Printer Friendly

Outcomes in critical care delivery at Jimma University Specialised Hospital, Ethiopia.


The aim of this study was to assess outcomes following intensive care unit (ICU) admissions at Jimma University Specialised Hospital, Ethiopia. This was a retrospective observational study. Data were collected regarding all ICU admissions and discharges during a 12-month period beginning August 2011. Demographic data and information regarding diagnosis, length-of-stay and outcome were gathered and data analysed. There were 370 admissions to the ICU during the study period. Median age (interquartile range) was 32.0 (22.0-47.0) years and 56.2% were males. The median length-of-stay (interquartile range) was 3.0 (1.0-7.0) days. The overall ICU mortality rate was 50.4% and major causes included trauma, cardiac disease, acute abdominal presentations, septic shock, tetanus and hysterectomy secondary to uterine rupture. Medical diagnoses accounted for 50.1% of admissions followed by surgery (43.2%) and obstetrics (5.8%). Corresponding mortality rates were 53.6, 48.0 and 42.9%, respectively. The main cause for surgical admission was trauma, with head injury carrying a mortality of 52.1%. The principal cause for medical admission was cardiac disease. In children, trauma, upper airway obstruction and communicable diseases were most common. Critical care mortality rates at this Ethiopian university hospital reflect the challenges facing critical care delivery in the developing world. Delayed presentation to hospital secondary to poor access to healthcare plays a predominant role. This is confounded by inadequate staffing, training, diagnostic and interventional limitations. Despite resource restraints, simple cost-effective measures may improve morbidity and mortality.

Key Words: intensive care: outcomes, mortality, developing countries


In the developed world, critical care is an essential element of the continuum of patient care that can ultimately lead to improved outcomes. However, in the developing world setting, intensive care delivery faces multifaceted challenges. While interventions such as rapid administration of fluids and antibiotics are relatively inexpensive, it is a commonly held belief that intensive care is too costly for developing countries (1,2). However, sustainable and appropriate critical care is possible in resource-poor sub-Saharan African countries (3). In comparison to Western intensive care populations, equivalent developing world cohorts are often younger with fewer comorbidities and thus have a greater chance of recovery (4).

Ethiopia has a population of over 83 million people and life expectancy is 54 years (5). In 2010, 38% and 12% of the population had access to improved drinking water and sanitation, respectively. Access to healthcare is poor and physician density is <0.5 per 10,000 people, while the equivalent European figure is over >30 per 10,000 (5). Jimma University Specialised Hospital (JUSH) is a 450-bed teaching university and referral hospital, situated 350 km to the south-west of Addis Ababa. It provides services for 90,000 patients per annum and has a catchment population of 15 million people. Despite this, facilities are limited with no magnetic resonance imaging or computed tomography scanning, and no specialised neurosurgical, cardiothoracic or vascular services. In the event that it is affordable, patients requiring such investigations are referred to Addis Ababa.

The six-bed intensive care unit (ICU) at JUSH was created in 2006 and while it has capacity for three ventilated patients, additional supporting services, equipment and pharmaceuticals are extremely limited, primarily due to expense and lack of trained staff. While electrocardiography and non-invasive blood pressure monitoring are routinely employed, and continuous pulse oximetry is available for ventilated patients, there is no capnography. Likewise, blood gas analysis is not possible, and biochemical and haematological investigations are seldom performed due to prohibitive cost to patients. The only routine investigation on ICU admission is haematocrit. Similarly, due to the previously mentioned constraints and infection risks, there is no access to invasive monitoring. Inotropes are rarely used but, if necessary, adrenaline is titrated using a drop-counting technique according to non-invasive blood pressure and clinical improvement. In the event of major haemorrhage, blood products are limited to whole blood donation from family members.

Developments in critical care depend on successive introduction of service improvement strategies with emphasis on sustainability, training and a cost-effectiveness analysis. Thus research in resource-poor settings is critical, focussing on assessment of needs and economic planning. There have been no previous studies of morbidity and mortality conducted in JUSH ICU. The aim of this study was to investigate outcomes of ICU admission and to identify areas for improvement in care.


This was a retrospective observational study conducted in JUSH ICU. Data were collected from a record routinely completed by nursing staff, documenting all ICU admissions and discharges. Demographic data and information regarding diagnosis, length-of-stay and outcome were drawn for the 12-month period from 23 August 2011 to 22 August 2012. Data were anonymous and subsequently analysed using Excel 2011 Version 14.2.3. Incomplete datasets were excluded from additional analysis.


