Page 343 - HIVMED_v21_i1.indb

Page 2 of 5 Original Research

The initial (induction phase) management of HIV/CCM history of previous CCM, data of concurrent tuberculosis
requires the following: (1) combination antifungal therapy (TB) and details of the index admission. Patients’ folders
including IV amphotericin B and oral flucytosine (first week were checked for the results of therapeutic LPs and to
only) and high-dose oral fluconazole 1200 mg daily (second document the indications for the procedure. The National
week), after that an 8-week consolidation phase of oral Health Laboratory Services (NHLS) computer records
fluconazole 800 mg daily, (2) control of raised intracranial were accessed for admission bloods, CD4 count, HIV viral
pressure (rIP) with therapeutic lumbar punctures (LPs) to load (VL) and CSF results. Identifying patient material
maintain the ‘opening-pressure’ (CSF-OP) at < 25 cm of water was anonymised at the time of the collection and storage
and (3) minimising the risk of immune reconstitution of data.
inflammatory syndrome (IRIS) by delaying the initiation of
ART until 4 to 6 weeks after the start of antifungal Statistical analysis
therapy. 8,9,10,11,12
Continuous variables are reported as mean (± standard
deviation, [s.d.]) when normally distributed and as median
In the absence of a spinal manometer, the SA guidelines for
the prevention, diagnosis and management of CCM (interquartile range, IQR) when not normally distributed.
recommend performing an LP to remove 20 mL – 30 mL of Discrete data are presented as number and percentages.
12
CSF if the symptoms and signs of rIP are present. Pearson’s chi-square test and corresponding 95%
Confidence Intervals (CI) were used to calculate the
However, in clinical practise only 23% – 30% of CCM mortality difference between the study and comparator
patients with signs and symptoms receive ‘therapeutic’ groups. A p-value of < 0.05 defined significance. Patients
LPs. 13,14 A symptom guided approach has the potential to who died or left hospital within the first 7 days were
miss asymptomatic patients who might benefit from excluded from the mortality analysis. Statistical analyses
therapeutic LPs. were performed by using IBM SPSS Statistics for Macintosh
version 24.0.
We hypothesised that four or more LPs in the first 7 days of
treatment could facilitate CSF drainage and reduce in-
hospital mortality in PLWH and HIV/CCM in a resource Ethical consideration
constrained setting where there are no spinal manometers. Approval to conduct the study was obtained from Walter
We, therefore, conducted a single centre retrospective Sisulu University Human Research Committee. Ethical
cohort study to determine the impact on in-hospital clearance number: 027/2018.
mortality of four or more LPs in the first 7 days of antifungal
therapy compared with PLWH/CCM who received fewer Results
than four LPs.
Clinical characteristics of the study population
Methods From 01 January 2016 to 31 December 2016, a total of
Study design 116 patients received a diagnosis of CCM. After exclusion of
11 (two were younger than 18 years, two had missing hospital
The study was designed as a retrospective cohort review of files/records and seven demised or left the hospital before
PLWH/CCM admitted to the department of medicine at the 7 days of hospitalisation), 105 patients were available for
Dora Nginza Hospital from 01 January 2016 to 31 December study analysis (Figure 1).
2016. The Dora Nginza Hospital is a district hospital located
in the Zwide township of the Nelson Mandela Bay The mean age of patients was 39.4 (s.d. ± 9.7) years. A total of
Municipality (Port Elizabeth), SA. The Nelson Mandela Bay 65.2% were male. All the patients were PLWH and had a
region has a population of 1 152 115 and an unemployment median CD4 count = 37 (IQR, 13–77) cells/µL. Human
15
rate of 36.6%. The internal medicine department consists
of a 120-bed unit without access to intensive or high care
services and with limited access to radiological imaging 116 patients diagnosed and
apart from plain chest radiography. treated for CCM from
01/01/2016 – 31/12/2016
Study population Excluded patients
2 patients younger than 18 years of age
Clinical notes, discharge summaries and death notification 2 patients with missing files
registries were reviewed to identify patients who received a 7 patients who demised or left the hospital
before 7 days
primary or secondary diagnosis of CCM.
105 patients available for
Data collection the final analysis
A standardised data collection form was prepared. This CCM, cryptococcal meningitis.
included patient demographic details, comorbidities, FIGURE 1: Flow diagram showing patients enrollment in the study.

http://www.sajhivmed.org.za 335 Open Access