PEDIATRICS Vol. 121 No. 5 May 2008, pp. e1434-e1437 (doi:10.1542/peds.2007-1750)
EXPERIENCE & REASON |
Tuberculosis Mimicking Ileocecal Intussusception in a 5-Month-Old Girl
a Department of Medical Microbiology and Infectious Diseases, Erasmus MC, University Medical Centre, Rotterdam, Netherlands
b Department of Paediatric Infectious Diseases, Erasmus MC, University Medical Centre, Rotterdam, Netherlands
c Subdivision Tuberculosis Control, Municipal Health Service, Rotterdam, Netherlands
d Department of Urology, Albert Schweitzer Hospital, Dordrecht, Netherlands
ABSTRACT
A 5-month-old girl was diagnosed with tuberculosis, mimicking ileocecal intussusception. The mother of the patient was later diagnosed with renal tuberculosis attributable to the same (unique) Mycobacterium tuberculosis strain. Possibly, that transmission occurred by aspiration or ingestion of infected amniotic fluid or urine, which could occur before or during birth. This case illustrates that tuberculosis can mimic other common diseases and, therefore, can be a difficult diagnosis to make. Because respiratory infection was very unlikely in this case, congenital tuberculosis or postnatal infection via infected urine or breast milk should be in the differential diagnosis. In this article, we focus on different (nonrespiratory) transmission routes of Mycobacterium tuberculosis and give a short review of the recent literature on congenital tuberculosis.
Key Words: tuberculosis • mycobacterium • diagnosis • urogenital tuberculosis • congenital tuberculosis
Abbreviations: MTB, Mycobacterium tuberculosis • TB, tuberculosis • CTB, congenital TB • CT, computed tomography • TST, tuberculin skin test • RFLP, restriction fragment length polymorphism
Worldwide, there are an estimated 2 billion Mycobacterium tuberculosis (MTB)-infected persons, of whom 8.8 million developed tuberculosis (TB) in 2005.1 This enormous amount of patients is in sharp contrast with the just-over-300 reported cases of congenital tuberculosis (CTB). In 1994, Cantwell et al2 revised the diagnostic criteria as set by Beitzke3 and set the "Cantwell criteria and the prevalence of symptoms" (Table 1).
|
The first symptoms of CTB typically occur within 3 months of birth. In most cases, onset of the disease is in the second half of the first month.4,5 The presenting signs and symptoms are often nonspecific, mimicking other more common diseases. Cantwell et al2 described the most prevalent symptoms (Table 1).
We describe a 5-month-old child suspected to have an ileocecal intussusception, who seemed to have miliary TB. This is an apparent example of the diverse and difficult ways TB can present. Because CTB should be in the differential diagnosis in such cases, we focused on the different (nonrespiratory) transmission routes of MTB and the literature on CTB.
CASE REPORT
A 5-month-old girl presented at a primary care hospital with a 5-day history of nonbloody diarrhea. The patient had started vomiting 2 days before admission and had subsequently developed a distended abdomen. She had developed a fever 1 day before admission.
The patient was the first child of an Indonesian mother and Dutch father and was born in the Netherlands. The mother became pregnant spontaneously after several failed attempts of in vitro fertilization. Pregnancy and delivery were uneventful. Mother was HIV-negative, hepatitis B surface antigen-negative, and syphilisnegative.
On physical examination, she seemed moderately ill, with a temperature of 37.5°C, pulse rate of 150 beats per minute, and respiratory rate of 30 beats per minute. She had a distended abdomen with sparse bowel sounds. The liver was slightly enlarged without splenomegaly. A nontender mass was palpable in the right lower abdominal quadrant. Initial laboratory examination showed a hemoglobin level of 6.9 mmol/L, a leukocyte level or 22.1 x 109/L, with 43% granulocytes and 19% bandforms, and a C-reactive protein level of 286 mg/L. The abdominal radiograph was normal, but the ultrasound showed a crescent-in-doughnut sign or "multiple-concentric-ring-sign," which was compatible with the initial diagnosis of ileocecal intussusception. There were no signs of appendicitis or abdominal abscesses. In an attempt to correct the intussusception with hydrostatic pressure, a barium enema was performed, which revealed a complete obstruction at the ileocecal level. After the procedure, the patient developed a more distended abdomen, with deterioration of the clinical condition, metabolic acidosis, and hyponatremia. She was subsequently referred to the intensive care unit of our tertiary care hospital.
