Indian Pediatrics 1999; 36:697-700 

Serology in Congenital Infections: Experience in Selected Symptomatic Infants

The role of many microbial agents in the causation of congenital disease has been extensively studied over the last five decades. The magnitude of the problem in India has not, however, been adequately investigated. There have been studies in which congenital diseases due to rubella virus (RV)(1,2) and cytomegalovirus (CMV)(3) have been described. However, few studies have attempted to comprehensively identify the role of a spectrum of teratogenic agents known to be vertically transmitted from mother to infant. We investigated the frequency of the presence of immunoglobulin M (IgM) antibody to Toxoplasma gondii, RV, CMV, herpes simplex virus (HSV) ("TORCH" agents) and parvovirus B 19 (PB 19) in infants clinically suspected to have congenital infection.

Subjects and Methods

Single serum samples were collected from 92 infants at the Christian Medical College Hospital Neonatology and Pediatrics Out-Patient Clinics and Wards between January 1996 and December 1997. These infants presented with neonatal cholestasis, hematological, cardiac, neurological, ophthalmic, dysmorphic and/or other abnormalities compatible with congenital infections. The sera were tested for IgM antibodies using commercial kits to T. gondii (Eurogenetics, Tessenderlo, Belgium), RV (Centocor, Malvern, USA), PB 19 (Biotrin International, Dublin, Ireland) and CMV (IMx system, Abbott Laboratories, Chicago, USA). Anti-HSV IgM was checked for by indirect immunofluorescence assay (IFA) (4).

Results

None of the infants tested for HSV IgM (n = 90) or T. gondii IgM (n = 86) was positive. IgM antibody positivity to RV, PB 19 and CMV is shown in Table I. One infant positive for RV IgM and another for PB 19 IgM were additionally positive for CMV IgM. Sixteen infants were found to have CMV IgM antibody; 13 of them had earlier received blood transfusion which could have been a possible source of CMV infection. After scrutiny of transfusion history, 4 infants (4.3%) were considered likely to have been congenitally infected with CMV.

Table 1

Virus-Specific IgM in Infants with Suspected Congenital Infection

Age (mo)	No. tested	IgM antibody positive No. (%)
		RV	PB19	CMV
<1	54	3 (5.6)	0	1 (2.0)*
1-2	19	3 (15.8)	0	6 (33.3)**
3-6	17	3 (17.6)	1 (5.9)	9 (52.9)
7-12	2	0	0	0
All	92	9 (9.8)	1 (1.1)	16 (18.2)***

* Only 51 of 54 tested.
** Only 18 of 19 Tested.
***Only 16 of 88 Tested.

The 9 RV IgM positive infants had typical manifestations of congenitally acquired rubella (Table II). The PB 19 IgM positive infant also had CMV IgM, both tests being repeated. It is likely that PB 19 was the etiological agent as the clinical presentation was with upper gastrointestinal (GI) bleed and ecchymotic patches from birth. The four CMV IgM positive infants had neonatal hepatitis, hepatosplemomegaly, congenital cataract and! or coagulation disorder, features compatible with congenital CMV infection.
 

TABLE II

Clinical Profile of Congenital Infection of Viral Etiology

Predominant clinical features	No.	Virus Etiology
1. Neonatal    Cholestatis	23	RV:1. PB19:1, CMV:3
2. Dysmorphism*	21	RV:5, CMV: 1
3. hematological	14	-
4. Neurological	13	RV: 1
5. Hepatosplenomagely	11	RV:2
6. Cardiac	2	-
Miscellaneous**	8	-
All	92	14**

*  PDA, microcephaly ± cataract, other malformations.
**  Respratory, metabolic
*** Congenital cataract seen in 11 patients (5 RV IgM positive)

Discussion

Among the infants suspected to have congenital infection in this series, 9.8% were found to be RV IgM positive. As all these infants presented with features compatible with congenitally acquired RV infection, the possibility of postnatal acquisition of RV in the older infants can be ruled out. However, false negativity with the use of RV specific IgM alone for diagnosis in symptomatic infants upto 6 months of age is 10% while between 6 and 12 months, it is more than 50%(5). This degree of false negativity may be overcome by the use of RT-PCR(6). Detection of persistent RV specific IgG in serum or saliva in infants above 8 months may be useful when infants present too late for detection of RV specific IgM(5).

Although, 16 infants were found to be CMV IgM positive in this series, 13 had earlier received blood transfusion and interpretation needs to be cautious. CMV infection may be transmitted through blood transfusion and the lag period to development of IgM antibody is about 4 weeks(7). One infant (1-2 mo) who had not received transfusion was CMV IgM and RV IgM positive. As this child had bilatieral pigmentary mottling of the optic fundi and hepatosplenomegaly, we have assigned significance to the RV IgM positivity. Another infant transfused five days before testing was CMV IgM and PB 19 IgM positive and presented with neonatal hepatitis, GI bleed and ecchymotic patches from birth. It is likely that PB 19 is the etiological agent in this child as infection with PB 19 in the later phases of pregnancy may share common but nonspecific features of other congenital infections(8). Only four other infants were considered likely to be congenitally infected with CMV: 1 was a 10 day old neonate transfused 2 days before testing, 1 child (1-2 mo) was not transfused and 2 infants (1-2 mo) were transfused the day before testing. In the other 10 older infants, post- natal infection cannot be ruled out.

