Академический Документы
Профессиональный Документы
Культура Документы
com
Original article
of Infectious
Diseases and Microbiology,
Institute of Child Health,
London, UK
2Department of Infectious
Diseases, Great Ormond
Street Hospital for Children,
London, UK
Correspondence to
Sarah Eisen, Department
of Infectious Diseases and
Microbiology, Institute of Child
Health, 30 Guilford St, London
WC1N 1EH, UK; saraheisen@
hotmail.com
Received 19 December 2011
Accepted 18 May 2012
Published Online First
25 June 2012
ABSTRACT
Objectives To review our experience of spinal
tuberculosis (TB) at a major UK paediatric tertiary
referral centre.
Methods The authors performed a retrospective case
survey of 21 patients admitted to Great Ormond Street
Hospital over a 15-year period (19952010) with confirmed
or presumed spinal TB. Data were collected concerning
demographics, clinical, laboratory and radiological
characteristics, treatment and clinical outcome.
Results Only one patient was of Caucasian origin.
Four (19%) had a previous diagnosis of TB, 11 (52%) a
known contact, 10 (48%) had received BCG vaccine
and none were HIV-positive. Clinical presentations
included systemic symptoms (18 patients), back pain
(16 patients), deformity (five patients) and neurological
deficits (12 patients). Mycobacterium tuberculosis was
isolated from 14 patients (67%) including one multidrug resistant strain. Spinal cord compression or critical
stenosis was demonstrated in eight patients (38%).
All received TB treatment for at least 12 months; six
patients received treatment for a longer period. Seven
(33%) underwent surgical intervention. Seventy-five
per cent showed clinical and radiological resolution
after treatment. No patients died or suffered long-term
neurological deficit.
Conclusions Spinal TB in children needs a high index of
suspicion for diagnosis. Early referral to an expert centre
allows a multidisciplinary approach to management.
The authors recommend that treatment should be
individually tailored and may need to exceed 12 months
in cases of poor adherence, extensive disease or drug
resistance.
INTRODUCTION
Tuberculosis (TB) is the most common infectious disease worldwide, and, in 2009, affected
four per 100 000 UK-born children under 5 years
old in the UK.1 Spinal TB refers to infection of
one or more vertebral bodies with Mycobacterium
tuberculosis, with or without involvement of the
spinal cord.2 3 This condition is believed to arise
by haematogenous dissemination, often without an obvious primary focus in children, who
may be vulnerable to discitis because of the
persisting anastomosis between the vertebral
endplate and disc. Septic embolus may result in
thrombosis and bony necrosis, often with extension to adjacent vertebral bodies, vertebral endplate destruction, subligamentous spread and
cord involvement.4 There is limited evidence to
guide treatment of spinal TB in children, which
remains controversial. This review of our experience of childhood spinal TB over 15 years is, to
our knowledge, the only paediatric case series
reported from the UK.
724
RESULTS
Demographics and clinical characteristics
Twenty-one patients were identied. Median
age was 9.7 years (range 3.415.9 years). Eleven
patients were Black African, seven Asian, two
Middle Eastern and one Caucasian. Nine were
born in the UK, one in the Netherlands and the
remainder outside Europe. Ten patients reported
Arch Dis Child 2012;97:724729. doi:10.1136/archdischild-2011-301571
Original article
travel to a country where TB is endemic within the year preceding diagnosis. Four (19%) had a previous diagnosis of TB
disease (three of whom had been diagnosed within the previous 3 years in the UK, only one of whom had completed a
full treatment course), while 11 (52%) had a history of recent
active TB disease in a relative. Ten patients (48%) had received
BCG. None were known to be HIV-positive (although only
seven were tested). Of 14 patients with data available, four
(29%) were <0.4th centile for weight.
Table 1 shows the symptoms and signs recorded in each of
the 21 patients on admission.
Eighteen patients presented with systemic and 12 with neurological symptoms, and ve with spinal deformity. Back pain
was common and associated with systemic symptoms in all
but three (night sweats in 11, weight loss in nine, fever in eight
and anorexia in ve). The ve patients without back pain all
reported one or more systemic symptoms. Median symptom
duration before diagnosis was 6 weeks (range 216 weeks).
Skin tests
Tuberculin test results (before 2003: 10TU Evans tuberculin; after 2003: 2TU Statens Serum Institut tuberculin) were
recorded for 20 patients, of whom 18 (90%) were positive and
two (10%) negative (both in patients without microbiological
diagnosis and with a negative QuantiFERON-TB Gold test but
histology suggestive of TB). All positive results were >10 mm
(widest diameter); 17 were >15 mm (eight of these patients had
previously received BCG).
