INT J TUBERC LUNG DIS 18(9):1062–1065 Q 2014 The Union http://dx.doi.org/10.5588/ijtld.14.0160
Congenital tuberculosis and management of exposure in neonatal and pediatric intensive care units G. Grisaru-Soen,* M. Savyon,† E. Sadot,‡ V. Schechner,§ Y. Sivan,‡ D. Schwartz,¶ J. Tarabeia,§ Z. Amitai,† I. Yoabov,† Y. Carmeli‡ *Pediatric Infectious Disease Unit, Tel Aviv Sourasky Medical Center, Tel Aviv, †Tel Aviv Health District, Ministry of Health, Tel Aviv, ‡Department of Pediatric Intensive Care, Dana-Dwek Children’s Hospital, §Division of Epidemiology and Preventive Medicine, and ¶Microbiology Laboratory, Tel Aviv Sourasky Medical Center, Tel Aviv, affiliated to the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Faculty of Nursing, Yezreel Valley College, Yezreel Valley, Israel SUMMARY S E T T I N G : This report describes the management and outcome of neonatal intensive care unit (NICU) and paediatric ICU (PICU) exposure to a 26-day-old premature infant with congenital tuberculosis (TB). D E S I G N : The infant’s mother underwent chest X-ray (CXR) and sputum culture. Contacts of the infant were identified. Tuberculin skin tests (TSTs) were performed on 97 infants and children, 156 NICU and PICU visitors and 115 health care workers. R E S U LT S : The mother’s sputum culture was positive for Mycobacterium tuberculosis. No TST conversion occurred in the exposed NICU infants. All neonates received prophylactic isoniazid (INH). One exposed child in the PICU had TST conversion with normal CXR
and completed 9 months of INH without developing active disease; 22 (14%) PICU and NICU visitors and 3 NICU personnel had TST conversion without evidence of disease. C O N C L U S I O N S : The sequence of events described here demonstrates the difficulty in diagnosis and management of TB in this age group. Transmission of TB in NICU and PICUs is unusual but can occur, and calls for a systematic approach to investigation of the exposed infants, family members and health care providers. K E Y W O R D S : tuberculosis; contact investigation; infant
CONGENITAL TUBERCULOSIS (TB) is rare, with a mortality rate of as high as 50%. It can result from haematogenous dissemination of Mycobacterium tuberculosis after rupture of a placental tubercle into the foetal circulation or by ingestion of amniotic fluid or maternal blood at delivery.1–5 We describe an infant who was admitted with recurrent vomiting and respiratory failure, and the investigations performed in the neonatal (NICUs) and paediatric intensive care units (PICUs), demonstrating the need for a high index of suspicion for TB and the difficulty in diagnosis and management in this age group.
NICU for 18 days and her course was uneventful. She was re-admitted at 26 days with a 5-day history of vomiting after every feed. A full sepsis workup was performed and treatment with cefotaxime and ampicillin was started. During her first day of admission to the paediatric ward the infant had recurrent apnoea, tachypnoea and desaturation. Her chest X-rays (CXR) showed multiple infiltrates in both lungs, and she was transferred to the PICU, where she was mechanically ventilated. She underwent an extensive diagnostic workup that included blood, urine, cerebrospinal fluid and sputum cultures, human immunodeficiency virus serology, urine Legionella pneumophila Ag x2, polymerase chain reaction (PCR) for Bordetella pertussis, PCR for enterovirus in stool, cytomegalovirus (CMV) culture in urine, and a respiratory viruses panel from a nasopharyngeal swab (influenza, respiratory syncytial virus, adenovirus, parainfluenza, human meta pneumovirus). All results were negative. Bronchoalveolar lavage (BAL) specimens
METHODS Case review and description In May 2012, a female infant was born at 33 weeks’ gestation to an ostensibly healthy mother. Delivery was spontaneous and vaginal and the birth weight was 2 kg. The mother had arrived in Israel from Eritrea 7 months previously. The infant stayed in our
Correspondence to: Galia Grisaru-Soen, Dana-Dwek Children’s Hospital, Tel Aviv Sourasky Medical Center, 6 Weizman St, Tel Aviv, Israel 6423906. e-mail: [email protected]. Article submitted 24 February 2014. Final version accepted 9 May 2014.