There were 370 admissions to ICU during the 12-month study period. Median age (interquartile range) was 32.0 (22.0-47.0) years and 56.2% were male. The median length-of-stay (interquartile range) was 3.0 (1.0-7.0) days. Data were incomplete for an outcome in 13 cases and therefore were excluded from mortality rate calculations. There were 180 deaths giving an all cause ICU mortality rate of 50.4% (n=357). Data were missing or illegible for cause of admission in nine cases and these data were excluded from further analysis.

When data were analysed according to specialty (Table 1), medical diagnoses accounted for the greatest number of admissions and the poorest outcomes with an overall mortality rate of 53.6%. The leading causes of medical admission were cardiac disease, septic shock, non-tuberculous respiratory disease and tetanus (Table 2). The median age of all patients suffering upper airway obstruction was 5.5 years (Table 3). All cases of malaria, diabetic ketoacidosis, cerebrovascular accident and status epilepticus admitted to the ICU in the study period proved fatal. Mortality rates and length-of-stay according to the major causes for medical admission are shown in Table 3. The longest lengths-of-stay were in patients suffering advanced retroviral disease and tetanus cases requiring ventilation for prolonged periods.

The major surgical causes of admission to ICU are detailed in Tables 2 and 4. All cause surgical mortality was 48.0%. The leading single cause was severe head injury (n=50), which carries a mortality of 52.1% and comprises a predominantly young male population. Trauma other than severe head injury included blunt abdominal trauma, penetrating trauma, orthopaedic trauma and burns. Other causes for surgical admissions included thyroidectomy (3.2%) and malignant disease (2.6%).

There were 21 obstetric cases admitted to the ICU. The majority (n=13) were cases of hysterectomy secondary to uterine rupture, which carried a mortality of 46.2% (including one 15-year-old girl). The mortality rate for eclampsia (n=5) was 20.0%.

There were marked differences in the gender of patients admitted to the ICU according to some causes of presentation (Tables 1, 3 and 4). The most striking differences were predominance for male tetanus (86.7%) and trauma (76.8%) patients, while female pateints accounted for a higher proportion of cardiac cases (61.5%).

Fifty-one patients admitted to the ICU were <16 years (Table 5). The mortality rate for these patients was 42.9% and median age (interquartile range) was 9.0 (3.0-13.0) years. Medical, surgical and obstetric cases accounted for 54.9, 43.1 and 2.0%, respectively. Severe head injury accounted for 54.5% of all surgical admissions (n=12). The remainder comprised of acute abdominal complaints requiring operative intervention (n=6) and 'other surgical' causes (n=4) including burns and thyroidectomy. The predominant medical cause for admission was upper airway obstruction (n=10). The other major cause for admission was communicable disease (n=12), which comprised pneumonia, tetanus, meningitis and severe malaria.


This study identified an all-cause ICU mortality rate of 50.4%. Internationally utilised severity sickness scores permitting outcome comparisons with similar patients managed in developed world ICUs necessitate laboratory investigations. Laboratory support in most developing world facilities is poor, precluding the use of such scores in this study. Admission physiological parameters in JUSH ICU patients are often severely deranged. This is usually secondary to delays in presentation as a result of limited access to healthcare throughout Ethiopia (5). This partially accounts for the remarkable ICU mortality rate observed in this study. While this study was purely a retrospective analysis, it has also highlighted other likely contributory factors in such poor outcomes.

The leading single cause of surgical mortality was severe head injury, with trauma accounting for the majority of paediatric and adult admissions. Trauma represents a major burden in sub-Saharan Africa. It is often severe and poorly managed due to inadequately prepared health systems, delays in presentation and failure to apply Advanced Trauma Life Support principles (6-8). In Ethiopia, head injury management presents further challenges due to non-availability of diagnostic imaging techniques and lack of rehabilitative support. Future interventions should initially aim to implement effective early trauma management.

In contrast to the developed world, tetanus is endemic throughout many sub-Saharan countries due to a lack of effective vaccination schedules and inappropriate injury treatment. In a study of 171 tetanus cases conducted at JUSH by Amare et al, the mortality rate was 38% (9). Since only the more severe tetanus cases are referred to the ICU, the mortality rate in this study was correspondingly higher (53.3%) than other Ethiopian studies (27-55%) (10-12). Universally, young males are most affected, reflecting population demographics and the occupational work undertaken by this cohort. Respiratory failure due to a lack of early respiratory support in tetanus cases has been identified as a major cause of mortality (9).

