49 / 76
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 28, No 2, March/April 2017
AFRICA
115
Discussion
This is the first observational study of acute HF in Botswana,
and it confirms the findings of previous studies that are unique
to the African setting. Contrary to the situation in developed
countries, where HF patients present at a much older age, with
most cases recorded around the seventh and eighth decades of
life, our cohort comprised relatively young patients (mean age 54
years).
21
Similar to previous studies of AHF patients in Africa,
our data have shown that HF affects young and middle-aged
Africans in their most productive period of life.
7,18,22-25
The young age at presentation is not surprising because of
the predominance of non-ischaemic causes of HF among black
Africans, such as hypertension, which often occur early in life
and remain undetected, or if detected are inadequately treated.
In our cohort, ischaemic HF was found in only 5.7% of the
patients. As expected, a significant proportion of our patients
was symptomatic, with left ventricular systolic dysfunction, and
was predominantly men.
7,22,23
More than three-quarters (77.5%)
presented in NYHA functional class III or IV, a finding which is
in agreement with previous reports.
6
One of the most striking features of this cohort was the
relatively high prevalence of HIV-seropositive patients. The
prevalence of HIV seropositivity (33.9%) reported in this study
is far higher than in the previous largest HF study among
HIV-poitive patients in South Africa.
20
The observed high
prevalence of HIV in our cohort may partly be explained by a
relatively higher prevalence of HIV in Botswana than in South
Africa.
26
Botswana has the world’s third-highest HIV infection
rate in the world after Swaziland and Lesotho, with an adult
prevalence rate of about 24.4% in 2012.
26
This high prevalence of HIV infection in our cohort provides
additional evidence of the confluence of non-communicable and
infectious diseases as co-morbidities amongHFpatients inAfrica.
HIV-associated cardiomyopathy, pulmonary hypertension,
pericardial disease and accelerated atherosclerosis are known
to occur frequently in HIV-positive patients.
27
Our data reaffirm
that hypertension, diabetes, renal failure and anaemia are still
common and contribute significantly to the aetiological burden
of HF in Africa. These co-morbidities are not only the likely
aetiologies of HF, but are also factors that affect the clinical
course of the disease, and should be concurrently addressed at
the time of admission.
28
Consistent with previous reports from sub-Saharan Africa,
hypertensive heart disease, dilated cardiomyopathy, cor pulmonale
(right heart disease), peripartum cardiomyopathy, pericardial
disease and valvular heart disease were the common causes of HF
in our cohort.
6,7,18,20,22,23,25,27
Ischaemic HF, however, was uncommon
in our cohort, and patients with pericardial disease were more
likely to be HIV positive than those with other types of HF.
In our study, peripartum cardiomyopathy was the third
commonest aetiology of HF among females. Although the mean
age of patients with peripartum cardiomyopathy was similar
to previous studies in Africa, the reported prevalence (9%) was
lower than that reported in other African studies, where 13 to
60% of admissions for HF in females were related to peripartum
cardiomyopathy.
25,29,30
Nevertheless, consistent with the above
studies, it is clear that peripartum cardiomyopathy is one of the
important causes of HF among female patients in Africa.
The present study demonstrates that patients with HF are at
risk for adverse clinical outcomes, which ranks HF among the
major causes of death of cardiovascular origin in Africa.
6
The
in-hospital mortality rate (10.9%) in our cohort is similar to the
nine to 12.5% case fatality rates reported elsewhere in Africa, but
was higher than reports from high-income countries.
6,21,24,31
Generally, African-Americans have been reported to have
a lower in-hospital mortality rate compared to Caucasians,
and the same results could be expected among HF patients in
Africa if the quality of care was similar.
31
However, in resource-
poor settings such as ours, there is inconsistent availability
of HF medications, limited access to intensive care services,
Table 2. Outcomes and discharge medications of patients admitted
with acute heart failure at Princess Marina Hospital
Discharge medication
Number (%)
Diuretics
148 (86)
Beta-blockers
124 (72.1)
ACE inhibitors
116 (67.40
Angiotensin receptor blockers
10 (5.8)
Spironolactone
103 (59.9)
Digoxin
38 (22.1)
Nitrate
8 (4.7)
Hydralazine
3 (1.7)
Outcome
Median length of hospital stay (IQR)
9 (5–15)
In-hospital mortality (
n
= 193)
21 (10.9)
30-day mortality
(
n
= 190)
28 (14.7)
90-day mortality (
n
= 182)
47 (25.8)
180-day mortality (
n
= 181)
56 (30.9)
Table 3. Associations between demographic and medical
characteristics of the patients and mortality outcome at the end
of the follow-up period (six months post-admission)
Outcome
Characteristic
Died (
n
= 56)
Survived (
n
= 125)
p-
value
Age (years) mean (SD)
59.8 (16.5)
51.93 (16.547)
0.004
†
Male gender,
n
(%)
32 (57.1)
65 (52)
0.60*
Medical history,
n
(%)
Hypertension
27 (48.2)
72 (57.6)
0.200*
Diabetes
7 (12.5)
21 (16.8)
0.515*
Rheumatic heart disease
4 (7.1)
10 (7.9)
1.00*
Ischaemic heart disease
3 (5.4)
15 (12)
0.281*
Stroke/TIA
4 (7.1)
13 (10.4)
0.591*
Atrial fibrillation
7 (12.5)
11 (8.8)
0.431*
HIV positive
16 (28.6)
41 (32.8)
0.306*
Clinical history
Median SAP (mmHg)
120.5 (108–131.4)
120.0 (101.5–141.3)
0.887
‡
Mean DAP (mmHg)
75.3 (67–79.9)
74 (66.0– 85.3)
0.878
‡
LVEF (%)
41.9
±
20.7
41.8
±
20.2
0.975
†
Haemoglobin (g/dl)
11.2
±
3.1
12.4
±
2.7
0.010
†
MCV (%)
87.9
±
9.5
89.3
±
10.9
0.337
†
eGFR (ml/min/1.73 m
2
),
median (IQR)
70.95 (41.7–95.6)
84.8 (55.9–113.6)
0.043
‡
Sodium (mEq/l)
132.0
±
8.0
135.1
±
6.1
0.010
†
LOS (days), median (IQR)
11 (6.0–19.8 )
7 (5–12.3)
0.005
‡
Creatinine (µmol/l),
median (IQR)
116.5 (80.5–149.0)
96.0 (66.8–130.8)
0.041
‡
Urea (mmol/l), median (IQR)
11.4 (6.6–18.9)
7.1 (4.7–12.0)
0.002
‡
NT-proBNP (pg/ml),
median (IQR)
6597 (4340.0–18810.3) 2739.0 (998.5–4656.0)
<
0.001
‡
*
Chi- squared;
‡
Kruskal–Wallis,
†
Student’s
t
-test.
TIA, transient ischaemic attack; SAP, systolic arterial pressure; DAP, diastolic
arterial pressure; LVEF, left ventricular ejection fraction; MCV, mean corpuscular
volume; eGFR, estimated glomerular filtration rate; LOS, length of stay; NT-proB-
NP, N-terminal pro-brain natriuretic peptide.