Because of suspected intestinal ischemia or bowel perforation, a laparotomy was performed. This procedure showed an inflammatory process in the ileocecal region with multiple adhesions, but indications for an intussusception or enlarged mesenteric lymph nodes were absent. Mainly in that region, the visceral and parietal peritoneum were covered with multiple white-grayish nodules with a diameter of 2 to 5 mm. Auramine staining of multiple biopsies was mycobacteria-positive. A culture of the nodules, ascites, and gastric aspirate yielded MTB.
A thoracic computed tomography (CT) scan showed miliary lesions throughout the lungs. Retrospectively, these fine lesions were already present on the preoperative chest radiograph, but unnoticed at that time. A cerebral CT scan and lumbar puncture excluded the presence of tuberculous meningitis. On retinoscopy, a few choroidal tubercles were visible. The tuberculin skin test (TST) was negative initially, but became positive 1 month later.
Treatment was initiated with intravenous rifampicin and ciprofloxacin. Within 1 week, the patient was able to take oral medication, so treatment switched to oral rifampicin, isoniazide, ethambutol, and pyrazinamide. Isoniazide was discontinued when the MTB strain showed resistance to isoniazide. Pyrazinamide was stopped after 2 months, and treatment with rifampicin and ethambutol was continued for another 7 months. She fully recovered.
The TST of the father measured 22 mm; nonetheless, his chest radiograph was normal. Because he did not have signs of active TB, he was diagnosed with a latent TB. The initial treatment was isoniazide but changed to rifampicin for 4 months when the strain, isolated from the child, seemed to be isoniazide resistant.
The mother, aged 30 years, immigrated to the Netherlands at the age of 22. She was not bacille Calmette-Guerin vaccinated. The TST, on arrival in the Netherlands, showed an induration of 18 mm. The chest radiograph showed a small pleural adhesion and calcification that was suspect for primary TB complex. The mother was not treated, but received follow-up chest radiographs every 6 months for 2 years, which remained unchanged.
After diagnosing TB in the child, we advised the mother to consult an internist and gynecologist to evaluate the presence of extrapulmonary TB. However, the mother refused additional examination, but when she developed fever and backache, investigations showed right-sided hydronephrosis. The pyelum was drained by nephrostomy, and urine samples showed acid-fast rods and MTB was confirmed by polymerase chain reaction.
At that time, it became apparent that the mother had consulted a gynecologist because of fertility problems. An abdominal ultrasound, at that time, showed calcification in the right ovary and tube. The mother refused additional analysis of the ovarian lesions, and no final diagnosis was made.
Because the child's MTB strain was isoniazide resistant, the mother received rifampicin, ethambutol, and pyrazinamide for 2 months, followed by rifampicin and ethambutol for 7 months. After 2 months, when infection parameters had normalized, nephrectomy of the right kidney was performed. Histopathology analysis revealed extended granulomatous inflammation and caseous necrosis of the kidney. The strain recovered from the mother's urine sample had the same restriction fragment length polymorphism (RFLP) pattern as the strain isolated from the child (Fig 1). These 2 strains together form cluster 7195 and are unique in the Netherlands. The exclusive RFLP pattern is indicative for the transmission of the MTB strain from the mother to the child.6 Because the RFLP pattern is unique in the Netherlands, 1 can argue that the TB infection of the mother took place in another country or before RFLP analysis was done on all MTB strains in the Netherlands.