Radiology
Laboratory data
Table 1
The presenting features on admission in 21 children with spinal tuberculosis, shown in decreasing order of frequency
Patient
Presenting features
Back pain
Night sweats
Fever
Weight loss
Cough
Anorexia
Limp
Weakness/reduced
power
Lymphadenopathy
Deformity
Sensory change
Abdominal pain
Hyperreflexia
Discharging sinus
Features relating to
extra-spinal site*
Identified through
contact tracing
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
10
11
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
14
15
16
17
18
19
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
13
Y
Y
Y
Y
Y
12
Y
Y
Y
Y
Y
Y
Y
Y
Y
20
21
Total
16
15
12
11
10
9
7
7
Y
Y
Y
Y
Y
Y
Y
Y
7
5
5
2
1
1
1
Y
Y
Y
725
Original article
paraspinal abscesses in 15 (71%) and disc involvement in 12
(57%). Cord compression or critical stenosis was demonstrated
in eight (38%), although none showed actual involvement of
the spinal cord. Figure 1 shows a spinal MRI demonstrating
vertebral collapse, compression, discitis and subligamentous
spread at the level of T11.
Chest x-ray was abnormal in eight patients (38%) (two
showed pleural effusions, three mediastinal lymphadenopathy
and three consolidation). Twelve patients (57%) had co-existent
extra-spinal TB disease (pulmonary, osteoarticular, pericardial
or TB meningitis) of whom two had multiple extra-spinal disease foci. Five patients (24%) had coexisting psoas abscess.
Treatment
Table 2 summarises treatment and outcome for each patient.
All patients received inpatient care (median duration 12
days, range 242) and antituberculous drugs. Ten (48%)
received standard quadruple therapy for fully sensitive TB
(rifampicin, isoniazid, pyrazinamide, ethambutol). The three
patients with isoniazid-resistant isolates (or contacts) received
rifampicin and ethambutol (one also received ciprooxacin).
The patient with an MDR isolate received ethambutol, ciprooxacin, cycloserine, prothionamide and intravenous amikacin. Six patients changed drugs during treatment, two due
to clinical or radiological deterioration (moxioxacin added),
one due to side effects (clarithromycin substituted for ethambutol following a documented reduction in visual acuity) and
three based on subsequently available sensitivities.
Table 2
Patient
Treatment
1
2
3
4
5
6
7
8
9
10
11
RHZ, streptomycin
E, ciprofloxacin*, cycloserine,
prothiamide, amikacin
RHZ
RE
RHZE
RHZ, streptomycin
RHZE
RE, ciprofloxacin*
RHZE
RE
RZE
12
RHZE, moxifloxacin
13
RHZE
14
15
RHZE
RHZE, clarithromycin,
ciprofloxacin*, streptomycin
16
17
18
19
20
21
RHZE
RHZE
RHZE
RHZE
RHZE
RHZE, moxifloxacin
Corticosteroids
Treatment
duration
(months)
Adherence
difficulties
N
N
N
N
12
48
Y
Y
Y
N
N
Y
N
N
N
Y
N
N
N
Y
Y
Y
N
N
N
N
12
24
12
12
12
12
12
16
12
N
N
N
N
N
N
N
Y
N
18
18
Y
N
Y
Y
12
12
N
N
N
N
N
N
N
Y
N
N
Y
Y
N
Y
12
12
12
12
12
18
N
N
Y
N
N
N
Surgery
Unknown
MDR-TB
Unknown
H resistance
Unknown
H resistance, ciprofloxacin sensitive
H resistant contact
Unknown
Failure to resolve with further lumbar
vertebral collapse
Outcome
Resolved
Two relapses on
treatment
Resolved
Resolved
Resolved
Resolved
Resolved
Resolved
Resolved
Resolved
Residual deformity
and pain
Resolved
Two relapses then
resolved
Resolved
Relapse and
deafness
Resolved
F/u
F/u
F/u
F/u
F/u
E, ethambutol; F/u, follow-up ongoing; H, isoniazid; MDR, multi-drug resistant; R, rifampicin; Z, pyrazinamide.
*It should be noted that ciprofloxacin has the weakest anti-TB activity of all quinolones and is no longer recommended for the treatment of drug resistant TB.
726
Original article
Fifteen patients (71%) received therapy for 12 months only.
Those with isoniazid-resistant strains or extensive extraspinal disease received longer treatment courses (16 and 24
months and 18 months, respectively), as did those whose condition deteriorated on treatment (18 months). The patient with
MDR-TB received 4 years of treatment. None received directly
observed therapy, but adherence difculties are documented in
six patients, including those who deteriorated on treatment.
Nine patients (43%) received steroid therapy, of whom eight
had radiologically demonstrated cord compression and one
associated raised intracranial pressure (with co-existent tuberculous meningitis con rmed on lumbar puncture).
Seven patients underwent surgical intervention (33%).
Debridement and decompression was performed to resolve
cord compression in the four patients with neurological manifestations. Spinal instrumented stabilisation was performed in
three patients with instability. There were no surgical complications or failures of procedure. Time from diagnosis until
surgery was a median of 7 days (range 125 days), determined
by clinical severity, and all surgery was performed after initiation of medical therapy (median 7 days after treatment started,
range 022 days).