Congenital TB exposure in NICU/PICU
1063
Figure 1 Sputum gram stain on day 19. This image can be viewed online in color at http://www. ingentaconnect.com/content/iuatld/ijtld2014/00000018/00000009/art00009.
on day 12 included direct smear, culture, Pneumocystis carinii pneumonia (PCP) Ag immunofluorescence, fungal direct smear and culture, Ziehl-Neelsen (ZN; acid-fast) stain, Xpertw (Cepheid, Inc, Sunnyvale, CA, USA) for TB, PCR for 16S, fungus, mycobacteria, herpes simplex virus PCR and cultures. All were negative. Due to a distended abdomen, an abdominal ultrasound was performed on day 15 and demonstrated a large amount of fluid. A yellowish fluid was drained: gram stain and culture, ZN and Xpert for TB from the fluid were all negative. On day 19, sputum sent for gram stain and culture to the microbiology laboratory showed suspicious bacteria (Figure 1). The acid-fast stain of the same sputum showed numerous acid-fast bacilli (AFB) (Figure 2). The mycobacterium PCR from sputum (Xpert) was positive for M. tuberculosis. The CSF acid-fast stain was negative, as was PCR for M. tuberculosis.
The patient was put in airborne isolation. Antimycobacterial treatment was initiated with INH, rifampicin, ethambutol, pyrazinamide and pyridoxine. Due to the pancytopaenia, a bone marrow biopsy was performed on day 20 and the results showed haemophagocytosis, consistent with a diagnosis of haemophagocytic lymphohistiocytosis (HLH). Further investigation revealed a high ferritin level (1501 ng/ml), a low fibrinogen level (114 mg/dl), no natural killer (NK) activity, a high soluble interleukin-2 receptor level and a high triglyceride level (876 mg/ dl). Treatment for HLH was started with steroids and anti-CD52 (Campathw Genzyme Corporate, Waltham, MA) for 1 month, and there has been no recurrence since then. Cultures from the BAL and peritoneal fluid were positive for M. tuberculosis. The last positive sputum culture was on day 45 of treatment. She was discharged in April 2013 (10
Figure 2 Acid-fast stain of the same sputum as in Figure 1. This image can be viewed online in color at http://www.ingentaconnect.com/content/iuatld/ijtld2014/00000018/00000009/ art00009.
1064
The International Journal of Tuberculosis and Lung Disease
Table 1 TST results among exposed infants and visitors in the NICU and PICU Contacts identified
TST 1
66 43
61/61 negative 36/36 negative 84/100 negative 16/100 positive 40/56 negative 16/56 positive
NICU infants PICU infants and children NICU visitors PICU visitors
TST 2 51/51 23/24 50/58 19/22
negative negative 1/24 positive negative 8/58 positive negative 3/22 positive
TST ¼ tuberculin skin test; NICU ¼ neonatal intensive care unit; PICU ¼ paediatric intensive care unit.
months after admission) and completed 12 months of therapy for TB. Due to chronic lung injury, she still has a tracheostomy. Transmission route and contact identification The route of TB transmission to the infant was not known, nor was the degree or duration of possible transmission to others. A multidisciplinary team was urgently assembled to investigate and prevent further transmission. The patient’s mother was asymptomatic during her child’s hospitalization, as was the father. Both parents underwent CXR, and the mother was referred for further investigation. Health care workers, family members and visitors to the NICU and PICU and exposed infants and children were evaluated as possible contacts of the infected neonate and her mother. Those who had spent at least 24 cumulative hours in the same room in the NICU or PICU during her hospitalisation were assessed. The infants and children underwent physical examinations and tuberculin skin tests (TSTs) at the Dana Children’s Hospital. All exposed infants aged ,3 months (corrected age of 3 months) were offered treatment with INH at a dosage of 10 mg/kg per day and pyridoxine at a dosage of 2 mg/day. Children aged .3 months with primary or secondary immune deficiency were also offered the same treatment. Children aged between 3 months and 5 years of age with no immune deficiency were offered the same treatment only if they had been exposed for more than 3 cumulative days. Children aged .5 years were only screened initially with TST; a second TST was performed at 3 months from the last day of exposure or 3 months after the expected date of delivery (whichever came later). Treatment was continued until the final evaluation and was terminated if the TST results were negative at that time. As this study reports on a required public health investigation, ethics approval was not required. Table 2
RESULTS The patient’s mother showed pleural effusion on CXR, and pulmonary TB was suspected. Her sputum was negative in acid-fast stain, as was PCR for M. tuberculosis, but culture was later positive for M. tuberculosis. The father had a normal CXR. The results of the TSTs of the contacts in the NICU and PICU are shown in Table 1. There had been no TST conversion among the exposed infants in the NICU. There was only one TST conversion among the exposed children in the PICU: a 2-year-old boy who had been hospitalised for a brain tumour and was exposed to the patient for 24 hours. His CXR was normal, and he completed 9 months of INH without developing active disease. A total of 11 visitors had TST conversion (these comprised 14% of the total number of visitors to the NICU and the PICU). All had normal CXRs. Of these 11, eight were born and lived in Israel, two were new immigrants from Russia and one was a new arrival from Africa. The TST results among exposed personnel in the NICU (n ¼ 82) and the PICU (n ¼ 33) are presented in Table 2. Altogether, 7% of all the health care workers (3/41) had TST conversion.
