CHARACTERISTICS AND OUTCOME OF STROKE PATIENTS WITH CEREBROVASCULAR ACCIDENT AT THE INTENSIVE CARE UNIT OF A TERTIARY HOSPITAL IN NIGERIA

Abstract

Background: Patients with severe stroke defined as NIHSS score >17 constituting about 15-20% of cerebrovascular accident require admission into the Intensive Care Unit (ICU). However the benefit of ICU admission for stroke patients remains controversial.

Aim & Objectives: To determine the characteristics and outcome of patients with cerebrovascular accident managed at the Intensive Care Unit of University of Benin Teaching Hospital.

Methodology: Demographic characteristics, clinical features and course, treatment options and outcome of all stroke patients admitted in ICU from January 2002 to January 2012 were retrieved from the hospital records and analyzed. A patient before and after each stroke patient were selected as controls for the study. Primary outcome variable was ICU mortality, type of stroke whether ischemic or haemorrhagic, duration of stay, whether patients were transferred from the medical/stroke ward or from the accident and emergency department of the hospital.

Results: A total of thirty six (36) stroke patients were admitted into the ICU within the study period accounting for 5.6% of the total ICU admissions. The male: female ratio is 2:1 and patients aged >60 years accounted for 55.6%. Stroke patients admitted into ICU had a mortality rate of 77.8%. Patients with severe stroke admitted into the ICU were 4 times more likely to die compared to non-stroke patients in the ICU (p=0.002, OR=4.472). However, severe stroke had no significant impact on duration of ICU stay (p=0.454, OR=1.464).

Conclusion: Stroke patients have a high mortality in the intensive care unit that is independent on the type and route of admission. Provision of the support equipment and instruments required for high dependency service in the intensive care unit and early admission should improve the outcome.

Introduction

Stroke patients are often managed in the stroke ward. However patients with severe stroke, defined as National Institute of Health Stroke Scale (NIHSS) score >17 constituting about 15-20% will require admission into the Intensive Care Unit1. In the United State of America for example, severe stroke accounts for about 3-6% of admissions in the ICU.2^-5 Riachy et al reported a similar finding, where severe stroke accounted for 7% of admission into their medical ICU6. About 2 decades ago, Odusote reported that stroke accounted for 2.83-4.52% of total death in Nigeria7, while Obiako et al8 observed that acute stroke accounted for 33% of medical coma at the University College Hospital Ibadan, Nigeria

Despite this, the benefit of ICU admission of stroke patients remains controversial. It has been previously reported that there was no significant reduction in mortality and morbidity following admission of stroke patients in the ICU9^-12. In most cases, ICU admission has been found only to prolong patients’ inevitable demise after an episode of severe stroke13s.

Mortality following an episode of severe stroke in the ICU varies from one centre to another. In a study by Marik14 at University of Massachusetts, USA where 40 patients were studied, ICU mortality was found to be 28%, with 15% in hospital deaths. Burtin and co-workers15 Centre Hospitalier, Universitaire de Nancy, France reported a higher ICU mortality rate of 73% and one year mortality rate of 92%. Patients with severe stroke suffer a high mortality despite care in the intensive care unit. Many of these patients are admitted into the unit for various reasons ranging from unstable blood pressure that is difficult to control, arrhythmias, myocardial infarction, and impaired level of consciousness or a massive hemispheric or cerebellar infarct1.

Previously, Adudu and colleagues16 conducted an 18 year (1985-2003) survey of mortality and morbidity pattern among neurological patients in ICU in our centre. A total of 187 patients were studied; head injury and tetanus constituted 63.7% and 13.9% respectively. The mortality rate was 52.4% but the authors did not report any case of stroke among the patients studied. This study therefore looked at stroke patients admitted into the ICU with the aim of determining the outcome in terms of ICU mortality, duration of stay and demographics of the patients.

Reports

Methodology

Setting: The Intensive Care Unit of the University of Benin Teaching Hospital, Benin, Nigeria is a 7 bed facility. Medical, surgical and neurological patients are admitted and managed in the unit. The unit is run by a group of anaesthetists in the department of Anaesthesiology of the hospital.

