[Downloaded free from http://www.sjkdt.org on Tuesday, July 24, 2018, IP: 36.72.231.

171]

Saudi J Kidney Dis Transpl 2017;28(6):1362-1368

© 2017 Saudi Center for Organ Transplantation Saudi Journal
of Kidney Diseases
and Transplantation

Renal Data from the Arab World

The Diagnosis of Tuberculosis in Dialysis Patients

Hela Jebali1, Sana Barrah1, Lamia Rais1, Rania Kheder1, Nihal Khouja2, Salma Nadia Mhiri2,
Majed Beji2, Rim Abdelmalek3, Hanene Tiouiri3, Wided Smaoui1, Soumaya Beji1, Fethi Ben
Hmida4,5, Lilia Ben Fatma1, Mohamed Karim Zouaghi1

Departments of 1Nephrology, Dialysis and Transplantation, 2Pneumology and 3Infectiology, La

Rabta Hospital, 4Research Laboratory of Kidney Diseases (LR00SP01), Charles Nicolle Hospital,
5
Faculty of Medicine, Tunis El Manar University, Tunis, Tunisia

ABSTRACT. The incidence of tuberculosis (TB) is high in patients undergoing chronic dialysis
than it is in the general population. The diagnosis of TB is often difficult and extrapulmonary
involvement is predominant. This study investigates the spectrum of clinical presentations and
outcome in dialysis patients during a nine-year period. TB was diagnosed in 41 patients. Anti-TB
drugs, adverse effects of therapy, and outcome were noted. Thirty-eight patients (92.6%) were on
hemodialysis and three were on peritoneal dialysis (7.3%). The mean age at diagnosis was 50.8
years and the male/female ratio was 1.16. Four patients had a history of pulmonary TB.
Extrapulmonary involvement was observed in 32 (78 %) patients. The bacteriological confir-
mation was made in 41.46% and histological confirmation was made in 26.83%, and in the rest,
the diagnosis was retained on the criterion presumption. Nineteen patients (46.34%) developed
adverse effects of antitubercular drugs. Eight patients (19.51%) died during the study from TB or
adverse effects of treatment. Low urea reduction ratio and female sex were associated with poor
prognosis in our study. The clinical manifestations of TB in patients on dialysis are quite non-
specific, making timely diagnosis difficult, and delaying the initiation of curative treatment,
which is a major determinant of the outcome.

