Secondary Logo

Journal Logo

Brief Reports

The natural course of nonsuppurative Calmette-Gu↑erin bacillus lymphadenitis

Singla, Anupam M.D.; Singh, Surjit M.D.; Goraya, Jatinder S. M.D.; Radhika, Srinivasan M.D.; Sharma, Meera M.D.

Author Information
The Pediatric Infectious Disease Journal: May 2002 - Volume 21 - Issue 5 - p 446-448
  • Free

In a tertiary level hospital in North India, we studied the outcome of infants with nonsuppurative Calmette-Guérin bacillus (BCG) lymphadenitis managed conservatively. Twenty-three infants with nonsuppurative BCG lymphadenitis diagnosed on the basis of clinical presentation, evidence of granulomatous inflammation and demonstration of acid fast bacilli in the aspirated material were followed prospectively without being offered specific antitubercular drug therapy. Twenty patients were available for final analysis; three were lost to follow-up. Seventeen (85%) had spontaneous regression of BCG lymphadenitis without progression and drainage. The mean time to resolution was 9.1 months. Three (15%) patients developed suppuration and drainage, but even these children had uneventful healing on conservative management only. Nonsuppurative BCG lymphadenitis follows a benign course in most individuals. Most of the cases regress spontaneously with conservative management.

Enlargement of regional lymph nodes, with or without suppuration, is a well-known complication of Calmette-Guérin bacillus (BCG) vaccination, occurring in 0.013 to as many as 23% of the vaccinated children. 1–5 Its incidence is believed to be influenced by a variety of host- and vaccine-related factors. 5, 6 Even though it is well-known that the natural course of BCG lymphadenitis, whether suppurative or not, is spontaneous regression, it is common practice to treat these children with various combinations of antitubercular drugs. 7, 8 Some national expert groups also suggest the use of such drugs in this condition. 9 There is, however, no evidence in literature to suggest that these interventions hasten recovery. 7, 8, 10–14 This fact has also been corroborated in a recent metaanalysis. 15 Conservative management with expectant follow-up is the acceptable alternative. We describe our experience with managing nonsuppurative BCG lymphadenitis conservatively.

Patients and methods.

This prospective study enrolled patients attending the Pediatric Outpatient Department of Nehru Hospital, Post Graduate Institute of Medical Education and Research, Chandigarh, with axillary or supraclavicular lymph node enlargement after BCG vaccination. Patients entered the study from January, 1994, to July, 1995. In India the Danish 1331 strain is used for vaccination, and the injection is given intradermally over the left deltoid area.

BCG lymphadenitis was defined as the development of an ipsilateral axillary or supraclavicular lymph node swelling (≥2cm) within 6 months of BCG vaccination. Fine needle aspiration cytology was performed at presentation using a 23-gauge needle attached to a 20-ml syringe in a specially designed holder (Cameco). The aspirated material was subjected to bacteriologic (Gram stain, routine bacterial culture on blood and MacConkey agar, Ziehl-Neelsen stain and culture for mycobacteria on Lowenstein-Jensen medium) as well as cytologic examinations (May-Grünwald-Giemsa and hematoxylin-eosin stains). A patient was entered into the study protocol only if there was evidence of acid-fast bacilli (AFB) and granulomatous inflammation in the fine needle aspirate from the lymph node. Infants with other clinically obvious causes of lymphadenitis or those who had received antitubercular drugs for >1 month before presentation were excluded, as were those who had already developed suppuration at the time of presentation.

All babies with nonsuppurative BCG lymphadenitis were subjected to a Mantoux test (1 tuberculin unit of purified protein derivative) and a chest roentgenogram. The patients were followed up at 2-month intervals (or earlier if signs of suppuration developed) until the time of resolution. Defaulters were sent postal reminders. Resolution was defined as the node becoming either nonpalpable or <0.5 cm in size, and there being no tenderness, redness, sinus tract or ulcer formation. Abscess formation (suppuration), defined by the development of redness of the overlying skin and/or fluctuations, was taken to indicate failure of conservative management. No antitubercular treatment was administered to any of the children during follow-up. The Institute Ethics Committee approved the study protocol.


