Professional Documents
Culture Documents
CITATIONS
READS
41
102
3 authors:
Andrew Currie
University of Oxford
SEE PROFILE
SEE PROFILE
OVERVIEW
S.K. has received a travel grant from Roche. P.J.M. chairs a Data, Safety and
Monitoring Board for Bristol Myers Squibb and has received honoraria and
travel grants in the past from Novartis, Roche, Astellas, and Genezyme
(P.J.M.).
1
Centre for Evidence in Transplantation, Royal College of Surgeons of England and London School of Hygiene and Tropical Medicine, University
of London, London, United Kingdom.
2
Address correspondence to: Peter J. Morris, A.C., F.R.S., F.R.C.S., Centre
for Evidence in Transplantation, Royal College of Surgeons of England,
35-43 Lincolns Inn Fields, London WC2A 3PE, United Kingdom.
E-mail: pmorris@rcseng.ac.uk
P.J.M. conceived the idea for the paper. A.C. and P.J.M. undertook the literature
search and reviewed the papers. S.R.K. assisted with the meta-analysis. All
authors were involved with writing the manuscript.
Received 11 January 2010. Revision requested 10 March 2010.
Accepted 14 June 2010.
Copyright 2010 by Lippincott Williams & Wilkins
ISSN 0041-1337/10/9007-695
DOI: 10.1097/TP.0b013e3181ecea8d
695
696 |
www.transplantjournal.com
TABLE 1. The prevalence of Mycobacterium tuberculosis drawn from observational studies of tuberculosis in renal
transplant recipients in both single and multiple centers and in the general population of the same country
Country
TB prevalence in general
population % (51)
N (transplant
recipients)
N (patients
with TB)
Prevalence
(%)
References
Argentina
Belgium
Brazil
China
Hong Kong
India
0.048
0.011
0.055
0.201
Not listed
0.299
384
2502
982
1947
440
305
1414
77
163
202
1510
12820
1350
545
850
273
487
1261
756
404
151
368
359
443
261
274
520
880
283
935
633
15870
3921
456
14
9
44
25
23
36
166
10
21
27
8
44
52
10
130
8
22
27
29
6
5
5
9
20
8
16
22
36
10
19
11
66
3
13
3.64
0.36
4.5
1.28
5.2
11.8
13
13
12.96
13
1.4
0.3
3.9
1.8
15.2
2.9
4.5
2.1
3.8
1.5
3.3
1.3
2.5
4.5
3.06
5.8
4.2
4.1
3.1
2.0
1.7
0.4
0.07
3.13
Iran
Mexico
Pakistan
Slovenia
South Africa
Spain
Taiwan
0.028
0.025
0.263
0.015
0.692
0.024
Not listed
Thailand
Tunisia
0.197
0.028
Turkey
0.032
UK
USA
0.012
0.003
Yugoslavia
0.041
Chemoprophylaxis with INH has been shown in randomized controlled trials (RCTs) to be effective in reducing
the development of TB by 60% to 90% in immunocompetent
individuals in large-scale studies (14, 15) and also in HIV positive individuals (16, 17). Given the difficulties in diagnosing and
treating established TB infection in the immunocompromised
transplant recipient, along with the risks to graft function that it
poses, there is interest in similar chemoprophylaxis regimens.
This review will critically appraise the published literature on
prophylaxis in renal transplantation, because there is a paucity of
data in the transplantation of other organs.
MATERIALS AND METHODS
A systematic literature search was performed using Ovid MEDLINE,
EMBASE, the Cochrane Central Registry of Controlled Trials, the Transplant
Library from the Centre for Evidence in Transplantation, and Clinical Trial
Currie et al.
RESULTS
A total of 11 reports met the inclusion criteria in the review
of INH prophylaxis (Fig. 1), of which six were nonrandomized
or observational studies (Table 2). Five reports of four RCTs
conducted from 1994 to 2009 were reviewed (Table 4). One of
these RCTs was reported twice, once in 2006 (21) and again in
2009 (22), in an analysis of the same patients but with a longer
follow-up, albeit with slightly discrepant numbers. After contacting the authors, we have used the later article (22) with the
longer follow-up in our analysis. The methodologic quality of
these RCTs was moderate (Table 3) and meta-analysis was undertaken of the data in these four trials.
Observational Studies
INH prophylaxis in renal transplant recipients has been
reported to be beneficial in observational studies by Yildiz et
al. (23), Spence et al. (24), Higgins et al. (25), and Matuck et
al. (26). In contrast studies by Apaydin et al. (27) and Sayiner
et al. (28) could not determine any benefit from chemoprophylaxis. Study characteristics and results are summarized in
Table 2.