Introduction Worldwide, TB infection in dialyzed patients

Tuberculosis (TB) remains a major health
problem, particularly in endemic regions.1
Correspondence to:
Dr. Sana Barrah,
Department of Nephrology, Dialysis and
Transplantation, La Rabta Hospital,
Tunis, Tunisia.
E-mail: barrahsana@gmail.com
ranges from 5%–25% and a 6.9–52.5-fold risk
of TB is reported as compared to the general
population.2,3 In a previously published
Tunisian prospective study, the authors found
10% as a rate of active TB among hemo-
dialysis (HD) patients; this rate represents 15
times of the general population TB incidence
in our country (23/100,000).4,5 Several factors
related to dialysis contribute to impairment of
cell-mediated immunity, including nutritional
[Downloaded free from http://www.sjkdt.org on Tuesday, July 24, 2018, IP: 36.72.231.171]
The diagnosis of tuberculosis in dialysis patients
abnormalities (reduction of protein intake and
micronutrient deficiency), iron overload, and
bio-incompatibility of the dialysis system. In
addition, frequent hospital contact, older age
of dialysis recipients, diabetes mellitus, low
body mass index (BMI), and the use of immu-
nosuppressive drugs are additional factors
which contribute to the higher prevalence of
TB in these patients.6-8 The diagnosis of TB
disease is often delayed in dialysis patients due
to nonspecific symptoms.2,9 This causes the
delay in initiation of TB treatment which
adversely patient outcomes.10,11
There are limited data about the prevalence
and the patient characteristics of TB cases
among dialysis patients in Tunisia. The aim of
this study was to evaluate clinical characte-
ristics, diagnostic tests, treatment modalities,
and outcomes of TB among patients with end-
stage renal disease undergoing chronic dialysis.
Patients and Methods
This was a retrospective study and included
all patients with diagnosed with TB under-
going either HD or peritoneal dialysis (PD)
over a period of nine years between April
2007 and July 2016. We identified 41 cases of
TB. We analyzed the clinical and laboratory
features at the time of TB diagnosis including
epidemiological characteristics and signs sug-
gestive of a pulmonary or extrapulmonary
involvement.
Diagnosis of TB was based on the following
criteria: (1) positive culture of any specimen or
a tissue for Mycobacterium tuberculosis and
(2) histopathological evidence of M. Tubercu-
losis (granulomatous with caseating necrosis).
In cases without bacteriological or pathological
confirmation, TB was defined as clinical (un-
explained fever, impaired general condition),
biological (high C-reactive protein (CRP),
hypoalbuminemia, lymphocyte predominance,
and exudative effusion of peritoneal, peri-
cardial, and pleural fluid), imaging (pleural
lesions, pulmonary lesions) and/or histopatho-
logical (noncaseating granulomas), and/or po-
sitive tuberculin skin test (TST) and/or history
of active TB, and the exclusion of other
1363
diseases. TST was carried out by the intra-
dermal method and was read 48-72 hours later.
Positivity was defined as an induration dia-
meter ≥10 mm. Malnutrition was defined as
BMI <20 kg/m2 and hypoalbuminemia <30
g/L. Iron overload was defined as hyper-
ferritinemia >500 µg/L.
Treatment response was evaluated according
to improvement of the general condition, func-
tional signs, radiological signs, and biological
data.
Statistical Analysis
Data were analyzed using the Statistical
Package for the Social Sciences for Windows
version 11.0 (SPSS Inc., Chicago, IL., USA).
The results were expressed in terms of number
of cases and/or percentage for categorical
variables and in terms of averages for quanti-
tative variables. We used Chi-square test for
the comparison of percentages and test ana-
lysis of ANOVA variances for comparison of
averages. P <0.05 was considered statistically
significant.
Results
Forty-one cases of TB were diagnosed during
the study period. The average age of this popu-
lation was 50.8 years ± 15.23 (22–85 years).
Twenty-two patients (53.6%) were male and
19 (46.3%) were female. Of these, 38 patients
(92.6 %) were on HD and three (7.3 %) on PD.
The mean interval between the onset of
dialysis and the time of diagnosis of TB was
about 1.7 years (range: 1 week–9.2 years).
Comorbidities, primary renal disease, and
corticoids and/or immunosuppressive therapy
are summarized in Table 1. Four patients
(9.7%) had a prior history of a pulmonary TB.
Thirty-two patients (78%) have been vacci-
nated against TB. The most common clinical
features were weight loss (97.5 %), anorexia
(97.5 %), and fever (51.2 %). The main biolo-
gical characteristics of our patients are summa-
rized in Table 2. High CRP (>10 mg/L) was
noted in 39 patients (95.12%). Hypoalbumi-
nemia (<30 g/L) was noted in 12 cases (29.2%).
[Downloaded free from http://www.sjkdt.org on Tuesday, July 24, 2018, IP: 36.72.231.171]

1364 Jebali H, Barrah S, Rais L, et al

Table 1. Epidemiological features

Number (%)
Age (years) 50.8 years (22–85 years)
Sex ratio (M/F) 22/19( 1.15)
Dialysis modality (HD/PD) 38/3 (92.6/7.3)
Primary renal disease
Glomerulonephritis 15 (36.5%)
Tubulointerstitial nephritis 9 (21.9%)
Nephroangiosclerosis 2 (4.8%)
Unknown origin 15 (36.5%)
Comorbidities
Diabetes 11(26.8%)
Hypertension 23(56.1%)
Neoplasia 3 (7.3 %)
Corticoids and/or immunosuppressive therapy 5(12.2%)
M: Male, F: Female, HD: Hemodialysis, PD: Peritoneal dialysis.

