Secondary Logo

Journal Logo

Original Article

Screening for retinopathy of prematurity—a comparison between binocular indirect ophthalmoscopy and RetCam 120

Shah, Parag K. DNB; Narendran, V. DNB; Saravanan, V. R. FRCS; Raghuram, A. FRCS; Chattopadhyay, Abhijit MS; Kashyap, Maithreyi DNB

Author Information
Indian Journal of Ophthalmology: Jan–Mar 2006 - Volume 54 - Issue 1 - p 35-38
doi: 10.4103/0301-4738.21612
  • Open

Abstract

Screening for retinopathy of prematurity (ROP) examination by traditional methods is difficult and requires an ophthalmologist with specialized training in pediatric retina. Many regional hospitals do not have staff to effectively screen patients for ROP. Early detection and treatment of threshold ROP has been shown to decrease severe visual loss in premature infants,1 and recently, early treatment of high-risk prethreshold ROP has shown to reduce unfavorable outcomes significantly.2 The binocular indirect ophthalmoscope (BIO) is the gold standard for the detection of ROP,345 but is time-consuming and stressful for the neonates to undergo this examination.6 We used RetCam 120 camera system (Massie Research Laboratories, Inc, Dublin, CA) as a screening tool and compared it with BIO in screening of ROP.

Materials and Methods

This prospective, comparative study included 87 examinations comprising 27 consecutive infants who were evaluated from March to June 2003 from a single neonatal intensive care unit (NICU). Fourteen infants underwent two examinations, one underwent three, eight underwent four, and four underwent six examinations. Infants with gestational age less than 36 weeks and/or birth weight of less than or equal to 2000 g were included in the study (the screening criteria were based on our personal experience). One infant with 2050 g was screened as requested by the neonatologist. Gestational age was determined by an experienced neonatologist based on menstrual history, obstetrical dating (including ultrasound data when available), and neonatal physical examination. Infants with hazy media or nondilating pupil whose RetCam pictures were of poor quality (owing to the initial learning curve) were excluded.

Screening was conducted by 31-week postconceptional age (PCA) or 4-week postnatal age, whichever was earlier.7 BIO was performed by a single, experienced vitreoretinal surgeon (VRS) and RetCam examination was performed by another vitreo retinal surgeon (PKS) specifically trained to take RetCam pictures. The examinations were continued until either the retina was completely vascularized or retinopathy developed. When ROP progressed to threshold or fulminate stage 3A,8 patients were treated with double-frequency neodymium-yttrium aluminum garnet laser with indirect ophthalmoscope delivery system within 72 h of diagnosis.

The infant's pupils were dilated with a combination of 0.5% cyclopentolate and 2.5% phenylephrine. They were dilated 30-60 min before the scheduled examination time. To reduce the risk of aspiration the next feeding was delayed for a period of 30 min to 2 h after examination; 0.5% proparacaine was used as topical anesthetic. A sterilized lid speculum and scleral depressor (Schoket type) were used for examination of each infant along with a BIO and 20D lens. The anterior segment was examined first with 20D lens by going close to the eye. Posterior pole was examined next, followed by peripheral retina with scleral depression.

The RetCam 120 was used to photographically document the fundus features at the same visit (Figure 1). A series of photographs were taken to adequately capture the posterior pole, and as much as possible, the periphery [Figure 2 (A,B)]. Each series was saved and the images were transferred to a file devoid of patient identifying information. All photographs taken from initial and follow-up examinations were mixed for reading purposes and each session was identified by the randomization number. The RetCam images were read in a masked manner by the same examiner (VRS) who performed clinical examination after the study period was over. The presence of laser photocoagulation scars, if seen, was noted. The distance between the optic nerve and fovea was measured for each posterior pole photograph to determine zone 1. The RetCam failed to capture zone 3 without scleral indentation. So, if ROP was seen in the photograph, it was assumed to be located in either zone 1 or zone 2.

F1-7
Figure 1:
RetCam 120 being employed to obtain digitized image
F2-7
Figure 2A:
RetCam picture of right eye showing threshold ROP
F3-7
Figure 2B:
RetCam picture showing regressing ROP post laser photocoagulation

For each patient, gestational age, birth weight and PCA at the time of each examination was noted. The presence or absence of plus disease with the presence or absence of ROP, the stage, zone, and number of clock hours involved according to the International Classification for ROP9 were extracted from both the clinical record as well as the RetCam sessions. These data were then tabulated and compared for both the methods of examination.

