Treatment of Tuberculous

Pericarditis

Bongani M Mayosi

Department of Medicine

GROOTE SCHUUR HOSPITAL AND

UNIVERSITY OF CAPE TOWN

Copyright ©2004 BMJ Publishing Group Ltd.

George, S et al. Heart 2004;90:1338-1339

Figure 1 Apical four chamber view of a two dimensional echocardiogram of a patient with tuberculous pericardial effusion showing multiple fibrin strands as linear or band like structures

crossing the pericardial space or protruding from the epicardium or parietal pericardium and exudates. LA, left atrium; LV, left ventricle; Per eff, pericardial effusion; RA, right atrium; RV, right

ventricle.

Alfieri, O., Mayosi B.M. et al. (2014) Exploring unknowns in cardiology

Nat. Rev. Cardiol. doi:10.1038/nrcardio.2014.123

Pericarditis is an important cause of acute heart failure in Africa

Mayosi BM et al. S Afr Med J 2008; 98:36-40

Low Survival Despite TB Rx

0

.25

.5.7

5

1

Cu

mula

tive P

ropo

rtio

n S

urv

ivin

g

0 50 100 150 200Survival time (days)

Clinical HIV No clinical HIV

Overall

17%

26%

40%

Relative risk (95%CI) of death for individuals treated with adjuvant prednisolone or placebo.

Ntsekhe M et al. QJM 2003;96:593-599

© Association of Physicians 2003; all rights reserved

Contemporary Use of Adjunctive

Steroids in TB Pericarditis

Wiysonge C et al. Int J Cardiol 2008; 124:388-390

17

0

21

0

22

13

4

19

10

48

2

29

0

10

20

30

40

50

60

Number of

participants

Cameroon Gauteng W Cape KZN E Cape Nigeria

No adjunctive corticosteroids Adjunctive steroids

The Investigation of the

Management of Pericarditis

Trial

The IMPI Trial

(Pronounced ‘ee-mp-ee’ for Zulu Warriors)

IMPI Trial: Hypothesis

•Hypotheses:

– Anti-inflammatory effect of Steroids may reduce morbidity

(i.e., cardiac tamponade and constrictive pericarditis) and

mortality in TB pericarditis.

– Immunotherapy with Mycobacterium indicus pranii (Mw),

a non-pathogenic environmental organism, may enhance

cure of TB.

•However, the effectiveness and safety of steroids and Mw in

TB pericarditis are uncertain.

IMPI Trial: Primary Objectives

To assess the effect of Prednisolone and Mw in

definite or probable TB pericardial effusion on:

(1) The composite outcome of death, cardiac

tamponade requiring pericardiocentesis, or

constrictive pericarditis;

(2) The incidence of opportunistic infection and

malignancy.

Follow-up data at hospital discharge, at weeks 2, 4 and 6, and

months 3, 6 and thereafter 6 monthly for 2 years, and annually

up to 4 years

Placebo X 5 Doses

PATIENTS WITH DEFINITE OR PROBABLE TUBERCULOUS PERICARDIAL EFFUSION

RANDOMIZATION

PREDNISOLONE

X 6 weeks

PLACEBO

X 6 weeks

M. w X 5 Doses

Placebo X 5 Doses

M. w X 5 Doses

IMPI Trial: Study Design

IMPI Trial Interventions

• Prednisolone or placebo

– 120 mg/day in 1st wk, followed by 90

mg/day in 2nd wk, 60 mg/day in 3rd wk, 30

mg/day in 4th wk, 15 mg/day in 5th wk, and

5 mg/day in 6th wk.

• M. w injection or placebo

– 5 doses of 0.1 ml intradermal injection at

enrolment, then 2 wks, 4 wks, 6 wks, and

3 mo.