In the study from Turkey by Yildiz et al. (23), 23 renal
transplant recipients who had a history, or TB sequelae on
chest radiograph, were given isoniazed 300 mg/day for 1 year
and one patient with a history of TB did not receive prophylaxis. None of the 23 patients who received prophylaxis developed TB, whereas the one patient with a history of TB, and
who did not receive prophylaxis, developed TB.
In the United Kingdom, Higgins et al. (25) identified 39
patients at increased risk of developing TB by virtue of previous residence in an endemic area or by exposure to a relative
697
22
0
44.4
27
23
12
Post
Turkey
Post
Turkey
Post
Post
Turkey
UK
Higgins
et al. (25)
Pre
Brazil
Previous infection/CXR
changes
Previous exposure/residence
in endemic area
223
51
NR
100
0
6
8.8
6
(1129)
7.7
0
36.7
13
23
Previous infection/CXR
changes
Variable risk
NA
3.4
36
0
30
Previous infection
NR
NA
0
Nil TB
32
0
14
Post
USA
Previous exposure
No
prophylaxis
Country
Commencement of
prophylaxis
(pre-/posttransplant)
Method of patient
selection
Prophylaxis
Follow-up
(mo) after
transplant
No. patients
Spence
et al. (24)
Matuck
et al. (26)
Sayiner
et al. (28)
Apaydin
et al. (27)
Yildiz et al. (23)
TABLE 2.
Summary of key findings in observational studies of INH prophylaxis in renal transplant recipients
Timing of TB
infection
posttransplant (mo),
mean (range)
Prophylaxis
No
prophylaxis
www.transplantjournal.com
Incidence of TB (%)
698 |
Randomized Studies
Four RCTs have been undertaken examining the effect of INH prophylaxis in renal transplant recipients.
Quality assessment is shown in Table 3 and, as can be seen,
the overall methodologic quality of the reports ranges
from poor to good. Study characteristics are shown in Table 4. All the studies were carried out in India or Pakistan, with
a mixture of pre- and posttransplant prophylaxis used. Primary
immunosuppression, when reported, consisted of cyclosporine,
azathioprine, and prednisolone. In the RCTs, the recipients received INH for 1 year. The reported follow-up posttransplantation ranged from 2 (29) to 4 years (22).
John et al. (29) in a study of primary INH prophylaxis
in dialysis and renal transplant recipients in India found a
trend toward lower incidence of TB in INH recipients. However, it did not reach statistical significance. In this study, 184
patients on hemodialysis were studied in a double-blind, randomized, placebo-controlled trial with INH. There were 92
patients in each group and the randomization was performed
using random number tables. No information is provided on
immunosuppressive protocol, and no assessment of LTBI is
mentioned. The INH 300 mg/day and placebo were continued after transplantation for duration of 1 year. Seven patients in the INH group developed TB in the first year (but in
four patients there was incomplete prophylaxis) compared
699
Currie et al.
TABLE 3. Methodologic quality of reports of randomized controlled trials of INH prophylaxis in postrenal transplant
tuberculosis
Randomizationa
Blindingb
Withdrawalsc
Jadad
score
ITT
analysis
Allocation
concealmentd
2
1
2
2
1
1
0
0
0
0
1
1
1
0
0e
4
2
3
2
2
Yes
Yes
Yes
No
Yesf
Yes
No
No
No
No
a
Jadad Score Key: 0 not randomized, or randomized but inappropriate method; 1 described as randomized; 2 described as randomized with an
appropriate method.
b
0 not blinded, or blinded but inappropriate method; 1 described as blinded; 2 described as blinded with appropriate method.
c
0 inadequate description of participant withdrawal and outcome; 1 adequate description of participant withdrawal and outcome.
d
The investigators are unable to determine to which intervention the patient has been allocated.
e
Although 12 withdrawals were reported, it was not stated in which arm of the trial these occurred.
f
No withdrawals occurred in the study by Naqvi et al. (21).
ITT, intention-to-treat.
Nil stated
NS
Nil stated
DISCUSSION
7.73
CSA, cyclosporin A; AZA, azathioprine; INH, isoniazid; TB, tuberculosis; NS, not significant; NR, not reported.
16
207
Placebo
Clinical history
CSA, prednisolone,
AZA
Naqvi
Pakistan Posttransplant
et al. (22)
4 yr
CSA, prednisolone,
AZA
Pretransplant
Vikrant
India
et al. (31)
2 yr
NR
NR
NR
Posttransplant
Agarwal
India
et al. (30)
NR
Pretransplant
2 yr
Mantoux, clinical
history
92
92
30
60
54
55
181
INH
Placebo
INH
Placebo
INH
Placebo
INH
3
4
3
15
9
27
1
3.25
4.34
10
25
16.7
49.1
0.55
NS
India
John et al.