DISCUSSION Our case demonstrates the difficulty in diagnosing TB during the neonatal period. TSTs are not reliable for infants aged ,6 weeks because of their immature immune systems.6 However, infants are more likely to have high bacillary loads than older children and produce more positive AFB smears and cultures, as their infections are widely disseminated and rapidly progressive.7 In our patient the first acid-fast smears from BAL and peritoneal fluid were negative, as were the Xpert PCRs from these specimens. A few days later, a smear from a tracheal aspirate was positive,
Tuberculin skin test results among exposed personnel in the NICU and PICU Identified contacts
Pre-exposure PPD
Post-exposure PPD
Treatment provided for LTBI
NICU personnel
82
3/24 conversion 0/6 positive
3/3
PICU personnel
33
38/82 negative 32/82 positive 12/82 NA 15/33 negative 14/33 positive 4/33 NA
0/11 conversion
NICU ¼ neonatal intensive care unit; PICU ¼ paediatric intensive care unit; PPD ¼ purified protein derivative; LTBI ¼ latent tuberculous infection; NA ¼ not available.
Congenital TB exposure in NICU/PICU
with a high mycobacterial load, and the cultures from the first specimens turned positive as well. This emphasises the difficulties in diagnosing TB in this age group. Notably, the mothers of these infants may be asymptomatic:7 our patient’s mother was asymptomatic and, although her CXR was abnormal, her sputum was negative on acid-fast smear. Brastianos et al.’s search of the English language literature revealed 37 other cases of M. tuberculosisassociated haemophagocytic syndrome.8 The reported mortality was approximately 50%. TB should therefore be considered early in the differential diagnosis of infectious aetiologies associated with haemophagocytic syndrome. Given the rapid disease progression and its association with high mortality, early diagnosis and timely implementation of antituberculosis medications are critical.8 Several reports have described the exposure of infants to TB in NICUs. The index cases with active TB could be the infants, the parents or the health care workers. The theoretical risk of transmission by health care workers with TB may be higher because they have close contact with the infants.5,6,9,10 Although infants with TB are unlikely to be infectious, prolonged exposure and therapeutic measures, such as respiratory suctioning, may increase the risk of transmission to health care workers.5,6,10,11 A review of the English literature revealed 13 documented occurrences of transmission of TB from children aged 610 years.10,11 Five cases occurred in children aged 61 year, four of whom were shown to have congenital TB. Factors that increased the risk of transmission included severe prematurity and mechanical interventions such as frequent suctioning, chest physiotherapy and/or mechanical ventilation.6 Our patient was ventilated most of the time during her PICU stay and had frequent suctioning. Fortunately, there was no documented TST conversion among our PICU personnel. Three of our NICU personnel had TST conversion, but we cannot determine whether the conversion was due to exposure to the infant and/or to the mother or to other reasons, as their baseline TSTs (all negative) had been performed long before the current exposure (2005, 2007 and 2011), leaving the possibility that conversion had already happened earlier. Infant-to-infant transmission of TB in a NICU was reported in only one case in the literature,5 and this was attributed to contaminated equipment rather than aerosolisation. In our study, TST conversion occurred in one child who was exposed to our patient in the PICU. That child’s second TST was read abroad
1065
in another hospital. After almost 2 years of follow-up, none of the exposed infants and children has developed active TB.
CONCLUSION Our case highlights how TB in the setting of a NICU and a PICU can lead to a huge epidemiological problem that demands a complex contact investigation and follow-up, with intensive communication between the medical centre and health departments. Similar to other reports, we found the risk of transmission to be minimal but not negligible, thus supporting the vital necessity for a systematic approach to investigate exposed infants, their family members and health care workers. Acknowledgements The authors are grateful to E Eshkol for editorial assistance. Conflict of interest: none declared.