Hematological abnormalities included anemia
in 85.3% (35 patients), hyperleukocytosis in
21.95% (9 patients), and leukopenia in 4.8%
of cases (2 patients). Six patients had hyper-
calcemia and parathyroid hormone levels
below 150 pg/mL (14.63%). An analysis of
peritoneal, pericardial, and pleural fluids
showed lymphocyte predominance and exuda-
tive effusion in all cases. TST was performed
on all patients before initiation of therapy. A
positive result was seen in 15 patients (36.5%).
The clinical presentation of TB was varied in
our study population. It was unifocal in 29
patients (70.73%), bifocal in nine patients
(21.95%), and three patients had multisystem
involvement (7.3%). Extrapulmonary TB was
observed in 78% of cases (32 patients).
Isolated pulmonary TB was noted in 21.95%
of cases (9 patients), whereas eight patients
presented pulmonary TB associated with
another placement (Tables 3 and 4). The mean
interval between the onset of symptoms and TB
diagnosis was about 113.52 days (30–360
days). Confirmation of TB was done in 68.3%
of cases (28 patients), 17 cases (41.46%) on
bacteriological findings, and 11 cases (26.83%)
on histological findings (Table 3). In the
remaining cases (13 patients, 31.7%), the diag-
nosis of TB was made on strong clinical, labo-
ratory, histopathological, and/or radiological
suspicion.
Thirty-nine patients were treated with rifam-
picin, isoniazid, ethambutol, and pyrazinamide,
whereas two patients received rifampicin, iso-
niazid, and pyrazinamide. Planned for all
patients involved the use of all antitubercular
drugs in the first two months, isoniazid and
rifampicin treatment continued thereafter. The
mean duration of treatment was 198.3 days (4–
365 days). Nonadherence of treatment was
observed in 34.1% of cases (14 patients). The
determination of acetylator phenotype was
carried out on 25 patients. Twenty-one patients
were slow acetylators, whereas four patients

Table 2. Biological features.

Parameters Minimum Maximum Average
C-reactive protein (mg/L) 5 298 65.8±55.4
Leukocytes 2530 16900 7494 ±3029
Hemoglobin (g/L) 4.6 11.8 8.4±1.7
PTH (pg/mL) 5 2238 397.6±482.3
Calcemia (mg/L) 76 133.6 97.7±13.1
Albuminemia (g/L) 16.2 43.20 30.9±6.7
Urea reduction ratio 47.9 77 64.7±7.4
Ferritin 20 2450 661.1±623.7
[Downloaded free from http://www.sjkdt.org on Tuesday, July 24, 2018, IP: 36.72.231.171]

The diagnosis of tuberculosis in dialysis patients 1365

Table 3. TB diagnosis.
TB site Number Bacteriological Histological Presumptive
HD PD
localization (%) diagnosis diagnosis criteria
Pulmonary 17 (41.4) 17 0 9 1 7
Urinary 9 (21.9) 7 2 7 1 1
Pleural 9 (21.9) 8 1 0 3 6
Lymph node 7 (17) 7 0 0 6 1
Peritoneum 5 (12.2) 5 0 1 0 4
Bone 3 (7.3) 3 0 0 0 3
Hematopoietic 3 (7.3) 3 0 0 0 3
Pericardial 1 (2.4) 1 0 0 0 1
TB: Tuberculosis, HD: Hemodialysis, PD: Peritoneal dialysis.