Outcome
All patients were alive at the end of follow-up (median duration 24 months; ve ongoing). Twelve of the 16 discharged
cases (75%) resolved fully, three relapsed (one off and two on
treatment, of whom one had MDR-TB). One patient is known
to have suffered long-term disease or treatment-related morbidity (deafness following streptomycin therapy).
DISCUSSION
Certain features should raise suspicion of spinal TB
Diagnosis of spinal TB is notoriously difcult. 5 Our ndings
indicate key features which should alert the clinician in primary, secondary or tertiary care. The demographics of our
cohort reect those seen locally and nationally, 5 the majority
originating from immigrant families, reecting higher prevalence in country of origin.6 A history of TB contact or travel
should be sought (and probably accounts for the nine affected
UK-born children in our cohort) as should a history of previous
TB, present in 19% of our group. Systemic symptoms were
present in almost all patients, consistent with other paediatric studies7 and over half presented with neurological symptoms. Rates of neurological features at presentation quoted in
the paediatric literature vary.7 8 It is clear from our cohort that
the association of back pain with systemic symptoms should
raise alarm. Likewise, long symptom duration is suspicious,
also reported elsewhere in both adults 5 and children.7 This is
likely to result from insidious onset of non-specic symptoms9
and is important as delayed presentation results in increased
complications.10
National Institute for Health and Clinical Excellence guidelines12 recommend bacteriological con rmation of disease,
which can be challenging where infective foci may be inaccessible and biopsy invasive. Microbiological diagnosis was
con rmed in two thirds of our patients, with a culture con rmation rate (where biopsy specimens were obtained) of 77%,
consistent with the published literature. 5 9 Histology can be
useful and allowed diagnosis in the absence of positive microbiology in three cases. Of note, the use of radiological guidance
to obtain specimens avoided open surgical biopsy in several
patients and may facilitate rapid treatment, avoid invasive
diagnostic procedures and decrease hospitalisation time.13
Single, multi and extensive drug resistance is increasing
nationally and globally. Identication of resistant strains is
vital for effective treatment and disease control. The rate of
single drug resistance in our cohort appears higher than in
published data,14 but our sample size is small. There are, as
yet, no data specically relating to resistance rates in isolates
in spinal TB in the UK, but adult data from India suggest that
this is an increasing concern.15 History of contact with drug
resistant TB, as with two patients here, should result in a high
index of suspicion of drug resistant TB in the patient.16
Radiological investigation remains crucial in diagnosis,
determining the extent of disease and the risk of complications. According to the Medical Research Council de nition,17
18 active disease is diagnosed on plain radiograph by loss of
cortical outline and rarefaction of affected vertebral bodies.
The disc is frequently affected, as in more than half of our
patients. The role of plan radiographs has been largely superseded by MRI to identify early soft tissue abnormality, spinal
cord compression, epidural involvement and subligamentous
spread.19 Cord compression was demonstrated radiologically
in nine patients (ve without clinical neurological abnormality), allowing identication of those in whom corticosteroids
or surgical intervention may be benecial. 5 Disease distribution (predominantly thoracic) is consistent with limited reports
within the adult literature, 5 although higher rates of lumbar
involvement have been shown in children.7
Original article
Table 3
Guideline
Use of steroids
Role of surgery
Follow-up
No comment
Standard combination
(RHZE; ethambutol may be
omitted) for 69 months.
DOT should always be
used in children
Not recommended
No comment
NICE, 201112
No comment
Not recommended
routinely but may be
useful if symptoms
worsen on therapy or if
spinal cord compression
No comment
CDC, Centers for Disease Control and Prevention; DOT, directly observed therapy; E, ethambutol; H, isoniazid; NICE, National Institute for Health and Clinical Excellence;
R, rifampicin; Z, pyrazinamide.
CONCLUSIONS
There is a need for further evidence from multi-site prospective studies to guide treatment in paediatric patients with
spinal TB. In the absence of this evidence, we recommend a
exible individualised approach, with careful follow-up, as
demonstrated in this case series. It is crucial that a high index
of suspicion for the diagnosis of spinal TB is maintained, with
early referral to an expert centre where liaison between infectious diseases specialists, radiologists, microbiologists and
surgeons is possible. Identication of drug resistant or extensive disease, complications which may benet from steroids or
surgical intervention, or adherence difculties, with monitoring for relapse, must inform appropriate management in the
absence of a robust evidence base.
Acknowledgements The authors are grateful to the patients and their families
Contributors SE analysed the data and wrote the paper. SE and LH acquired
and collated the data from original sources. DS and VN assisted in data analysis,
determining content and writing up the paper.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
Arch Dis Child 2012;97:724729. doi:10.1136/archdischild-2011-301571
Original article
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
729
doi: 10.1136/archdischild-2011-301571
These include:
References
Email alerting
service
Topic
Collections
Receive free email alerts when new articles cite this article. Sign up in
the box at the top right corner of the online article.
Notes