References 1 Mazade M A, Evans E M, Starke J R, Correa A G. Congenital tuberculosis presenting as sepsis syndrome: case report and review of the literature. Pediatr Infect Dis J 2001; 20: 439–442. 2 Peng W, Yang J, Liu E. Analysis of 170 cases of congenital TB reported in the literature between 1946 and 2009. Pediatr Pulmonol 2011; 46: 1215–1224. 3 Cantwell M F, Shehab Z M, Costello A M, et al. Brief report: congenital tuberculosis. N Engl J Med 1994; 330: 1051–1054. 4 Beitzke H. Uber die angeborene tuberkulose infektion. Ergeb Ges Tuberk Forsch 1935; 7: 1–30. [German] 5 Crockett M, King S M, Kitai I, et al. Outbreak Investigation Team. Nosocomial transmission of congenital tuberculosis in a neonatal intensive care unit. Clin Infect Dis 2004; 39: 1179– 1123. 6 Lee L H, LeVea C M, Graman P S. Congenital tuberculosis in a neonatal intensive care unit: case report, epidemiological investigation, and management of exposures. Clin Infect Dis 1998; 27: 474–477. 7 Smith K C. Congenital tuberculosis: a rare manifestation of a common infection. Curr Opin Infect Dis 2002; 15: 269–274. 8 Brastianos P K, Swanson J W, Torbenson M, Sperati J, Karakousis P C. Tuberculosis-associated haemophagocytic syndrome. Lancet Infect Dis 2006; 6: 447–454. 9 Laartz B W, Narvarte H J, Holt D, Larkin J A, Pomputius W F. Congenital tuberculosis and management of exposures in a neonatal intensive care unit. Infect Control Host Epidemiol 2002; 23: 573–579. 10 Winters A, Agerton T B, Driver C R, et al. Congenital tuberculosis and management of exposure in three neonatal intensive care units. Int J Tuberc Lung Dis 2010; 14: 1641– 1643. 11 Reynolds D L, Gillis F, Kitai I, et al. Transmission of Mycobacterium tuberculosis from an infant. Int J Tuberc Lung Dis 2006; 10: 1051–1056.
Congenital TB exposure in NICU/PICU
i
RESUME : Ce rapport d´ecrit la prise en charge et le devenir d’un b´eb´e pr´ematur´e ag´ ˆ e de 26 jours et atteint de tuberculose (TB) cong´enitale hospitalis´e en unit´e de soins intensifs n´eonatale (NICU) et p´ediatrique (PICU). S C H E´ M A : La m`ere du b´eb´e a eu une radio pulmonaire et une culture de crachats. Les contacts du b´eb´e ont e´ t´e identifi´es. Des tests cutan´es a` la tuberculine (TST) ont e´ t´e r´ealis´es sur 97 b´eb´es et enfants et 156 visiteurs des services de NICU et PICU ; 115 agents de sant´e ont e´ galement b´en´efici´e d’un d´epistage. R E´ S U LT A T S : La culture de crachats de la m`ere s’est re´ ve´ l´ee positive pour Mycobacterium tuberculosis. Aucune conversion de TST n’est survenue chez les b´eb´es expos´es du service de NICU. Tous les nouveau-n´es ont re¸cu un traitement pr´eventif par isoniazide (INH). CONTEXTE
Un enfant expos´e en service de PICU a eu une conversion du TST avec une radio pulmonaire normale ; il a re¸cu 9 mois d’INH sans d´evelopper de TB active, et 22 visiteurs des services de PICU et NCIU (14%) ainsi que trois membres du personnel de NICU ont eu une conversion du TST sans e´ vidence de maladie. C O N C L U S I O N S : La s´equence des e´ v´enements d´ecrits ici d´emontre la difficult´e du diagnostic et de la prise en charge de la TB dans cette tranche d’age. ˆ La transmission de la TB dans les services de NICU et de PICU est inhabituelle, mais peut survenir et requiert une approche syst´ematique de l’investigation des b´eb´es expos´es, des membres de leurs familles et du personnel soignant.
RESUMEN
En el presente art´ıculo se describe el manejo de un reci´en nacido prematuro que presentaba tuberculosis (TB) cong´enita y su desenlace cl´ınico tras el tratamiento durante 26 d´ıas en las unidades de cuidado intensivo neonatal (NICU) y pedia´trico (PICU). M E´ T O D O: Se practico ´ a la madre del reci´en nacido la radiograf´ıa de torax ´ y el cultivo de esputo. Se detectaron los contactos del reci´en nacido y se practico´ la prueba cuta´nea de la tuberculina (TST) a 97 lactantes y ninos ˜ y a 156 visitantes de las NICU y PICU. Se investigaron adema´s 115 profesionales de salud. R E S U LT A D O S: El resultado del cultivo de esputo de la madre fue positivo para Mycobacterium tuberculosis. No se observaron conversiones de la TST en los lactantes expuestos en la NICU. Todos los recie´ n nacidos M A R C O D E R E F E R E N C I A:
recibieron tratamiento profila´ ctico con isoniazida (INH). Un nino ˜ expuesto en la PICU que presento´ conversion ´ TST y una radiograf´ıa de torax ´ normal completo´ 9 meses de tratamiento con INH y no contrajo la enfermedad tuberculosa activa. Se observo´ una conversion ´ de la TST sin signos de enfermedad en 22 visitantes de las NICU y PICU (14%) y en tres miembros del personal de la NICU. C O N C L U S I O N E S: La secuencia de hechos que se describe en este art´ıculo demuestra la dificultad del diagnostico ´ y el tratamiento de la TB en este grupo de edad. La transmision ´ de la enfermedad en las NICU y PICU es infrecuente, pero posible, y exige un examen sistema´ tico de los ninos, los familiares y los ˜ profesionales de salud expuestos.
