|
Title: |
|
Authors:
|
|
Abstract: Background.
In
sub-Saharan Africa, little is known about what happens to stroke patients when
a neurosurgeon decides that an operation is needed. We followed this pathway
from indication to procedure, hypothesizing that the bottleneck lies in
delivering indications rather than identifying them.
Methods. Prospective
cohort of 250 consecutive adults admitted for imaging-confirmed stroke at
Bouaké University Hospital between 1 January 2022 and 31 December 2023. An
unmet indication was an indication made but not followed by a procedure during
admission. We used Fisher’s exact test and logistic regression (Firth in case
of separation).
Results. A surgical
indication was made in 107 patients (42.8%; 95%CI 36.8–49.0) and surgery
performed in 45 (18.0%): 62 indications (57.9%) went unanswered. Operated
patients underwent 54 procedures (24 external ventricular drainages, 24
decompressive craniectomies, 6 haematoma evacuations); no aneurysm was treated.
For the 37 unmet indications in intracerebral haemorrhage, the reason was
material (money or theatre unavailable) in 16, deterioration or death before
theatre in 7, and unknown in 10. In-hospital mortality was 24.0%: 16.8% without
indication, 33.3% in operated patients, 33.9% with an unmet indication, at
comparable baseline severity. After adjustment, unmet indication was not
significantly associated with mortality (adjusted odds ratio, aOR 2.10
[0.92–4.78]; p=0.077), nor was surgery (aOR 1.60 [0.65–3.91]); the direct
contrast gave an aOR of 1.31 [0.49–3.50]. Without the 7 patients who died or
deteriorated before theatre, crude mortality in the group fell to 27.3% (15/55)
and the aOR to 1.81 [0.76–4.28]. Good functional outcome at discharge (mRS ≤ 2)
concerned 31.2% of patients, with no difference between pathways. Only
subarachnoid haemorrhage predicted non-fulfilment.
Conclusion. The main problem is not spotting indications but honouring them, with access barriers as the main documented reason. Whether an unmet indication costs lives remains to be shown: the adjusted association is not significant, and part of the crude excess comes from patients who died or deteriorated before surgery. DOI: http://dx.doi.org/10.51505/ijmshr.2026.10509 |
|
PDF Download |