It must be noted that the standard diagnostic test for congenital infection with CMV is viral culture within the first three weeks of life. Where this is not possible, there may be a limited role for CMV IgM testing, bearing in mind the causes of false positivity as described above and false negativity (>35%) especially in asymptomatic infected infants and when maternal primary infection occurs late in pregnancy(9). In our study, we ruled out false positivity due to rheumatoid factor by absorption using the neutralization reagent provided by the manufacturer.

IgM antibody testing for PB 19 in serum of infants is also of low diagnostic value as the false negativity may be as high as 85%. The gold standard to confirm B 19 infection would be detection of viral DNA (70% detection rate), IgM antibodies or both(9).

None of the infants in this series had clinical features suggestive of HSV infection. In our collective experience of over 25 years in Vellore, the vesicular rash of congenital HSV infection has been very rarely seen in our hospital. Studies done elsewhere have also shown that neither the characteristic clinical findings nor elevated titres of antibody to HSV were found in autopsy reports of neonates (J 0). If lesions are present in the neonate, the method of choice for diagnosis of congenital HSV disease is virus culture. HSV IgM antibody may be seen in the first 2-3 weeks of illness and detectable for 6-8 weeks(9).
It is known that anti-toxoplasma IgM and/ or IgA detection has a sensitivity of 77% only as compared with 90% sensitivity with other tests like enzyme-linked immunofiltration assay (ELIFA) with anti- T.gondii IgM or IgA immunocaputre(11). A combination of tests may, hence, be needed to determine the magnitude of congenital toxoplasmosis in our region.

This study shows that the rORCH and PB 19 agents account for an etiology in only about 15% of infants clinically suspected to have congenital infection, using IgM detection assays. It may be that more sensitive tests will be needed to identify etiological agents in the other infants. Alternatively, the search will have to include other agents of congenital abnormalities.

Acknowledgements

We greatly appreciate the help given to us by Dr. Annie Sudarsanam, Professor and former Head of the Department of Clinical Pathology and Blood Bank, Christian Medical College and Hospital, Vellore, in obtaining transfusion histories of some of the infants in this series.
 

 References

1. Chaturvedi UC, Tripathi BN, Mathur A, Singh UK, Mehrotra RM. Role of rubella in congenital malformations in India. J Hyg Lond 1976; 76: 33-40.

2. Manjunath N, Balaya S. Serological study on congenital rubella in Delhi. Indian J Med Res 1984; 79: 716-721.
3. Sharma R, Bahl L, Goyal A, Sharma A, Sharma M, Thakur JS. Congenital cytomegalovirus infection in Shimla Hills, Himachal Pradesh, India. J Commun Dis 1995; 27: 23~26.

4. Rawls WE. Herpes simplex virus types I and 2 and Herpesvirus simiae. In: Diagnostic Procedures for Viral, Rickettsial and Chlamydial infections, 5th edn. Eds. Lenette EH, Schmidt NJ. Washington DC, American Public Health Association Inc, 1979; pp 309-373.

5. Banatvala JE, Best JM. Rubella. In: Topley and Wilson's Microbiology and Microbial Infections, Vol I: Virology, 9th edn. Eds. Mahy BWJ, Collier L. London, Arriold, 1998; pp 551- 577.

6. Bosma TJ, Corbett KM, Eckstein MB, O'Shea S, Vijayalakshmi P, Banatvala JE, et al. Use of PCR for prenatal and postnatal diagnosis of congenital rubella. J Clin Microbiol 1995; 33: 2881-2887.

7. Hirsch MS. Cytomegalovirus infection. In: Harrison's Principles of Internal Medicine, Vol. I, 13th edn. Eds. Isselbacher KJ, Braunwald E, Wilson JD, Martin JB, Fauci AS, Kasper DL. New York, McGraw-Hill Inc, 1994; pp 794- 797.

8. Vogel H, Kornman M, Ledet SC, Rajagopalan L, Taber L, McClain K. Congenital parvovirus infection. Pediatr Pathol Lab Med 1997; 17: 903-912.

9. Enders G. Viral infections of the fetus and neonate, other than rubella. In: Topley and Wilson's Microbiology and Microbial Infections, Vol. I: Virology, 9th edn. Eds. Mahy BWJ, Collier L. London, Arnold, 1998; pp 873-915.

10. Nakamura Y,Yamamoto S, Tanaka S, Yano H, Nishimura G, Saito Y, et al. Herpes simplex vi- ral infection in human neonates: An immunohistochemical and electron microscopic study. Hum Patho11985; 16: 1091-1097.

11. Pinon JM, Chemla C, Villena I, Foundrinier F, Aubert D, Puygauthier - Toubas D, et al. Early neonatal diagnosis of congenital toxoplasmosis: Value of comparative enzyme - linked immunofiltration assay immunological profiles and anti - Toxoplasma gondii immunoglobulin M (IgM) or IgA immunocapture and implications for postnatal therapeutic strategies. J Clin Microbiol1996; 34: 579-583.