Congenital tuberculosis and management of exposure in neonatal and pediatric intensive care units G. Grisaru-Soen,* M. Savyon,† E. Sadot,‡ V. Schechner,§ Y. Sivan,‡ D. Schwartz,¶ J. Tarabeia,§ Z. Amitai,† I. Yoabov,† Y. Carmeli‡ *Pediatric Infectious Disease Unit, Tel Aviv Sourasky Medical Center, Tel Aviv, †Tel Aviv Health District, Ministry of Health, Tel Aviv, ‡Department of Pediatric Intensive Care, Dana-Dwek Children’s Hospital, §Division of Epidemiology and Preventive Medicine, and ¶Microbiology Laboratory, Tel Aviv Sourasky Medical Center, Tel Aviv, affiliated to the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Faculty of Nursing, Yezreel Valley College, Yezreel Valley, Israel SUMMARY S E T T I N G : This report describes the management and outcome of neonatal intensive care unit (NICU) and paediatric ICU (PICU) exposure to a 26-day-old premature infant with congenital tuberculosis (TB). D E S I G N : The infant’s mother underwent chest X-ray (CXR) and sputum culture. Contacts of the infant were identified. Tuberculin skin tests (TSTs) were performed on 97 infants and children, 156 NICU and PICU visitors and 115 health care workers. R E S U LT S : The mother’s sputum culture was positive for Mycobacterium tuberculosis. No TST conversion occurred in the exposed NICU infants. All neonates received prophylactic isoniazid (INH). One exposed child in the PICU had TST conversion with normal CXR
and completed 9 months of INH without developing active disease; 22 (14%) PICU and NICU visitors and 3 NICU personnel had TST conversion without evidence of disease. C O N C L U S I O N S : The sequence of events described here demonstrates the difficulty in diagnosis and management of TB in this age group. Transmission of TB in NICU and PICUs is unusual but can occur, and calls for a systematic approach to investigation of the exposed infants, family members and health care providers. K E Y W O R D S : tuberculosis; contact investigation; infant
CONGENITAL TUBERCULOSIS (TB) is rare, with a mortality rate of as high as 50%. It can result from haematogenous dissemination of Mycobacterium tuberculosis after rupture of a placental tubercle into the foetal circulation or by ingestion of amniotic fluid or maternal blood at delivery.1–5 We describe an infant who was admitted with recurrent vomiting and respiratory failure, and the investigations performed in the neonatal (NICUs) and paediatric intensive care units (PICUs), demonstrating the need for a high index of suspicion for TB and the difficulty in diagnosis and management in this age group.
NICU for 18 days and her course was uneventful. She was re-admitted at 26 days with a 5-day history of vomiting after every feed. A full sepsis workup was performed and treatment with cefotaxime and ampicillin was started. During her first day of admission to the paediatric ward the infant had recurrent apnoea, tachypnoea and desaturation. Her chest X-rays (CXR) showed multiple infiltrates in both lungs, and she was transferred to the PICU, where she was mechanically ventilated. She underwent an extensive diagnostic workup that included blood, urine, cerebrospinal fluid and sputum cultures, human immunodeficiency virus serology, urine Legionella pneumophila Ag x2, polymerase chain reaction (PCR) for Bordetella pertussis, PCR for enterovirus in stool, cytomegalovirus (CMV) culture in urine, and a respiratory viruses panel from a nasopharyngeal swab (influenza, respiratory syncytial virus, adenovirus, parainfluenza, human meta pneumovirus). All results were negative. Bronchoalveolar lavage (BAL) specimens
METHODS Case review and description In May 2012, a female infant was born at 33 weeks’ gestation to an ostensibly healthy mother. Delivery was spontaneous and vaginal and the birth weight was 2 kg. The mother had arrived in Israel from Eritrea 7 months previously. The infant stayed in our
Correspondence to: Galia Grisaru-Soen, Dana-Dwek Children’s Hospital, Tel Aviv Sourasky Medical Center, 6 Weizman St, Tel Aviv, Israel 6423906. e-mail: [email protected]. Article submitted 24 February 2014. Final version accepted 9 May 2014.
Congenital TB exposure in NICU/PICU
1063
Figure 1 Sputum gram stain on day 19. This image can be viewed online in color at http://www. ingentaconnect.com/content/iuatld/ijtld2014/00000018/00000009/art00009.
on day 12 included direct smear, culture, Pneumocystis carinii pneumonia (PCP) Ag immunofluorescence, fungal direct smear and culture, Ziehl-Neelsen (ZN; acid-fast) stain, Xpertw (Cepheid, Inc, Sunnyvale, CA, USA) for TB, PCR for 16S, fungus, mycobacteria, herpes simplex virus PCR and cultures. All were negative. Due to a distended abdomen, an abdominal ultrasound was performed on day 15 and demonstrated a large amount of fluid. A yellowish fluid was drained: gram stain and culture, ZN and Xpert for TB from the fluid were all negative. On day 19, sputum sent for gram stain and culture to the microbiology laboratory showed suspicious bacteria (Figure 1). The acid-fast stain of the same sputum showed numerous acid-fast bacilli (AFB) (Figure 2). The mycobacterium PCR from sputum (Xpert) was positive for M. tuberculosis. The CSF acid-fast stain was negative, as was PCR for M. tuberculosis.