Respiratory support itself is fraught with difficulty, often exacerbating problems by introducing infection or damaging lung architecture. Decisions to ventilate patients may be delayed due to the relative inexperience of doctors typically staffing ICU and more timely initiation of positive pressure ventilation may prove beneficial (3). JUSH does not currently employ lung protection ventilation strategies and ventilator use is not universally safe due to inadequate staffing and training. The adoption of 'ventilator care bundles' may be a simple cost-effective intervention. However, two nurses with no specific intensive care training currently care for six critically ill patients, three of which may be ventilated. Inappropriate staffing limits the feasibility of safely introducing novel strategies. A study involving seven sub-Saharan African ICUs showed improvements in mortality with an increase in nurse-to-patient ratios and appropriate monitoring (3).

Surgical site and other soft tissue infections were associated with high mortality in this study (58.3%). They are common due to lack of basic resources including running water and sterile dressings. Hand hygiene, wound and intravenous cannula care all present challenges while infections are difficult to treat effectively due to a lack of microbiological support and sole employment of empirical antibiotics. Initiatives such as introducing the World Health Organization Safe Surgical Checklist's to JUSH operating theatres may reduce the incidence of surgical site infection.

Maternal mortality in Ethiopia is twice the global average (5). Huge variation exists between developed and developing nations in obstetric ICU admission, with the most common indications worldwide being haemorrhage followed by pre-eclampsia. There were few obstetric admissions to ICU in our study, suggesting that many patients are managed in the ward. While this may be appropriate, early escalation to critical care should be encouraged. The majority of patients were young, serving as a good prognostic indicator provided interventions are timely. The majority of our cases were hysterectomy following uterine rupture after obstructed labour, reflecting poor access to emergency healthcare.

Child mortality rates in developing countries are five to 15 times higher than those in the developed world (5,14). Efforts at reduction are often hindered by a high trauma burden, ignorance of signs of severe paediatric illness in the community and difficulties in accessing health services (15,16). This is supported by this study, in which trauma and communicable diseases were the leading causes for paediatric ICU admission. The relative under-representation of children in this study was due to paediatricians' preference to manage children in the high-care section of a dedicated paediatric ward, unless they needed ventilation.

This study highlighted several isolated causes of ICU admission with outstandingly poor outcomes. Universally fatal cases included severe malaria, diabetic ketoacidosis and status epilepticus. In the majority of these cases, the poor outcome was due to extremely delayed presentation to critical care in moribund patients. While pre-hospital delays are difficult to solve acutely, failure to escalate care from ward level may be improved by staff education and the employment of ward-based early warning scoring systems. The lack of standardised management of some conditions may play a role, and elsewhere the introduction of mandatory management protocols has reduced morbidity, mortality and length-of-stay in conditions such as diabetic ketoacidosis (17,18). Conversely, it may be inappropriate to admit cases with little potential for reversibility in which ICU is not necessarily a cost-effective destination. Carefully considered ICU admission criteria may enhance the utilisation of limited resources.

The limitations of this study include poor hospital record-keeping, which prevented accurate follow-up of patients discharged from the ICU. Thus, survival to hospital discharge is unknown and in-hospital mortality may be higher than quoted. The sole record of ICU admissions and discharges is kept by nursing staff. It does not include data regarding time of admission, ventilation, final diagnosis or cause of death, and there is an inevitable overlap in disease state classification. There is a clear demand for enhanced record-keeping and maintenance of a physician-led ICU database, in order to audit clinical practice and effectiveness of future interventions.

Key areas for consideration when developing intensive care in a resource-poor setting include adequate staffing levels and training, research and clinical audit, ethical considerations, and the provision of appropriate sustainable equipment and support services (4). At Jimma University, a new hospital is currently under construction. While it will have greater capacity and more modern equipment, this alone is unlikely to solve the problems identified. Despite resource limitations, simple cost-effective measures may be implemented in order to improve morbidity and reduce high mortality rates.


(1.) Sachdeva RC. Intensive care--a cost effective option for developing countries? Indian J Pediatr 2001; 68:339-342.

(2.) Engelhardt HT Jr, Rie MA. Intensive care units, scarce resources, and conflicting principles of justice. JAMA 1986; 255:1159-1164.

(3.) Towey RM, Ojara S. Intensive care in the developing world. Anaesthesia 2007; 62:32-37.

(4.) Riviello ED, Letchford S, Achieng L, Newton MW. Critical care in resource-poor settings: lessons learned and future directions. Crit Care Med 2011; 39:860-867.