IMPI: Organization

19 centres, 8 countries

African Coordinating Center

University of Cape Town

South Africa

International Coordinating Center

Population Health Research Institute

HHS and McMaster University, Hamilton,

Canada

Sponsors: Canadian Institutes for Health Research, Cadila

Pharma, South African Medical Research Council, Lily and Ernst

Hausmann Trust

IMPI Trial: Target Population

Inclusion Criteria

• Age ≥ 18 years

• Pericardial effusion on echocardiography

• Evidence of definite or probable TB pericarditis

• Within 1 week of starting of anti-TB treatment

Exclusion Criteria

• Presence of an alternative cause of pericardial disease

• Pregnancy

• Use of corticosteroids in the previous month

• Allergy to the M. w

IMPI Trial: Baseline Characteristics

Characteristics Prednisolone Placebo

N 706 694

Age in yrs 38.8 38.5

Female % 44.9 43.1

HIV positive % 67.1 67.0

Centesis done % 60.6 60.4

Definite TB % 26.7 26.7

Probable TBP % 71.7 72.9

Non-TB cause % 1.5 0.4

On anti-retrovirals % 14.0 15.0

0.5 2Prednisolone

BetterPlacebo Better

Primary efficacy outcome

Death

Tamponade

Constriction

Hospitalization

Opportunistic infection

Malignancy

AE, Not Hospitalized

Injection side effect

Prednisolone Placebo Hazard Ratio (95% CI) P

N(%) N(%)

168 (23.8) 170 (24.5 ) 0.95 (0.77 - 1.18 ) 0.66

133 (18.8) 115 (16.6 ) 1.15 (0.90 - 1.48 ) 0.26

22 ( 3.1 ) 28 ( 4.0 ) 0.77 (0.44 - 1.35 ) 0.37

31 ( 4.4 ) 54 ( 7.8 ) 0.56 (0.36 - 0.87 ) 0.01

146 (20.7) 175 (25.2 ) 0.79 (0.63 - 0.99 ) 0.04

78 (11.0) 68 ( 9.8 ) 1.16 (0.84 - 1.61 ) 0.36

13 ( 1.8 ) 4 ( 0.6 ) 3.27 (1.07 - 10.03 ) 0.03

171 (24.2) 149 (21.5 ) 1.15 (0.93 - 1.44 ) 0.20

140 (19.8) 137 (19.7 ) 0.98 (0.77 - 1.24 ) 0.84

Effect of Prednisolone on Outcomes

0.5 2Mycbacterium

BetterPlacebo Better

Primary efficacy outcome

Death

Tamponade

Constriction

Hospitalization

Opportunistic infection

Malignancy

AE, Not Hospitalized

Injection side effect

Mycobacterium Placebo Hazard Ratio (95% CI) P

N(%) N(%)

156 ( 25.0) 152 (24.3) 1.03 ( 0.82 - 1.29 ) 0.81

119 ( 19.0) 111 (17.8) 1.07 ( 0.83 - 1.39 ) 0.59

22 ( 3.5 ) 22 ( 3.5 ) 0.99 ( 0.55 - 1.79 ) 0.98

36 ( 5.8 ) 37 ( 5.9 ) 0.97 ( 0.61 - 1.53 ) 0.89

152 ( 24.3) 141 (22.6) 1.09 ( 0.87 - 1.37 ) 0.46

75 ( 12.0) 61 ( 9.8 ) 1.25 ( 0.89 - 1.75 ) 0.20

11 ( 1.8 ) 3 ( 0.5 ) 3.69 ( 1.03 - 13.24 ) 0.03

148 ( 23.7) 149 (23.8) 1.00 ( 0.79 - 1.25 ) 0.97

259 ( 41.4) 18 ( 2.9 ) 18.51 (11.47 - 29.87 ) <0.01

Effect of M. indicus pranii on Outcomes

IMPI Prednisolone: Time To

Constriction

IMPI Prednisolone:

Hospitalization

IMPI: Time To Malignancy

Conclusions

In those with definite or probable TB pericardial effusion:

1. Prednisolone for 6 weeks and M.w for three months had no

significant effect on the combined outcome of death from all

causes, cardiac tamponade requiring pericardiocentesis or

constrictive pericarditis.

2. Both therapies were associated with an increased risk of HIV-

associated malignancy.

3. However, use of prednisolone reduced the incidence of

constrictive pericarditis and hospitalization.

4. The beneficial effects of prednisolone on constriction and

hospitalization were similar in HIV-positive and HIV-negative

patients.