Inclusion criteria: All stroke patients admitted in the ICU from January 2002 to January 2012.

Ethical approval: Following approval from the Institution’s Ethics and Research Committee, data from all stroke patients admitted within the study period were retrieved from the hospital records and analyzed.

Data analysis: Data collected included socio-demographic characteristics such as age and sex. A patient before and after a stroke was selected as controls for the study. Primary outcome variable was ICU mortality, type of stroke whether ischemic or haemorrhagic, duration of stay, whether patients were transferred from the medical/stroke ward or from the accident and emergency department of the hospital.

All data retrieved were entered into SPSS version 16.0 and analyzed. Non-categorical data were analyzed using T test and categorical data were analyzed with chi square or Fisher exact test. P value < 0.05 was considered significant.

Results

A total of thirty six stroke patients were admitted into the ICU within the study period accounting for 5.6% of the total ICU admissions. Majority (83.3%) of the patients were admitted during the 2nd half of the study period. The male: female ratio was 2:1 and patients aged >60 years accounted for 55.6% as shown in Table 1.

Table 2 showed that 61.1% of stroke patients stayed in the intensive care unit between 1-5 days. Only six patients stayed between 6 -10 days while 3 patients stayed for more than 3 weeks. One patient stayed more than 30 days in the ICU. Stroke patients admitted into the unit within the study period had a mortality rate of 77.8%. In all, 22.2% were discharged to the ward.

Patients with severe stroke admitted into the ICU recorded four times more mortality when compared with non-stroke patients in the ICU (p=0.002, OR=4.472) as shown in Table 3.

Regarding the type of stroke suffered by the patients, ischaemic and haemorrhagic strokes were almost equally distributed representing 19(52.8%) and 17(47.2%) patients respectively. About 72.2% of stroke patients admitted into the ICU were transferred from medical wards in the hospital while 27.8% were admitted from the emergency room as shown in Table 4.

The type of stroke whether ishaemic or haemorrhagic had no significant impact on outcome, p value= 1.00. Seven (19.4%) patients were mechanically ventilated while the majority(80.6%) of these patients did not need ventilatory support.

In addition, the impact of the route of admission into the ICU on outcome is as shown in Table 4. There was no significant association between outcome and whether the patients were admitted from the medical or emergency ward (P value=1.000). Admission from the emergency ward mirrored the duration of stroke and ICU admission. Also, severe stroke had no significant impact on duration of ICU stay (p=0.454, OR=1.464). Table 5 shows the relationship between type of stroke and outcome.

Table 1: Age and sex distribution of stroke patients

Age	Frequency	Percentage
20 - 25	1	2.8
26 - 30	2	5.6
31 - 36	1	2.8
37 - 40	1	2.8
41 - 46	3	8.3
47 - 50	1	2.8
51 - 55	4	11.1
56 - 60	3	8.3
>60	20	55.6
Total	36	100
Sex
MALE	24	66.7
FEMALE	12	33.3

Table 2: Duration of hospitalization in the ICU

Duration of stay	Frequency	Percentage
1-5	22	61.1
6-10	6	16.7
11-15	1	2.8
16-20	1	2.8
21-25	3	8.3
26-30	2	5.6
>30	1	2.8

Table 3: Duration and outcome of Stroke and Non-stroke patients

	Mortality	Duration	Total
Stroke patients	28	12	36
Non-stroke patients	12	25	37
Total	36	37	73
P=0.002, OR=4.472, Duration in days.

Table 4: Route of admission and outcome

	Died	Discharge	Total
Transferred from medical ward	22	04	26
Transferred from the emergency room	08	02	10
TOTAL	30	06	36

Table 5: Relationship between type of stroke and outcome

	Died	Discharge	Total
Ischaemic stroke	14	05	19
Haemorrhagic stroke	13	04	17
Total	27	09	36
p=1.000

FIG 1 .