had fast phenotype. The mean dose of isonia- cessation of isoniazid in one case. Several
zid for slow group was 130 mg/day (ranging patients presented more than one adverse
from 50 mg and 200 mg/day), and the mean effect. According to acetylator phenotype, 11
dose for fast group was 200 mg/day (ranging slow acetylators developed adverse effects
from 150 mg and 300 mg/day). Nineteen related to isoniazid and only one fast acety-
patients developed adverse effects (46.34%). lator presented digestive disorders.
Cutaneous manifestations of drug allergy and The mean period of follow-up was 9.32
hypersensitivity occurred in seven patients; all months (3 days to 3 years). Clinical response
drugs were stopped and restarted gradually was achieved in 82.9% (34 patients) (Table 4).
with good tolerance. Hepatotoxicity developed One patient developed reactivation of pulmo-
in six patients (cytolysis in 2, cholestasis in 2, nary TB after cessation of treatment. Fifteen
and acute liver failure in 2), requiring cessation patients (36.58%) died during the treatment
of all drugs in two patients, and adjustment of from TB or adverse effects related to the treat-
doses in two cases. Hematologic abnormalities ment in eight patients (multiorgan failure in 5,
were observed in six patients (aggravation of acute liver failure in 1, isoniazid intoxication
anemia in 4, leukopenia in 1, and thrombo- in 1, and acute respiratory failure in 1 patient,
cytopenia in 1 patient). Five patients presented respectively). The remaining deaths were related
with reversible digestive disorders and required to cardiovascular and/or cerebrovascular events.
Table 4. Comparison with the literature.
Our study Hassine (Tunisia) Malik (KSA) Sen (Turquie)
Hemodialysis 92.6 % 100% 77% 77.8%
Peritoneal dialysis 7.3% 0 23% 22.2%
Mean age 50.8 42 55.65 52.33
Sex ratio 1.16 2 2 1.6
Mean interval for diagnosis
31 17.8 24 19.5
(months)
Diabetes 26.8% 16.6% 29.2% –
Weight loss 97.5 % 100% 18.5% –
Anorexia 97.5 % 50% 14.58% –
Fever 51.2 % 66.6% 66.6% –
Tuberculin skin test positive 36.5% 33.4% 35% –
Pulmonary 41.4% 33.3% 48% 22.2
Pleural 21.9% 0% _ 28.6
Lymph node 17% 16.6% 23% 35.7
Urinary 21.9% 0% 6.25% –
Peritoneum 12.1% 50% 31.25% –
Clinical response 82.9% 80% 72.92% 89%
[Downloaded free from http://www.sjkdt.org on Tuesday, July 24, 2018, IP: 36.72.231.171]

1366 Jebali H, Barrah S, Rais L, et al

Table 5. Prognosis factor.

Prognosis Factor P Odds Ratio
Female sex 0.009 6.4
Diabetes 0.15 0.357
Hypertension 0.79 1.19
Neoplasia 0.27 0.26
Corticoids 0.42 2.54
Immunosuppressive therapy 0.42 2.54
CRP >10 mg 0.28 1.62
Hyperleukocytosis 0.95 0.95
Leukopenia 0.058 0.33
Lymphopenia 0.11 2.5
Anemia 0.28 0.3
Hyperferritinemia 0.46 2.25
Hypercalcemia 0.37 0.4
PTH <150 pg 0.71 0.72
Albumin<30 g 0.69 1.4
Hypocholesterolemia <1.3 0.10 0.16
URR <65% 0.026 0.08

For the patients treated according to the

criteria of presumption, clinical response was
achieved in 84.61% (11/13) of cases. Overall
survival at one year was 62% (Figure 1).
Univariate analysis showed that the female sex
(P = 0.009), and urea reduction ratio <65% (P
= 0.026) were negative prognosis factors in
our patients (Table 5).

