

9. Sharma N, Goel M, Velpandian T, Titiyal JS, Tandon R,Vajpayee RB. Evaluation of umbilical cord serum therapy inacute ocular chemical burns. Invest Ophthalmol Vis Sci 2011;52(2):1087–1092.

0. Salman IA, Gündogdu C. Epithelial healing in experimentalcorneal alkali wounds with nondiluted autologous serum eyedrops. Cutan Ocul Toxicol 2010;29(2):116–121.

1. Dua HS, Gomes JA, King AJ, Maharajan VS. The amnioticmembrane in ophthalmology. Surv Ophthalmol 2004;49(1):51–77.

2. Joseph A, Dua HS, King AJ. Failure of amniotic membranetransplantation in the treatment of acute ocular burns. Br JOphthalmol 2001;85(9):1065–1069.

3. Ricardo JR, Barros SL, Santos MS, Souza LB, Gomes JA.Amniotic membrane transplantation for severe acute cases ofchemical ocular burn and Stevens-Johnson syndrome. ArqBras Oftalmol 2009;72(2):215–220.

4. ller D, Pires RT, Mack RJ, et al. Amniotic membranetransplantation for acute chemical or thermal burns. Oph-thalmology 2000;107(5):980–989.

5. Tejwani S, Kolari RS, Sangwan VS, Rao GN. Role ofamniotic membrane graft for ocular chemical and thermalinjuries. Cornea 2007;26(1):21–26.

26. Tamhane A, Vajpayee RB, Biswas NR, et al. Evaluation ofamniotic membrane transplantation as an adjunct tomedical therapy as compared with medical therapy alonein acute ocular burns. Ophthalmology 2005;112(11):1963–1969.

7. Tandon R, Gupta N, Kalaivani M, Sharma N, Titiyal JS,
Vajpayee RB. Amniotic membrane transplantation as an

adjunct to medical therapy in acute ocular burns. Br JOphthalmol 2011;95(2):199–204.

28. Tang X, Wang Z, Dong N. Tissue adhesive combined withamniotic membrane adhering in the treatment of ocularburns. Yan Ke Xue Bao 2005;21(2):74–78.

29. Sharma A, Pandey S, Sharma R, Mohan K, Gupta A.Cyanoacrylate tissue adhesive augmented tenoplasty: a newsurgical procedure for bilateral severe chemical eye burns.Cornea 1999;18(3):366–369.

30. Tuft SJ, Shortt AJ. Surgical rehabilitation following severeocular burns. Eye (Lond) 2009;23(10):1966–1971.

31. Merle H, Gérard M, Schrage N. Ocular burns. J Fr Ophtalmol2008;31(7):723–734.

32. Mendicute J, Bidaguren A, Martínez-Soroa I, Gibelalde A.Automatized large diameter lamellar keratoplasty and stemcell transplantation for the treatment of ocular surfacediseases with limbal insufficiency. Eur J Ophthalmol 2008;
18(4):641–644.

33. Dohlman CH, Cade F, Pfister R . Chemical burns to theeye: paradigm shifts in treatment. Cornea 2011;30(6):613–614.

34. Rihawi S, Frentz M, Reim M, Schrage NF. Rinsing withisotonic saline solution for eye burns should be avoided.Burns 2008;34(7):1027–1032.

35. Rihawi S, Frentz M, Schrage NF. Emergency treatment ofeye burns: which rinsing solution should we choose? GraefesArch Clin Exp Ophthalmol 2006;244(7):845–854.

36. Ikeda N, Hayasaka S, Hayasaka Y, et al. Alkali burns of theeye: effect of immediate copious irrigation with tap water ontheir severity. Ophthalmologica 2006;220(4):225–228.

37. Hall AH, Maibach HI. Water decontamination of chemicalskin/eye splashes: a critical review. Cutan Ocul Toxicol2006;25(2):67–83.

38. Davis AR, Ali QK, Aclimandos WA, Hunter PA. Topicalsteroid use in the treatment of ocular alkali burns. Br J
Ophthalmol 1997;81(9):732–734.

Biosketch

Namrata Sharma is working as an Additional Professor in Cornea and Refractive Surgery services at Dr Rajendra PrasadCentre for Ophthalmic Sciences, All India institute of Medical Sciences, New Delhi, India. She has published five booksand has more than 225 scientific articles in peer-reviewed international journals. She has won the “Best of Show” awardsfive times at the American Academy of Ophthalmology for innovations in corneal transplantation surgery and has alsowon the Achievement Award in 2005.

COURSE AND OUTCOME OF ACUTE ALKALI OCULAR INJURYVOL. 154, NO. 4 749.e1

Biosketch

Digvijay Singh, MD, specialized in ophthalmology from Dr Rajendra Prasad Centre for Ophthalmic Sciences, All IndiaInstitute of Medical Sciences, India, where he works in strabismus, neuro-ophthalmology and glaucoma services. He holdsvarious publications, book chapters and national and international presentations to his credit and also numerous awardsincluding Sorel Catherine Freymann prize for pediatrics, New Zealand High Commissioner’s prize for community medicineand Dr Atm Prakash gold medal for surgery by his institution.

AMERICAN JOURNAL OF OPHTHALMOLOGY749.e2 OCTOBER 2012

course and outcome of accidental sodium hydroxide ocular injury

Documents