We calculated the sensitivity and specificity for the RetCam, keeping the clinical examination with BIO as the gold standard. Positive predictive value (PPV) and negative predictive values (NPV) were also calculated to assess the clinical usefulness of the digital image readings assuming the observed prevalence in the study data.

Results

Of the 27 infants examined, the mean birth weight was 1468.88 g (range: 900-2050 g), mean gestational age was 32.33 weeks (range: 28-36 weeks), mean PCA at first RetCam examination was 35.63 weeks (range: 33.2-44 weeks), and mean PCA at last RetCam evaluation was 38.28 weeks (range: 33.5-44 weeks). A total of 87 consecutive BIO and RetCam examinations were performed. Twenty eight eyes of fourteen infants contributed one examination. BIO examination showed ROP in 63 examinations. RetCam examination showed ROP in 56 examinations (Table 1).

T1-7
Table 1:
Clinical and RetCam examination findings

Nine examinations with Retcam were false-negative and two were false-positive. The sensitivity of RetCam with 95% confidence interval was 85.71% (84.1, 87.32) and the specificity was 91.66% (90.05, 93.27). The PPV was 96.43% (94.81, 98.04) and NPV was 70.97% (72.58, 69.35).

Of the 54 examinations in which the clinical and RetCam examination both revealed the presence of ROP, 100% were located in zone 2 or zone 1. Ten examinations were judged to be fulminate ROP in zone 1, two examinations showed fulminate in zone 2, seven examinations showed stage 3 in zone 2, five examinations showed stage 2 in zone 2, and two examinations showed stage 4b in zone 1, both by BIO and RetCam examinations. Four examinations were evaluated as stage 3 in zone 2 by RetCam but fulminate ROP in zone 2 by BIO (all were treated). Two examinations were evaluated as stage 2 in zone 2 by RetCam but stage 1 in zone 2 by BIO (both were only observed). 22 examinations were postlaser (eight from zone 1 onwards and 12 from zone 2 onwards).

Of the nine false-negative results (ROP by clinical examination, no ROP by RetCam), two were in zone 2 (outer half) and seven in zone 3. Five had stage 1, four had stage 2, and two had plus disease. Of the two false-positive results (no ROP by clinical examination, ROP by RetCam) both were in zone 2 with stage 2 and none had plus disease. Clinical examination later revealed that photographic interpretation was correct.

Discussion

Intensive screening of premature babies in the NICU is the only way to detect ROP. At present its detection requires an ophthalmologist experienced in examining pediatric retina with BIO and scleral depression, which is stressful for the infants6 as well as time-consuming for the ophthalmologist. Moreover, the ophthalmologist has to coordinate his schedule with the NICU. Hand-drawn sketches are the only way to document the status of the premature retina.

An ideal screening modality would be one that is least traumatic to the infants, used by the NICU staff at their convenience, and permanently document any retinal finding in the new born. Any new test designed to replace the gold standard (BIO) must have an acceptable sensitivity and specificity. A sensitivity of 80% and a specificity of 90% of a test is acceptable in clinical medicine to diagnose disease.10 In our study, the sensitivity and specificity of RetCam were 85.71% and 91.66% respectively, which are similar to a previous study by Roth.11

The PPVs and NPVs were 96.43% and 70.97%, respectively, which is also comparable with the study by Roth.11 The PPVs and NPVs for threshold ROP and high risk pre-threshold ROP were 100%, which is comparable with the study by Ells et al.12 In another study screening was divided into two intervals of postnatal age, in which both early and the late stages of ROP have been compared. The sensitivity and specificity for detecting ROP were 76% and 100% respectively, and for detecting threshold ROP, both were 100% for the latter examination.13 In our study and also in the studies by Roth11 and by Schwartz,14 no cases of threshold ROP were missed.

RetCam missed nine cases of ROP. The diseases missed by imaging were mostly in zone 3 and some in outer zone 2. They were either stage 1 or stage 2. All these cases regressed well later without treatment. Although the occurrence of complications for disease in this location is low, they do occur.15 Moreover, the RetCam was extremely efficient in detecting and documenting ROP in posterior location.

Some of the drawbacks of this study are being described. First, sensitivity and specificity may have different numerical values when they are obtained using a group of patients with early stages of a disease compared with the sensitivity and specificity obtained in a group of patients with more advanced disease. In this study, screening of the infants could have been divided between two intervals of postnatal age, in which both early and the later stages of ROP could have been compared as done in the study by Yen.13 Second, the 50-50 split between the group of patients having ROP and the group not having ROP using the gold standard (BIO) could have given the greatest statistical power for the sample size. Third, 22 examinations were postlaser. As laser was done for posterior disease, the RetCam could easily pick it up. There was selection bias that could have been avoided.