Discussion

In this study, we diagnosed 41 active TB

patients undergoing chronic dialysis treatment
for a period of nine years. Our data suggested
that TB diagnosis was often difficult in these
patients. In fact, the mean interval between the
onset of symptoms and TB diagnosis was
about 113.52 days in our patients, and diag-
nosis of TB was made on presumption criteria
in 31.7% of cases.
In our study, 26.83% of patients were diabetic
and 12.19% were on immunosuppressive treat-
ment. The majority of our patients had iron
overload and malnutrition
In our study, 9.7% of patients had a history of
active pulmonary TB. This frequency varies
according to the authors between 0% and 27%.
TB infection in chronic dialysis patients is
probably secondary to reactivation of latent
TB.12 A negative TST due to its poor sensitivity
Figure 1. Survival curve.
in these patients cannot be used to eliminate
the possibility of latent or active TB.13-16 In our
study, only 36.5% of patients with a clinical
diagnosis of TB presented positive TST.
In this study, impaired general condition and
fever were noted in 97.56% and 51.22% of
cases, respectively, while increasing CRP and
anemia were noted in 95.12% and 85.3%,
respectively. Lymphocyte-predominant exudates
were presented in all cases with pleural,
pericardial, or peritoneal TB in our series. On
[Downloaded free from http://www.sjkdt.org on Tuesday, July 24, 2018, IP: 36.72.231.171]
The diagnosis of tuberculosis in dialysis patients
the other hand, hypercalcemia could be a sign
of TB in dialysis patients caused by an
increased form of Vitamin D released by gra-
nulomas. In our study, six patients developed
hypercalcemia (14.63%).17
In our study, urinary, pleural, and lympha-
denitis were the more frequent location,
whereas bifocal or multifocal TB occurred in
12 patients. In our study, the bacterial diag-
nosis was of TB which was positive in only 17
(41.46%) of pulmonary cases (Table 3). In our
study population, histological examination was
positive in 11/41 patients (26.83%) (Table 3).
Dialysis patients do not seem to have the
capacity to develop caseous lesions or even
granulomatous lesions,18,19 and the absence of
bacteriological or histological confirmation
should not exclude TB infection.20,21 There-
fore, initiation of antitubercular therapy must
be discussed if patients presented with unex-
plained debilitating symptoms, for example,
nonspecific fever, anorexia fatigue, cough, or
weight loss associated with specific signs such
as lymph node enlargement and pleural or
peritoneal effusion lesions.20-23 Thirteen pa-
tients in our series were treated according to
presumptive criteria with good response in the
majority of cases.
Treatment of TB in dialysis patients may be
complicated by an increased risk of toxicity
from antitubercular drugs.24 There are many
possible adverse effects including hepatitis,
skin rash, digestive disorders, and hematologic
complications. Hepatotoxicity and allergic
manifestations were the most frequent adverse
effects in our series. Our result showed that the
mortality of dialysis patients presenting TB
remain high. Delay in diagnosis and initiation
of treatment, nonobservance of treatment, and
the frequency of serious adverse effects of
antitubercular drugs were associated with worse
prognosis. The low urea reduction ratio was
identified as a negative prognostic factor in
our study. This finding was explained pro-
bably by the association between dialysis ade-
quacy and nutritional status and inflammation.
Due to the retrospective nature of this study,
we could not gather more information regar-
ding risk factors, additional comorbidities, or
1367
diagnostic procedures.
Conclusion
Our study highlights the predominance of
extrapulmonary TB in dialysis patients and the
frequency of adverse effects from antituber-
cular treatment and death related to TB or
treatment. This study demonstrates the chal-
lenges posed for diagnosing TB in this popu-
lation despite high index of suspicion.
References
1. Kayabasi H, Sit D, Kadiroglu AK, Kara IH,
Yilmaz ME. The prevalence and the charac-