|
According to the Dutch guidelines for contact and source tracing, the investigation of extrapulmonary TB patients focuses on the identification of a possible infectious source case. During these investigations by the Municipal Health Service, it became apparent that none of the 16 examined contacts (including the grandparents) had pulmonary TB. Ten of these contacts were examined by using a TST. As mentioned above, the father of the child was the only person with a positive TST and, thus, was treated for latent TB. Because there was no case of active (contagious) TB found, respiratory infection of the child became unlikely. This finding, together with similar and unique RFLP pattern of the strains isolated from the mother and child, directed the differential diagnosis strongly to neonatal or CTB.
DISCUSSION
To our knowledge, this is the first report of TB mimicking an intussusception. Only after laparotomy did the suspected intussusception manifest as TB. The obstruction of the bowel, which probably caused the symptoms, was attributable to the inflammatory process and the multiple adhesions. Because of the acute presentation, there was no substantial diagnostic delay in this patient, which is in contrast with many other cases with a gradual increase of symptoms. This diagnostic delay is of great importance, because it is an important determinant of poor disease outcome.4,5 The atypical presentation might lead to underestimation of the incidence of TB, especially in settings with inadequate access to health care and diagnostic facilities, and undiagnosed TB or CTB might be the cause of some unexplained infant death.7
The Royal Netherlands Tuberculosis Association registers the incidence of TB in the Netherlands. In their surveillance report on the incidence of TB from 1996 to 2005, they showed that the incidence of TB in children between 0 and 14 years of age is, on average, 3.0 per 100000 (range: 1.7–3.9).7 Because CTB is a rare presentation of TB, there is no specific registration of CTB and, therefore, correct incidence numbers are difficult to give. In the literature, there is only 1 Dutch case of CTB described in the last 50 years.5
Extrapulmonary TB, like genitourinary TB, occurs more often in developing countries, probably because of poorer treatment of pulmonary TB. Approximately 20% of TB patients develop extrapulmonary symptoms; 20% to 40% of those involve the genitourinary system.9 Most genitourinary TB cases are the result of hematogenous spread of mycobacteria from the primary pulmonary focus to the kidney and female genital organs. Genital TB often results in infertility. Consequently, the likelihood of vertical transmission decreases when the female genital organs are involved.
Possible routes of infection in utero are hematogenously spread through the umbilical vein or by the ingestion of infected amniotic fluid. Transmission during labor is possible through direct contact with an infected birth canal or the ingestion of infected amniotic fluid or urine. Nonrespiratory infection after birth can be caused by the ingestion of infected urine (which is highly contagious) from an infected carrier and is possible when appropriate precautions are not taken.
The best way to prevention transmission is to treat the mothers during pregnancy. Treatment of the (extra) pulmonary TB of the pregnant women should begin as soon as possible, to guarantee the best treatment outcome.10 The place of isoniazide in this treatment is still under debate, because the prevention of TB might result in isoniazide-induced hepatitis, which increases the death rate of the unborn child. Other (new) medication enabling the treatment of (latent) TB without isoniazide possible could solve this problem.10
Another route of infection is via infected mother's milk, although accurate risk analysis on this matter is not available. One can argue that women with a high MTB burden, who even have the bacteria in their blood, are at risk to transmit the disease this way. Also when there are extrapulmonary caseous lesions in the breast, the possibility of transmission is increased. Therefore, when a child is suspected of having TB, without a liable route of respiratory infection, breast examination of the mother should be performed.