Telemedicine is becoming more prevalent in the clinical practice of medicine and has been applied to ophthalmology as well.1416 RetCam provides a unique technology to enable NICU staff to obtain and transmit digital images to centers where ROP expertise is available. It is a versatile system that may be a welcome addition to NICU and the ophthalmology department. This is the first study on RetCam from India. However, at present, affordability is a barrier that will inhibit the adoption of RetCam in India.

Sensitivity and NPV of RetCam are insufficient. This is mainly because it is unable to photograph the pediatric retina till the ora, owing to the restricted mobility of the camera head due to its size and the metallic arms of the lid speculum (Figure 3). Both of these are mechanical problems which need to be fixed.

F4-7
Figure 3:
Shows incompatibility between the lid speculum arms and the camera head

In conclusion, RetCam is a new technique that can be an alternative to BIO for screening of ROP. However, BIO for screening infant eyes still stays as the primary key diagnostic tool. The major disadvantages include contact method of examination, large size of the lens that is incompatible with the small palpebral fissure in preterm infants, and huge costs involved, especially for a developing country like India.

1. Cryotherapy for Retinopathy of Prematurity Cooperative Group. . Multicenter trail of cryotherapy for retinopathy of prematurity. Preliminary results Arch Ophthalmol. 1988;106:471–9
2. Early Treatment for Retinopathy of Prematurity Cooperative Group. . Revised indications for the treatment of retinopathy of prematurity Arch Ophthalmol. 2003;121:1684–96
3. Cryotherapy for Retinopathy of Prematurity Cooperative Group. . Multicenter trial of cryotherapy for retinopathy of prematurity: three-month outcome Arch Ophthalmol. 1990;108:195–204
4. Cryotherapy for Retinopathy of Prematurity Cooperative Group. . Multicenter trail of cryotherapy for retinopathy of prematurity: one-year outcome: structure and function Arch Ophthalmol. 1990;108:1408–16
5. Pamer EA, Flynn JT, Hardy RJ, Phelps DL, Phillips CL, Schaffer DB, et al Incidence and early course of retinopathy of prematurity Ophthalmology. 1991;98:1628–40
6. Laws DE, Morton C, Weindling M, Clark D. Systemic effects of screening for retinopathy of prematurity Br J Ophthalmol. 1996;80:425–8
7. Reynolds JD, Dobson V, Quinn GE, Fielder AR, Palmer EA, Saunders RA, et al Evidence based screening criteria for ROP. Natural history data from the CRYO-ROP and LIGHT-ROP studies Arch Ophthalmol. 2002;120:1470–6
8. Shapiro MJKumar H, Shapiro MJ, Azad RV. Type 2 or fulminate ROP A Practical Approach to Retinopathy of Prematurity Screening and Management. 2001 New Delhi Malhotra Enterprises:23–33
9. The Committee for the classification of Retinopathy of Prematurity. . An international classification for retinopathy of prematurity Arch Ophthalmol. 1984;102:1130–4
10. Riegelman RK, Hirsch RP. Diagnostic discrimination of test Studying a Study and Testing a Test.2nd edn New York Little, Brown and Company:151–63
11. Roth DB, Morales D, Feuer WJ, Hess D, Johnson RN, Flynn JT. Screening for retinopathy of prematurity employing the RetCam 120 Arch Ophthalmol. 2001;119:268–72
12. Ells AL, Jonathan HM, Astle WF, Williams G, Leske DA, Fielden M, et al Telemedicine approach to Screening for Severe Retinopathy of Prematurity. A Pilot Study Ophthalmology. 2003;110:2113–7
13. Yen KG, Hess D, Burke B, Johnson RA, Feuer WJ, Flynn JT. Telephotoscreening to detect retinopathy of prematurity: Preliminary study of the optimum time to employ digital fundus camera imaging to detect ROP J AAPOS. 2002;6:64–70
14. Schwartz SD, Harrison SA, Ferrone PJ, Trese MT. Telemedical evaluation and management of retinopathy of prematurity using a fibreoptic digital fundus camera Ophthalmology. 2000;107:25–8
15. Carden SM, Good WV. Macular folds and poor vision associated with zone 3 retinopathy of prematurity Am J Ophthalmol. 1998;126:460–2
16. Sajeesh Kumar KR, Yogesan K, Constable IJ. Tele-ophthalmology in India? Is it here to stay Indian J Ophthalmol. 2003;51:295–6
Keywords:

RetCam 120; retinopathy of prematurity

© 2006 Indian Journal of Ophthalmology | Published by Wolters Kluwer – Medknow