The patient was put in airborne isolation. Antimycobacterial treatment was initiated with INH, rifampicin, ethambutol, pyrazinamide and pyridoxine. Due to the pancytopaenia, a bone marrow biopsy was performed on day 20 and the results showed haemophagocytosis, consistent with a diagnosis of haemophagocytic lymphohistiocytosis (HLH). Further investigation revealed a high ferritin level (1501 ng/ml), a low fibrinogen level (114 mg/dl), no natural killer (NK) activity, a high soluble interleukin-2 receptor level and a high triglyceride level (876 mg/ dl). Treatment for HLH was started with steroids and anti-CD52 (Campathw Genzyme Corporate, Waltham, MA) for 1 month, and there has been no recurrence since then. Cultures from the BAL and peritoneal fluid were positive for M. tuberculosis. The last positive sputum culture was on day 45 of treatment. She was discharged in April 2013 (10
Figure 2 Acid-fast stain of the same sputum as in Figure 1. This image can be viewed online in color at http://www.ingentaconnect.com/content/iuatld/ijtld2014/00000018/00000009/ art00009.
1064
The International Journal of Tuberculosis and Lung Disease
Table 1 TST results among exposed infants and visitors in the NICU and PICU Contacts identified
TST 1
66 43
61/61 negative 36/36 negative 84/100 negative 16/100 positive 40/56 negative 16/56 positive
NICU infants PICU infants and children NICU visitors PICU visitors
TST 2 51/51 23/24 50/58 19/22
negative negative 1/24 positive negative 8/58 positive negative 3/22 positive
TST ¼ tuberculin skin test; NICU ¼ neonatal intensive care unit; PICU ¼ paediatric intensive care unit.
months after admission) and completed 12 months of therapy for TB. Due to chronic lung injury, she still has a tracheostomy. Transmission route and contact identification The route of TB transmission to the infant was not known, nor was the degree or duration of possible transmission to others. A multidisciplinary team was urgently assembled to investigate and prevent further transmission. The patient’s mother was asymptomatic during her child’s hospitalization, as was the father. Both parents underwent CXR, and the mother was referred for further investigation. Health care workers, family members and visitors to the NICU and PICU and exposed infants and children were evaluated as possible contacts of the infected neonate and her mother. Those who had spent at least 24 cumulative hours in the same room in the NICU or PICU during her hospitalisation were assessed. The infants and children underwent physical examinations and tuberculin skin tests (TSTs) at the Dana Children’s Hospital. All exposed infants aged ,3 months (corrected age of 3 months) were offered treatment with INH at a dosage of 10 mg/kg per day and pyridoxine at a dosage of 2 mg/day. Children aged .3 months with primary or secondary immune deficiency were also offered the same treatment. Children aged between 3 months and 5 years of age with no immune deficiency were offered the same treatment only if they had been exposed for more than 3 cumulative days. Children aged .5 years were only screened initially with TST; a second TST was performed at 3 months from the last day of exposure or 3 months after the expected date of delivery (whichever came later). Treatment was continued until the final evaluation and was terminated if the TST results were negative at that time. As this study reports on a required public health investigation, ethics approval was not required. Table 2
RESULTS The patient’s mother showed pleural effusion on CXR, and pulmonary TB was suspected. Her sputum was negative in acid-fast stain, as was PCR for M. tuberculosis, but culture was later positive for M. tuberculosis. The father had a normal CXR. The results of the TSTs of the contacts in the NICU and PICU are shown in Table 1. There had been no TST conversion among the exposed infants in the NICU. There was only one TST conversion among the exposed children in the PICU: a 2-year-old boy who had been hospitalised for a brain tumour and was exposed to the patient for 24 hours. His CXR was normal, and he completed 9 months of INH without developing active disease. A total of 11 visitors had TST conversion (these comprised 14% of the total number of visitors to the NICU and the PICU). All had normal CXRs. Of these 11, eight were born and lived in Israel, two were new immigrants from Russia and one was a new arrival from Africa. The TST results among exposed personnel in the NICU (n ¼ 82) and the PICU (n ¼ 33) are presented in Table 2. Altogether, 7% of all the health care workers (3/41) had TST conversion.