In CTB, the route of transmission directs the presentation and pattern of infection.2,4 Intrauterine infection via the umbilical vein typically results in a primary complex in the liver.11 It is known that infection by aspiration or ingestion may show up as a primary infection in the lung or gut, and direct inoculation onto fragile epithelia even, for example, in the middle ear.12 Postnatal transmission of TB by inhalation of the bacteria is the most frequent mode, seen as a primary complex in the periphery of the lung. In adults, only a minority of MTB-infected persons develop active infection. In children, this proportion is not exactly known but is considered to be higher because of an immature immune response. Therefore, TB will also more easily disseminate in young children.12 Nonetheless, in neonates, protective immunity of the mother could help to prevent the development of active TB. This protective immunity through placental transmission of active elements of the immune system, such as transfer factors, might explain some cases of delayed-onset CTB.13 Immature immune response of infants also explains a reduced reliance on the TST. For that reason, in case of suspected CTB, specific staining and cultures of gastric aspirates, urine, and cerebral spinal fluid are mandatory. In addition, aspirates and/or biopsies from affected organs are required. Furthermore, cerebral and thoracoabdominal CT scans or and MRI and a fundoscopy are essential to detect dissemination. To confirm congenital transmission, a thorough evaluation of possible sources of TB is necessary, including evaluation for maternal genitourinary TB.
To further understand the difficulties of CTB, a literature review on the recent publications related to CTB provides the following observations. Singh et al14 described, in 2007, 4 different cases of TB in young infants, all presented with diverse, aspecific symptoms. Of these 4 cases, only 1 was classified as a CTB, because of the early onset of disease and the absence of an index patient at the time of presentation. In the CTB case, the mother of the child (as with our case) presented herself with neurologic complains because of multiple tuberculomas 1 week after the TB diagnosis of the child. Singh et al13 argue that it might be better to use the term perinatal TB instead of CTB or neonatal TB, because that differentiation is only of epidemiological importance and can be difficult.
Nicolaidou et al15 reported a classic case of CTB in 2005. They described a 1-month-old infant who was infected by endometrial TB during pregnancy. The mother did not have any symptoms during pregnancy, had no history of TB, and remained with a normal chest radiograph. The only sign of possible infection was a positive TST. This lack of symptoms is seen more often. In 
75% of the cases of CTB, the pregnant women remained asymptomatic during extrapulmonary TB and, as a result, were misdiagnosed.4,16 Again, this patient showed that CTB can present itself like a more common infection and a delay in the correct therapy can be the result.3 Therefore, when a child does not respond to broad-spectrum antibiotics and supportive treatment, CTB should be in the differential diagnosis and, unless a high index of suspicion is maintained, the diagnosis can be missed.13,17
The similarity of strains, active genitourinary TB in the mother, and absence of another source of TB, which is 1 of the Cantwell's criterion, support our hypothesis that transmission occurred prenatally or perinatally. Unfortunately, no endometrial biopsies were performed and, therefore, we cannot exclude intrauterine exposure. Furthermore, postnatal transmission of MTB via breast milk or infected urine of the mother cannot be excluded. Nonetheless, we hypothesize that in our case transmission of mycobacteria occurred by ingestion of infected urine during delivery, because of the massive renal infection of the mother and the possible high risk of exposure to the infected urine during labor. Because our patient was almost 5 months (136 days) on presentation, far above the average of previously reported cases of CTB, we would like to suggest that clinicians should consider CTB in their differential diagnosis even after the age of 3 months.
CONCLUSION
Our case demonstrates that TB can present in a nonspecific way. To our knowledge, this is the first report of TB in an infant, that mimics an intussusception. The way of infection of this child, where aerogenic infection is very unlikely, can be aspiration or ingestion of infected amniotic fluid, urine or breast milk, which could have taken place before, during, or after birth.
ACKNOWLEDGMENTS
We thank Dr D. van Soolingen (National Institute of Public Health and the Environment, National Reference Laboratory for Mycobacteriology, Bilthoven, Netherlands) for the Restriction Fragment Length Polymorphism-typing.
FOOTNOTES
Accepted Oct 1, 2007.