teristics of tuberculosis patients undergoing
chronic dialysis treatment: Experience of a
dialysis center in Southeast Turkey. Ren Fail
2008;30:513-9.
2. Ates G, Yildiz T, Danis R, et al. Incidence of
tuberculosis disease and latent tuberculosis
infection in patients with end stage renal
disease in an endemic region. Ren Fail 2010;
32:91-5.
3. Sav T, Tokgoz B, Sipahioglu MH, et al.
Extrapulmonary tuberculosis in a hemodialysis
patient with unusual clinical presentation. Int
Urol Nephrol 2010;42:223-6.
4. Hachicha J, Jarraya A. High incidence of
tuberculosis in chronic dialysis patients in
developing countries. Nephron 1989;52:189.
5. Ministry of Public Health. Basic Health Care
Management. Support Guide for Tuberculosis
in Tunisia; 2014.
6. Rieu PH, Touré F. Infections and immuno-
suppression during uraemia. News Nephrol
Flammarion 2006;229-50.
7. Christopoulos AI, Diamantopoulos AA,
Dimopoulos PA, Goumenos DS, Barbalias
GA. Risk factors for tuberculosis in dialysis
patients: A prospective multi-center clinical trial.
BMC Nephrol 2009;10:36.
8. Klote MM, Agodoa LY, Abbott KC. Risk
factors for Mycobacterium tuberculosis in US
chronic dialysis patients. Nephrol Dial
Transplant 2006;21:3287-92.
9. Borrajo Prol M, Perez Melon C, Novoa EF, et
al. Tuberculous peritonitis in peritoneal
dialysis. Nefrologia 2009;29:170-2.
10. Lee SS, Chou KJ, Su IJ, et al. High prevalence
of latent tuberculosis infection in patients in
end-stage renal disease on hemodialysis: Com-
[Downloaded free from http://www.sjkdt.org on Tuesday, July 24, 2018, IP: 36.72.231.171]
1368
parison of Quantiferon-TB GOLD, ELISPOT,
and tuberculin skin test. Infection 2009;37:96-
102.
11. Kazancioglu R, Ozturk S, Gursu M, et al.
Tuberculosis in patients on hemodialysis in an
endemic region. Hemodial Int 2010;14:505-9.
12. Taskapan H, Utas C, Oymak FS, Gülmez I,
Ozesmi M. The outcome of tuberculosis in
patients on chronic hemodialysis. Clin Nephrol
2000;54:134-7.
13. Dinnes J, Deeks J, Kunst H, et al. A systematic
review of rapid diagnostic tests for the
detection of tuberculosis infection. Health
Technol Assess 2007;11:1-96.
14. Passalent L, Khan K, Richardson R, et al.
Detecting latent tuberculosis infection in
hemodialysis patients: A head-to-head compa-
rison of the T-SPOT.TB test, tuberculin skin
test, and an expert physician panel. Clin J Am
Soc Nephrol 2007;2:68-73.
15. Fang HC, Chou KJ, Chen CL, et al. Tuberculin
skin test and anergy in dialysis patients of a
tuberculosis-endemic area. Nephron 2002;91:
682-7.
16. Foster R, Ferguson TW, Rigatto C, et al. A
retrospective review of the two-step tuberculin
skin test in dialysis patients. Can J Kidney
Health Dis 2016;3:28.
17. Ko YC, Lee CT, Cheng YF, et al. Hypercal-
caemia and haemophagocytic syndrome: Rare
concurrent presentations of disseminated tuber-
Jebali H, Barrah S, Rais L, et al
culosis in a dialysis patient. Int J Clin Pract
2004;58:723-5.
18. Langlois S, Barré P. The different modes of
presentation of tuberculosis in hemodialysis
patients. Union Med Can1983;112:1084-7.
19. Ferrara E, Lemire J, Grimm PC, et al. Myco-
bacterial peritonitis in pediatric peritoneal
dialysis patients. Pediatr Nephrol 2004;19:114-7.
20. Vadivel N, Tucker JK, Trikudanathan S, Heher
E, Singh AK. Tuberculous peritonitis: A race
against time. Kidney Int 2006;70:969-72.
21. Ninet B, Roux-Lombard P, Schrenzel J,
Janssens JP. New tests for the diagnosis of
tuberculosis. Rev Mal Respir 2011;28:823-33.
22. American Thoracic Society, Centers for
Disease Control and Prevention, Infectious
Diseases Society of America. American
thoracic society/Centers for disease control and
prevention/Infectious diseases society of
America: Controlling tuberculosis in the
United States. Am J Respir Crit Care Med
2005;172:1169-227.
23. Mouhamadou MC, Rachid EK, Yaya K, Sidy
MS, Ahmed TL. Tuberculosis among chronic
hemodialysis patients: A Senegalese single
center experience. Open J Nephrol 2015;5:
117-22.
24. Tahar G, Goucha-Louzir R, Rachid LM.
Tuberculosis in children undergoing hemo-
dialysis. Int J Nephrol Renovasc Dis 2010;3:
47-50.