A total of 23 patients met the inclusion criteria. Three were lost to follow-up and were excluded from the analysis. Among the remaining 20 patients there were 15 boys and 5 girls. The time of BCG vaccination was at birth (n = 8), within 1 month of birth (n = 7), between 1 and 2 months (n = 3) and at 5 months of age (n = 2). Mean age at presentation was 7.6 months (range, 2.5 to 16 months). The interval between BCG vaccination and development of lymphadenitis was <2 months in 2, between 2 and4 months in 10 and between 4 and 6 months in 8 (mean, 3.8 months; range, 1 to 6 months). The interval between the development of BCG lymphadenitis and presentation to the hospital was <15 days in 5, between 15 days and 3 months in 9 and >3 months in 6.

All infants had left axillary lymphadenitis, and one infant had supraclavicular lymph node enlargement as well. In 18 infants only 1 lymph node was palpable, and 2 infants had 2 enlarged lymph nodes. The size of lymph node varied from 2 to 4 cm (mean, 2.7 cm), and the consistency was firm. All infants showed tuberculin reactivity of ≥5 mm, 1 had between 6 and 10 mm, 9 had between 10 and 15 mm and it was ≥15 mm in 10 infants. Evidence of granulomatous inflammation and AFB positivity in the smears made from the aspirate was present in all (by definition). Mycobacterium bovis was cultured from only 1 patient, and Staphylococcus aureus was isolated from 2 cases. Chest radiographs were normal in all children.

Three infants could not be traced after being enrolled into the study protocol. Of the remaining 20 spontaneous lymph node regression occurred in 17 (85%). Mean time for resolution was 9.1 months (range, 2.5 to 19 months). The remaining 3 infants developed suppurative changes on follow-up at intervals of 3, 4.5 and 19 months, respectively, after the onset of lymphadenopathy. Two of these infants had received BCG vaccination at birth, whereas the third had received it at 5 months of age. Perforation and drainage occurred in these infants at 5, 2 and 4 weeks, respectively, after the development of suppuration. After drainage complete healing occurred at 3, 10 and 12 weeks, respectively. The mean resolution period for suppurative lymphadenopathy was 11.8 months (range, 5 to 23 months). None of the patients was given any specific antimicrobial or antitubercular therapy.


Two forms of BCG lymphadenitis can be recognized in its natural course. The nonsuppurative, also called simple or hypertrophic BCG lymphadenitis, 3, 16, 17 is characterized by palpable glands firm in consistency. These nodes usually regress spontaneously but sometimes might enlarge and develop into abscesses (suppurative BCG lymphadenitis). 7, 8, 10–14 Suppurative lymphadenitis can also develop abruptly within 2 to 4 months of BCG vaccination. 11, 16, 18 Suppuration is an undesirable outcome in BCG lymphadenitis.

Our study has shown that a benign outcome with spontaneous regression can be expected in the majority of cases who present with nonsuppurative BCG lymphadenitis. This is significant when one considers the fact that treatment with antitubercular drugs 10–14 or antimicrobials (like erythromycin) 11, 12, 14 neither hastens the resolution of BCG lymphadenitis nor prevents its progression into suppuration. The rate of spontaneous regression in our study is higher compared with the previously published reports (Table 1). This could be because of the following two factors: (1) we used stringent inclusion criteria for enrollment to prevent contamination of results that would necessarily have occurred if the diagnosis of BCG lymphadenitis were to be made on clinical grounds alone; (2) according to our protocol all affected lymph nodes were aspirated at presentation for diagnostic purposes. This might have had some therapeutic effect as well, resulting in the low rate of spontaneous drainage in our patients. 18, 19 We excluded patients with suppurative BCG lymphadenitis because suppuration was an important therapeutic end point in our study and patients with suppurative lymphadenitis form a distinct subgroup that is managed differently, either by needle aspiration 18, 19 or by surgical excision. 18–20

Table 1
Table 1:
Spontaneous regression rates in BCG lymphadenitis among various studies

Although AFB positivity was present in all the cases, the yield of Mycobacterium culture was negligible. Previous studies 12, 13, 14 also have reported a low yield of Mycobacterium for reasons not entirely clear. There is some evidence that the isolation rate of mycobacteria may be high early in the course of lymphadenopathy 20 or in patients with disseminated disease. 3 Isolation of S. aureus, which was seen in two of our patients, has also been reported in previous studies, 1, 12 but the significance of this organism in BCG lymphadenitis is not clearly understood. None of the two aforementioned children had clinical symptoms or signs of bacterial infection and improved without any specific antimicrobial therapy.