DISCUSSION Our case demonstrates the difficulty in diagnosing TB during the neonatal period. TSTs are not reliable for infants aged ,6 weeks because of their immature immune systems.6 However, infants are more likely to have high bacillary loads than older children and produce more positive AFB smears and cultures, as their infections are widely disseminated and rapidly progressive.7 In our patient the first acid-fast smears from BAL and peritoneal fluid were negative, as were the Xpert PCRs from these specimens. A few days later, a smear from a tracheal aspirate was positive,
Tuberculin skin test results among exposed personnel in the NICU and PICU Identified contacts
Pre-exposure PPD
Post-exposure PPD
Treatment provided for LTBI
NICU personnel
82
3/24 conversion 0/6 positive
3/3
PICU personnel
33
38/82 negative 32/82 positive 12/82 NA 15/33 negative 14/33 positive 4/33 NA
0/11 conversion
NICU ¼ neonatal intensive care unit; PICU ¼ paediatric intensive care unit; PPD ¼ purified protein derivative; LTBI ¼ latent tuberculous infection; NA ¼ not available.
Congenital TB exposure in NICU/PICU
with a high mycobacterial load, and the cultures from the first specimens turned positive as well. This emphasises the difficulties in diagnosing TB in this age group. Notably, the mothers of these infants may be asymptomatic:7 our patient’s mother was asymptomatic and, although her CXR was abnormal, her sputum was negative on acid-fast smear. Brastianos et al.’s search of the English language literature revealed 37 other cases of M. tuberculosisassociated haemophagocytic syndrome.8 The reported mortality was approximately 50%. TB should therefore be considered early in the differential diagnosis of infectious aetiologies associated with haemophagocytic syndrome. Given the rapid disease progression and its association with high mortality, early diagnosis and timely implementation of antituberculosis medications are critical.8 Several reports have described the exposure of infants to TB in NICUs. The index cases with active TB could be the infants, the parents or the health care workers. The theoretical risk of transmission by health care workers with TB may be higher because they have close contact with the infants.5,6,9,10 Although infants with TB are unlikely to be infectious, prolonged exposure and therapeutic measures, such as respiratory suctioning, may increase the risk of transmission to health care workers.5,6,10,11 A review of the English literature revealed 13 documented occurrences of transmission of TB from children aged 610 years.10,11 Five cases occurred in children aged 61 year, four of whom were shown to have congenital TB. Factors that increased the risk of transmission included severe prematurity and mechanical interventions such as frequent suctioning, chest physiotherapy and/or mechanical ventilation.6 Our patient was ventilated most of the time during her PICU stay and had frequent suctioning. Fortunately, there was no documented TST conversion among our PICU personnel. Three of our NICU personnel had TST conversion, but we cannot determine whether the conversion was due to exposure to the infant and/or to the mother or to other reasons, as their baseline TSTs (all negative) had been performed long before the current exposure (2005, 2007 and 2011), leaving the possibility that conversion had already happened earlier. Infant-to-infant transmission of TB in a NICU was reported in only one case in the literature,5 and this was attributed to contaminated equipment rather than aerosolisation. In our study, TST conversion occurred in one child who was exposed to our patient in the PICU. That child’s second TST was read abroad
1065
in another hospital. After almost 2 years of follow-up, none of the exposed infants and children has developed active TB.
CONCLUSION Our case highlights how TB in the setting of a NICU and a PICU can lead to a huge epidemiological problem that demands a complex contact investigation and follow-up, with intensive communication between the medical centre and health departments. Similar to other reports, we found the risk of transmission to be minimal but not negligible, thus supporting the vital necessity for a systematic approach to investigate exposed infants, their family members and health care workers. Acknowledgements The authors are grateful to E Eshkol for editorial assistance. Conflict of interest: none declared.