Address correspondence to Jurriaan E. M. de Steenwinkel, MD, MSc, Department of Medical Microbiology and Infectious Diseases, Erasmus MC, s-Gravendijkwal 230 (Room L327), 3015 CE Rotterdam, Netherlands. E-mail: j.desteenwinkel{at}erasmusmc.nl
Financial Disclosure: The authors have indicated they have no financial relationships relevant to this article to disclose.
REFERENCES
- World Health Organization (WHO). WHO report 2007: global tuberculosis control: surveillance, planning, financing (WHO/HTM/TB/2007.376). Available at: www.who.int/tb/publications/global_report/2007/en/. Accessed March 27, 2008
- Cantwell MF, Shehab ZM, Costello AM, et al. Brief report: congenital tuberculosis.
N Engl J Med. 1994;330
(15):1051
–1054
[Free Full Text] - Beitzke H. Ueber die angeborene Tuberkuloese Infektion. Ergeb Gestamten Tuberkulose Forsch. 1935;7 :1 –30
- Mazade MA, Evans EM, Starke JR, Correa AG. Congenital tuberculosis presenting as sepsis syndrome: case report and review of the literature. Pediatr Infect Dis J. 2001;20 (4):439 –442[CrossRef][Web of Science][Medline]
- Hageman J, Shulman S, Schreiber M, Luck S, Yogev R. Congenital tuberculosis: critical reappraisal of clinical findings and diagnostic procedures.
Pediatrics. 1980;66
(6):980
–984
[Abstract/Free Full Text] - Kiers A, Drost AP, van Soolingen D, Veen J. Use of DNA fingerprinting in international source case finding during a large outbreak of tuberculosis in The Netherlands. Int J Tuberc Lung Dis. 1997;1 (3):239 –245[Web of Science][Medline]
- Kumar R, Gupta N, Sabharwal A, Shalini. Congenital tuberculosis. Indian J Pediatr. 2005;72 (7):631 –633[Medline]
- Erkens CGM, Kalisvaart NA, Slump E, Sebek M, van Soolingen D. Tuberculose in Nederland 2005: surveillancerapport over de tuberculosesituatie in Nederland. KNCV Tuberculosefonds; 2007
- Lenk S, Schroeder J. Genitourinary tuberculosis. Curr Opin Urol. 2001;11 (1):93 –98[CrossRef][Medline]
- Smith KC. Congenital tuberculosis: a rare manifestation of a common infection. Curr Opin Infect Dis. 2002;15 (3):269 –274[Web of Science][Medline]
- Berk DR, Sylvester KG. Congenital tuberculosis presenting as progressive liver dysfunction. Pediatr Infect Dis J. 2004;23 (1):78 –80[CrossRef][Web of Science][Medline]
- Skevaki CL, Kafetzis DA. Tuberculosis in neonates and infants: epidemiology, pathogenesis, clinical manifestations, diagnosis, and management issues. Paediatr Drugs. 2005;7 (4):219 –234[CrossRef][Medline]
- Fabre RA, Perez TM, Aguilar LD et al. Transfer factors as immunotherapy and supplement of chemotherapy in experimental pulmonary tuberculosis. Clin Exp Immunol. 2004;136 (2):215 –223[CrossRef][Web of Science][Medline]
- Singh M, Kothur K, Dayal D, Kusuma S. Perinatal tuberculosis a case series.
J Trop Pediatr. 2007;53
(2):135
–138
[Abstract/Free Full Text] - Nicolaidu P, Psychou F, Stefanaki K, Tsitsika A, Syriopoulou V. Congenital tuberculosis: a case report.
Clin Pediatr (Phila). 2005;44
(5):451
–453
[Free Full Text] - Abughali N, Van der Kuyp F, Annable W, Kumar ML. Congenital tuberculosis. Pediatr Infect Dis J. 1994;13 (8):738 –741[Web of Science][Medline]
- Chen A, Shih SL. Congenital tuberculosis in two infants.
AJR Am J Roentgenol. 2004;182
(1):253
–256
[Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
