Intended for healthcare professionals


Congenital rubella in south India: diagnosis using saliva from infants with cataract

BMJ 1996; 312 doi: (Published 20 January 1996) Cite this as: BMJ 1996;312:161
  1. M B Eckstein, research fellowa,
  2. D W G Brown, director enteric and respiratory virus laboratoryb,
  3. A Foster, senior lecturera,
  4. A F Richards, medical laboratory scientific officerb,
  5. C E Gilbert, lecturera,
  6. P Vijayalakshmi, consultantc
  1. a Department of Preventive Ophthalmology, Institute of Ophthalmology, London EC1V 9EL
  2. b PHLS Virus Reference Division, Colindale, London NW9 5HT
  3. c Aravind Eye Hospital and Postgraduate Institute of Ophthalmology, Madurai, south India
  1. Correspondence to: Dr Eckstein.
  • Accepted 1 November 1995

Congenital rubella is a preventable disease which has been largely controlled by immunisation in the developed world.1 Serological surveys in India indicate that up to 45% of women of childbearing age are susceptible to rubella and potentially at risk of infection during pregnancy.2 We tested affected infants to see whether detection of rubella specific IgM from saliva is as reliable as from serum for diagnosing rubella infection and whether rubella is an important cause of congenital cataract in south India.

Patients, methods, and results

We studied 95 consecutive infants with congenital cataract presenting to the paediatric department of the Aravind Eye Hospital in 1993-4. Thirty six age matched children attending the same clinic over the same period with a diagnosis of watering eyes acted as controls. Serum samples were taken from 61 children with cataract and saliva samples from all 131 children. Saliva samples rich in crevicular fluid were collected using the Orasure device (Epitope Inc, Beaverton) according to the manufacturer's instructions. Serum and saliva samples were tested for rubella specific IgM by antibody capture.3 4 Specimens were considered positive if the test to negative control ratio exceeded 3:1.

The mean (range) age of the 95 cases was 6.2 months (1-11) and of the 36 controls 7.3 months (1-11). Rubella specific IgM was detected in saliva and serum in 17 paired samples and was absent in 44 paired samples (sensitivity 100%, 95% confidence interval 80.5% to 100%; specificity 100%, 95% confidence interval 92% to 100%). Saliva testing gave no false positive and no false negative results compared with serum.

Twenty five of the 95 infants with cataract (26.3%) had congenital rubella infection confirmed by detection of rubella specific IgM in saliva. None of the controls had raised rubella specific IgM (χ2=11.71, P<0.005). Congenital rubella was suspected clinically in 19 of the 25 cases of congenital cataract in which rubella specific IgM was detected (sensitivity 76%, specificity 100%). The other six children had clinical features compatible with congenital rubella syndrome but rubella had not been diagnosed before laboratory confirmation. The general and ocular chracteristics of the 25 children with rubella cataract are shown in table 1.

Table 1

Characteristics of 25 cases of serologically proved cataract due to congenitally acquired rubella. Values are numbers (percentages)

View this table:

Sequential saliva samples taken from seven infants with congenital rubella showed that in all cases raised concentrations of rubella specific IgM persisted up to 6 months of age and in five they persisted up to 14 months.


This study showed excellent agreement between test results in saliva and serum in the 61 paired samples, suggesting that saliva is as reliable as serum in allowing the detection of rubella specific IgM in infants. Although the infants tested had few teeth and therefore produced little crevicular fluid, adequate saliva samples were collected from all cases. The storage and transport of samples was not ideal, but this study indicates the potential of the Orasure device for undertaking studies remote from testing centres.

The study also suggests that congenital rubella is a significant cause of congenital cataract in south India. Nevertheless, this group of children admitted to hospital may not be representative of all children with congenital cataract in the community and therefore more extensive studies will be needed to establish the true scale of the problem.

Improved surveillance is required for congenital rubella in developing countries. The identification of rubella specific IgM in saliva offers a simple, non-invasive test that can be used in infants to enhance surveillance based on clinical case finding.

We thank Epitope Inc for supplying the Orasure devices.


  • Funding This study was supported by a grant from Sight Savers International and the British Council for Prevention of Blindness.

  • Conflict of interest None.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
View Abstract