References 1 Mazade M A, Evans E M, Starke J R, Correa A G. Congenital tuberculosis presenting as sepsis syndrome: case report and review of the literature. Pediatr Infect Dis J 2001; 20: 439–442. 2 Peng W, Yang J, Liu E. Analysis of 170 cases of congenital TB reported in the literature between 1946 and 2009. Pediatr Pulmonol 2011; 46: 1215–1224. 3 Cantwell M F, Shehab Z M, Costello A M, et al. Brief report: congenital tuberculosis. N Engl J Med 1994; 330: 1051–1054. 4 Beitzke H. Uber die angeborene tuberkulose infektion. Ergeb Ges Tuberk Forsch 1935; 7: 1–30. [German] 5 Crockett M, King S M, Kitai I, et al. Outbreak Investigation Team. Nosocomial transmission of congenital tuberculosis in a neonatal intensive care unit. Clin Infect Dis 2004; 39: 1179– 1123. 6 Lee L H, LeVea C M, Graman P S. Congenital tuberculosis in a neonatal intensive care unit: case report, epidemiological investigation, and management of exposures. Clin Infect Dis 1998; 27: 474–477. 7 Smith K C. Congenital tuberculosis: a rare manifestation of a common infection. Curr Opin Infect Dis 2002; 15: 269–274. 8 Brastianos P K, Swanson J W, Torbenson M, Sperati J, Karakousis P C. Tuberculosis-associated haemophagocytic syndrome. Lancet Infect Dis 2006; 6: 447–454. 9 Laartz B W, Narvarte H J, Holt D, Larkin J A, Pomputius W F. Congenital tuberculosis and management of exposures in a neonatal intensive care unit. Infect Control Host Epidemiol 2002; 23: 573–579. 10 Winters A, Agerton T B, Driver C R, et al. Congenital tuberculosis and management of exposure in three neonatal intensive care units. Int J Tuberc Lung Dis 2010; 14: 1641– 1643. 11 Reynolds D L, Gillis F, Kitai I, et al. Transmission of Mycobacterium tuberculosis from an infant. Int J Tuberc Lung Dis 2006; 10: 1051–1056.
Congenital TB exposure in NICU/PICU
i
RESUME : Ce rapport d´ecrit la prise en charge et le devenir d’un b´eb´e pr´ematur´e ag´ ˆ e de 26 jours et atteint de tuberculose (TB) cong´enitale hospitalis´e en unit´e de soins intensifs n´eonatale (NICU) et p´ediatrique (PICU). S C H E´ M A : La m`ere du b´eb´e a eu une radio pulmonaire et une culture de crachats. Les contacts du b´eb´e ont e´ t´e identifi´es. Des tests cutan´es a` la tuberculine (TST) ont e´ t´e r´ealis´es sur 97 b´eb´es et enfants et 156 visiteurs des services de NICU et PICU ; 115 agents de sant´e ont e´ galement b´en´efici´e d’un d´epistage. R E´ S U LT A T S : La culture de crachats de la m`ere s’est re´ ve´ l´ee positive pour Mycobacterium tuberculosis. Aucune conversion de TST n’est survenue chez les b´eb´es expos´es du service de NICU. Tous les nouveau-n´es ont re¸cu un traitement pr´eventif par isoniazide (INH). CONTEXTE
Un enfant expos´e en service de PICU a eu une conversion du TST avec une radio pulmonaire normale ; il a re¸cu 9 mois d’INH sans d´evelopper de TB active, et 22 visiteurs des services de PICU et NCIU (14%) ainsi que trois membres du personnel de NICU ont eu une conversion du TST sans e´ vidence de maladie. C O N C L U S I O N S : La s´equence des e´ v´enements d´ecrits ici d´emontre la difficult´e du diagnostic et de la prise en charge de la TB dans cette tranche d’age. ˆ La transmission de la TB dans les services de NICU et de PICU est inhabituelle, mais peut survenir et requiert une approche syst´ematique de l’investigation des b´eb´es expos´es, des membres de leurs familles et du personnel soignant.
RESUMEN
En el presente art´ıculo se describe el manejo de un reci´en nacido prematuro que presentaba tuberculosis (TB) cong´enita y su desenlace cl´ınico tras el tratamiento durante 26 d´ıas en las unidades de cuidado intensivo neonatal (NICU) y pedia´trico (PICU). M E´ T O D O: Se practico ´ a la madre del reci´en nacido la radiograf´ıa de torax ´ y el cultivo de esputo. Se detectaron los contactos del reci´en nacido y se practico´ la prueba cuta´nea de la tuberculina (TST) a 97 lactantes y ninos ˜ y a 156 visitantes de las NICU y PICU. Se investigaron adema´s 115 profesionales de salud. R E S U LT A D O S: El resultado del cultivo de esputo de la madre fue positivo para Mycobacterium tuberculosis. No se observaron conversiones de la TST en los lactantes expuestos en la NICU. Todos los recie´ n nacidos M A R C O D E R E F E R E N C I A:
recibieron tratamiento profila´ ctico con isoniazida (INH). Un nino ˜ expuesto en la PICU que presento´ conversion ´ TST y una radiograf´ıa de torax ´ normal completo´ 9 meses de tratamiento con INH y no contrajo la enfermedad tuberculosa activa. Se observo´ una conversion ´ de la TST sin signos de enfermedad en 22 visitantes de las NICU y PICU (14%) y en tres miembros del personal de la NICU. C O N C L U S I O N E S: La secuencia de hechos que se describe en este art´ıculo demuestra la dificultad del diagnostico ´ y el tratamiento de la TB en este grupo de edad. La transmision ´ de la enfermedad en las NICU y PICU es infrecuente, pero posible, y exige un examen sistema´ tico de los ninos, los familiares y los ˜ profesionales de salud expuestos.