(29)
Country
Timing of
prophylaxis
Follow-up
Immunosuppression posttransplant
NR
Latent TB
screening
Intervention
TB
TB (%)
P (INH vs.
placebo)
Hepatic
dysfunction (%)
P (INH vs.
placebo)
www.transplantjournal.com
TABLE 4. Summary of key findings in randomized controlled trials of INH prophylaxis in postrenal transplant tuberculosis
700 |
TB in the renal allograft recipient is a common problem, especially when the incidence and prevalence are extensive in the general population. The nonrandomized studies
examined in this review suggested that there is a potential
value of INH similar to chemoprophylaxis in populations of
renal transplant recipients at risk of development of active TB
infection, namely patients in regions in which TB is endemic
and in patients with evidence of previous TB infections or
exposure.
There are only four RCTs assessing the benefits of INH
prophylaxis. The methodologic quality of the articles is moderate. Our meta-analysis demonstrated a statistically significant benefit of administering INH prophylaxis to prevent TB
in renal transplant recipients. However, two of the studies
analyzed prophylaxis from the pretransplantation period,
while the patients were still on dialysis. This would seem appropriate in view of the immunocompromised state of all
forms of renal replacement therapy, but does further confound any analysis. Furthermore, all four RCTs were carried
out in high-risk patients in endemic areas (India and Pakistan). Thus, firm conclusions about the role of prophylaxis
are difficult to make based on such a small number of trials
and certainly the findings cannot be extrapolated to renal
transplant populations in regions in which TB is not endemic.
The major risk factors for TB in a large prospective
analysis in an Indian population were chronic liver disease
(2 times increased risk), age, other coexisting infections, particularly deep mycoses, pneumocystis pneumonia and nocardia (1.6 times), OKT3 usage (1.8 times), and cytomegalovirus
infections (2.25 times) (32). Disseminated disease may also be
more common with the use of OKT3 and in the presence of
cytomegalovirus infection. Cyclosporine use seems to advance the time of onset to an earlier date, and patients on
cyclosporine seldom develop TB later than 6 years after transplantation (32). The factors associated with death on univariate analysis were the recipient age, human leukocyte
antigen mismatching, prednisoloneazathioprine immunosuppression, pretransplantation TB, posttransplant TB
(after 2 years), chronic liver disease (6 years), diabetes
mellitus, posttransplant diabetes mellitus (5 years), and
the presence of other coexisting infections (32). A recent study
by Basiri et al. (33) from Iran also reported an increased risk of
posttransplant TB with increasing numbers of rejection episodes
and duration of pretransplant dialysis, both being associated
with decreased immunocompetence.
A number of the nonrandomized and randomized
studies used radiographic changes suggestive of TB as a definition of an LTBI in addition to clinical indicators such as
history or previous treatment. The radiographic indicators of
previous TB contact include apical cavitation and other parenchymal changes. Studies in the general population suggest
that chest radiograph can appear normal in up to 15% of cases
of TB (1). Furthermore, studies of renal transplant patients
have shown the expected changes of previous TB infection in
less than 60% (33). Therefore, relying on radiographic
changes as an indicator of previous disease may miss a number of cases. The ability of tuberculin skin testing (Mantoux)
in dialysis patients or patients with incipient renal failure,
who are both often anergic, to detect LTBI has shown a low
Currie et al.
701
FIGURE 2. Meta-analysis and forest plot of development of tuberculosis in RCTs of INH prophylaxis for postrenal transplant tuberculosis. Blue boxes represent individual studies, black diamond represents the overall effect size, and horizontal
lines demonstrate 95% confidence intervals. INH, isoniazid; CI, confidence interval; RCT, randomized controlled trial.
FIGURE 3. Meta-analysis and forest plot of development of hepatitis in RCTs of INH prophylaxis for postrenal transplant
tuberculosis. Blue boxes represent individual studies, black diamond represents the overall effect size, and horizontal lines
demonstrate 95% confidence intervals. INH, isoniazid; CI, confidence interval; RCT, randomized controlled trial.
sensitivity and specificity (34, 35). The newer assays that measure the release of INF- may offer a solution to this dilemma.
However, limited studies have addressed the diagnostic performances of an IGRA for LTBI in renal transplant recipients.
Several studies addressed the diagnostic performances of
IGRA for LTBI in renal dialysis patients (36 38) or in patients
with chronic liver disease awaiting transplantation (39, 40).