Only three (15%) patients in our study developed suppuration with discharge and sinus formation followed by rapid healing. However, the interval between the development of lymphadenopathy and ultimate healing was longer compared with that of nonsuppurative BCG lymphadenitis. Because of the small number of children who developed suppuration, it is not possible for us to recommend conservative management for this form of BCG lymphadenitis. Needle aspiration may be a better option in such cases. 18, 19

Although expectant treatment for nonsuppurative BCG lymphadenitis has been recommended previously, it was borne out of failure of medical therapy in these studies. 10–14 Our study has provided more convincing evidence in support of conservative treatment of such patients. BCG lymphadenitis therefore appears to be a harmless condition. Most of the cases regress spontaneously on expectant follow-up. Progression to suppuration and drainage is uncommon, and healing occurs even in these cases without any surgical intervention.

1. Hsing CT. Local complications of BCG vaccination in preschool children and newborn babies. Bull WHO 1954; 11: 1023–9.
2. Ustvedt HJ. Local reactions in BCG vaccination. Bull WHO 1950; 2: 441–68.
3. Lotte A, Wasz-Hockert, Poisson N, et al. Second IUATLD study on complications induced by intradermal BCG-vaccination. Bull IUATLD 1988; 63: 47–59.
4. Praveen NK, Smikle MF, Prabhakar P, Pande D, Johnson B, Ashley D. Outbreak of Bacillus Calmette-Guérin-associated lymphadenitis and abscesses in Jamaican Children. Pediatr Infect Dis J 1990; 19: 890–3.
5. Milstien JB, Gibson JJ. Quality control of BCG vaccine by WHO: a review of factors that may influence vaccine effectiveness and safety. Bull WHO 1990; 68: 93–108.
6. Lugosi L. Theoretical and methodological aspects of BCG vaccine from the discovery of Calmette and Guérin to molecular biology: a review. Tuberc Lung Dis 1992; 73: 252–61.
7. Chaves-Carbello E, Sanchez GA. Regional lymphadenitis following BCG vaccination. Clin Pediatr 1972; 11: 693–7.
8. Victoria MS, Shah BR. Bacillus Calmette-Guérin lymphadenitis: a case report and review of the literature. Pediatr Infect Dis J 1985; 4: 295–6.
9. Consensus statement of IAP Working Group. Treatment of childhood tuberculosis. Indian Pediatr 1997; 34: 1093–6.
10. Close GC, Nasiiro R. Management of BCG adenitis in infancy. J Trop Pediatr 1985; 31: 286.
11. Caglayan S, Yegin O, Kayran K, Timocin N, Kasirga E, Gun M. Is medical therapy effective for regional lymphadenitis following BCG vaccination? Am J Dis Child 1987; 141: 1213–4.
12. Noah PK, Pande D, Johnson B, Ashley D. Evaluation of oral erythromycin and local isoniazid instillation therapy in infants with Bacillus Calmette-Guérin lymphadenitis and abscesses. Pediatr Infect Dis J 1993; 12: 136–9.
13. Oguz F, Sidal M, Alper G, Alev F, Neyzi O. Treatment of Bacillus Calmette-Guérin-associated lymphadenitis. Pediatr Infect Dis J 1992; 10: 887–8.
14. Kuyucu N, Kuyucu S, Öcal B, Teziç; T Comparison of oral erythromycin, local administration of streptomycin and placebo therapy for non-suppurative Bacillus Calmette-Guérin lymphadenitis. Pediatr Infect Dis J 1998; 17: 524–5.
15. Goraya JS, Virdi VJS. Treatment of BCG-lymphadenitis: a metaanalysis. Pediatr Infect Dis J 2001; 20: 632–4.
16. Helmick CG, D’souza AJ, Goddard RN. An outbreak of severe BCG axillary lymphadenitis in Saint Lucia, 1982–1983. West Indies Med J 1986; 35: 12–17.
17. Baki A, Öncü; M, Usta S, Yildiz K, Karagüzel A. Therapy of regional lymphadenitis following BCG vaccination. Infection 1991; 19: 414–6.
18. Cagalayn S, Arikan A, Yaprak I, Aksoz K, Kansoy S. Management of suppuration in regional lymph nodes secondary to BCG vaccination. Acta Paediatr Jpn 1991; 33: 699–702.
19. Banani SA, Alborzi A. Needle aspiration for suppurative post-BCG adenitis. Arch Dis Child 1994; 71: 446–7.
20. Hengster P, Solder B, Fille M, Menardi G. Surgical treatment of Bacillus Calmette-Guérin lymphadenitis. World J Surg 1997; 21: 520–3.
© 2002 Lippincott Williams & Wilkins, Inc.