Outcome of stroke patients in ICU

Discussion

This study revealed that stroke patients constituted a small proportion(5.6%) of patients admitted into the intensive care unit of the University of Benin Teaching Hospital, Benin, Nigeria. This is similar to other studies which reported that severe stroke constituted between 3 -6%2^-5 and 7%6 of all ICU admissions.

A previous study from our centre16 on the morbidity and mortality pattern of neurological patients in the ICU did not include stroke patients. In that study, 187 patients were studied within the period of 1985-2003; head injury and tetanus patients constituted 63.7% and 13.9% respectively. With the availability of invasive monitoring and mechanical ventilation and increased level of awareness, it is therefore expected that the number of stroke patients admitted into the ICU will continue to be on the increase.

The ICU mortality of stroke patients from this study was 77.8%. Also, we found that severe stroke patients were 4 times more likely to die in the ICU than non-stroke patients. A previous study by Burthin et al15 also reported a mortality rate of 73% which was comparable to our finding. It is of note that this study was limited to mortality in the ICU. The benefit of ICU care on outcome of stroke patients remains controversial. It has been opined that ICU care of stroke patients may only serve to prolong the date of demise, without any significant favourable impact on eventual outcome14. Considering the doubtful outcome benefit and associated increased cost of care, there is a need to re-evaluate the protocol for ICU admission for stroke patients especially in a resource limited environment. Most(61%) of the patients in this study stayed in the unit between 1-5 days before discharge or death. Length of stay in the ICU has been found to have conflicting effect on patient’s outcome. Some authors have reported a favourable outcome with patients’ prolonged ICU stay17, others did not find any such association18. Other than length of stay, the severity of illness has been found to be more predictive of outcome in the ICU19. Although, the ICU in our centre could hardly be described as providing a level III care, this study has further strengthened the need to review the risk: benefit ratio of ICU admission of stroke patients as most of these patients ended up spending a few days before their demise.

Furthermore, many of these patients were transferred from the medical ward to the ICU due to worsening clinical status requiring intensive and specialized care it is therefore not surprising that majority of these patients only stayed for few days in the unit before they expired. It is debatable however if early ICU admission of this group of patients could offer some benefit in terms of reduction of morbidity and mortality. The provision of patient monitors and upgrading of the stroke ward to a high dependency unit (HDU) could help in this regard. This will leave only stroke patients requiring airway management and ventialtory support for ICU care.

The above finding differs significantly from that of Riachy et al6. In their study of 62 patients with severe stroke admitted into the ICU, they reported that majority of their patients were admitted from the accident and emergency ward, 8.1% each from the neuro-ward, from other medical ward in the hospital and from other hospitals. This difference could be due to different hospital protocols. In our centre, most stroke patients are initially admitted into the stroke ward. We however did not find any significant difference in ICU outcome between patients admitted via the accident and emergency ward and the stroke ward.

Despite the fact that ventilatory support is one of the indications for ICU admission in severe stroke, only 19% of our patients in this study were mechanically ventilated. One of the reasons for this finding could include inadequate number of mechanical ventilators in the unit. To compound this dilemma is the increased number of other none-stroke patients requiring ventilatory assistance. The provision of more mechanical ventilators in the ICU and a dedicated neuro-ICU with all its paraphilenia would go a long way in the care of our stroke patients.

This study revealed that both types of stroke, ishaemic and haemorrhagic were equally distributed in the ICU. This could be due to the fact that admission of stroke patients into the ICU is independent of the type of stroke. Apart from this, patient outcome in the ICU was not significantly affected by the type of stroke.

One obvious limitation of this study arises from the fact that it is a retrospective study. However, we have established that although, stroke patients constituted a small proportion of our ICU admission, they have a high mortality rate. This is independent of the type of stroke and whether transferred from ward or direct admission from the emergency room.

Conclusions

Stroke patients have a high mortality in the intensive care unit that is independent on the type and route of admission. Provision of the support equipment and instruments required for high dependency service in the intensive care unit and early admission should improve the outcome.