Manuel et al. (39) showed that Quantiferon-TB Gold (a form
of IGRA) results were associated with high clinical risk factor
for LTBI in liver transplant candidates. In hemodialysis patients (38), no association was found between clinical risk for
LTBI and positivity on T-SPOT.TB (an alternative IGRA
technique) or TST in renal transplant recipients. In contrast,
Quantiferon-TB Gold results showed a statistically significant
association with clinical risk factor for LTBI (38). However,
the evidence for IGRAs in renal transplant recipients and in
those awaiting transplant requires further clarification in
long-term studies. Thus, the definition of patients at risk of
developing TB after renal transplantation must be based on
combination of a relevant clinical history of exposure or
abode in an endemic region as well as available diagnostic
investigation.
In the included studies, both randomized and nonrandomized, a definitive diagnosis of M. tuberculosis infection
was not established by culture in all cases. Often a positive
acid-fast bacilli smear was all that was used, or, on occasions,
a positive response to antituberculous therapy. As the numbers of TB cases involved in both the randomized and nonrandomized studies are small, it is imperative to accurately
diagnose all possible TB and definitively rule out all non-TB
cases to effectively establish the effect of any potential prophylactic therapy. Direct microscopy for acid-fast bacilli is
not specific for M. tuberculosis, and nontuberculous myco-
bacteria cannot be ruled out (1). Any future trials would have
to ensure effective and accurate diagnosis.
The duration of INH prophylaxis was 1 year in the four
RCTs included in this review. The current recommendations
from the European Best Practice Guidelines for Renal Transplantation (41) and the American Society of Transplantation
(42) are for a 9-month course of INH prophylaxis. There is no
evidence suggesting the appropriate time to commence prophlyaxis. The RCTs reviewed commenced INH prophylaxis
in both the pre- and posttransplant periods. However, some
commentators believe that pretransplant would be the most
appropriate and, considering the immunosuppressive state of
renal replacement therapy, that would seem justified when
the recipient is to have a living donor transplant, but is not
practical in those awaiting a deceased donor transplant.
Hepatotoxicity remains the major problem associated
with INH administration, especially in the elderly, the malnourished, and in the alcoholic patient (6), and it generally
occurs within 4 to 8 weeks of therapy. Our meta-analysis
demonstrated no statistically significant difference in the risk
for hepatotoxicity between patients receiving INH and those
not. In patients in whom INH was discontinued because of
significant hepatitis, the majority if them were subsequently
found to have hepatitis C and B infection in the study by
Vikrant et al. (31). Unlike the situation with significant hepatitis, INH was safely continued in patients with mild hepatitis, which resolved over time in both prophylaxis and
control groups and no patient in either group had a positive viral marker. This suggests that significant hepatic dysfunction is likely to be associated with viral hepatitis rather than INH.
A significant incidence of hepatitis was reported in the
study by John et al. (29). Half of the patients who developed
TB were actually ones who had been removed from INH ther-
702 |
www.transplantjournal.com
CONCLUSION
TB is a significant clinical infection in renal transplant
patients. Both nonrandomized and randomized studies point
to a possible benefit of INH chemoprophylaxis in preventing
posttransplant TB. Our meta-analysis of the four RCTs provides further evidence of the potential of INH as TB prophylaxis after renal transplantation in the endemic areas of India
and Pakistan.
Nevertheless, the available evidence is not robust and
there is need for a large multicenter randomized and preferably blinded controlled trial performed in an endemic area,
such as India or Pakistan, Any such study should use, if possible, more recent diagnostic procedures for LTBI. Such studies should also concentrate on helping to determine which
subsets of patients will derive the greatest benefit from prophylaxis. In reviewing the literature, the most significant risk
factors are previous contact or previous TB treatment, and any
prospective trial should focus on this population. However, the
available evidence does suggest that INH prophylaxis should be
considered in potential renal transplant recipients in endemic
areas or in recipients in developed countries who are at risk for
TB infection.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Currie et al.
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.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
703
704 |
68.
69.
70.
71.
www.transplantjournal.com
72.
73.
74.
75.
Koseoglu F, Emiroglu R, Karakayali H, et al. Prevalence of mycobacterial infection in solid organ transplant recipients. Transplant Proc 2001; 33: 1782.
Klote MM, Agodoa LY, Abbott K, et al. Mycobacterium tuberculosis
infection incidence in hospitalized renal transplant patients in the
United States, 1998 2000. Am J Transplant 2004; 4: 1523.
Jie T, Matas AJ, Gillingham KJ, et al. Mycobacterial infections after
kidney transplant. Transplant Proc 2005; 37: 937.
Lezaic V, Radivojevic R, Radosavljevic G, et al. Does tuberculosis after
kidney transplantation follow the trend of tuberculosis in general population? Ren Fail 2001; 23: 97.