(5.) World Health Organization. World Health Statistics 2010. From

(6.) Hofman K, Primack A, Keusch G, Hrynkow S. Addressing the growing burden of trauma and injury in low- and middle-income countries. Am J Public Health 2005; 95:13-17.

(7.) American College of Surgeons Committee on Trauma. Advanced Trauma Life Support, 8th ed. Chicago: American College of Surgeons 2008.

(8.) Hyder AA, Wunderlich CA, Puvanachandra E Gururaj G, Kobusingye OC. The impact of traumatic brain injuries: a global perspective. NeuroRehabilitation 2007; 22:341-353.

(9.) Amare A, Yami A. Case-fatality of adult tetanus at Jimma University Teaching Hospital, Southwest Ethiopia. Afr Health Sci 2011; 11:36-40.

(10.) Habte-Gabr E, Mengistu M. Tetanus in Gondar Public Health College Hospital, Ethiopia: a review of 72 cases. Ethiop Med J 1978; 16:53-61.

(11.) Ramos JM, Reyes F, Tesfamariam A. Tetanus in a rural Ethiopian hospital. Trop Doct 2008; 38:104-105.

(12.) Hodes RM, Teferedegne B. Tetanus in Ethiopia: analysis of 55 cases from Addis Ababa. East Afr Med J 1990; 67:887-893.

(13.) World Health Organization. World Alliance for Patient Safety: WHO guidelines for safe surgery. From

(14.) Gwatkin DR. How many die? A set of demographic estimates of the annual number of infant and child deaths in the world. Am J Public Health 1980; 70:1286-1289.

(15.) Okwaraji YB, Cousens S, Berhane Y, Mulholland K, Edmond K. Effect of geographical access to health facilities on child mortality in rural Ethiopia: a community based cross sectional study. PLoS One 2012; 7:e33564.

(16.) Paxton LA, Redd SC, Steketee RW, Otieno JO, Nahlen B. An evaluation of clinical indicators for severe paediatric illness. Bull World Health Organ 1996; 74:613-618.

(17.) Bull SV, Douglas IS, Foster M, Albert RK. Mandatory protocol for treating adult patients with diabetic ketoacidosis decreases intensive care unit and hospital lengths of stay: results of a nonrandomized trial. Crit Care Med 2007; 35:41-46.

(18.) Kitabchi AE, Umpierrez GE, Fisher JN, Murphy MB, Stentz FB. Thirty years of personal experience in hyperglycemic crises: diabetic ketoacidosis and hyperglycemic hyperosmolar state. J Clin Endocrinol Metab 2008; 93:1541-1552.

Z. A. SMITH *, Y. AYELE ([dagger]), E MCDONALD ([double dagger])

Department of Anaesthesiology, Jimma University Specialised Hospital, Jimma, Ethiopia

* MB, BCh, BSc, DTM&H, DiMM, Visiting Lecturer in Anaesthesiology.

([dagger]) MD, Anaesthesiologist.

([double dagger]) MB, ChB, FRCA, FFICM, Consultant in Anaesthetics and Intensive Care Medicine, Department of Anaesthetics, St Richard's Hospital, Chichester, West Sussex, United Kingdom.

Address for correspondence: Dr Z. A Smith. Email: drzoesmith@gmail. com

Accepted for publication on March 4, 2013.
Table 1
Outcomes, sex, median age and length-of-stay according to specialty
(n = 361)

 Medicine Surgery

Total cases (%) 184 (50.1) 156 (43.2)
Deaths (%) 97 (52.7) 72 (46.2)
Condition improved * (%) 84 (45.7) 78 (50.0)
Outcome unknown (%) 3 (3.6) 6(3.8)
Male (%) 97 (52.7) 113 (72.4)
Female (%) 87 (47.3) 43 (27.6)
Median age in years (IQR) 35.0 (25.0-46.3) 31.0 (20.0-50.0)
Median LOS in days (IQR) 3.0 (1.0-7.0) 3.0 (2.0-7.0)


Total cases (%) 21 (5.8)
Deaths (%) 9(42.9)
Condition improved * (%) 12 (57.1)
Outcome unknown (%) 0(0.0)
Male (%) --
Female (%) 21 (100.0)
Median age in years (IQR) 25.0 (24.0-30.0)
Median LOS in days (IQR) 2.0 (1.0-4.0)

* Sufficient improvement to meet ICU discharge criteria. IQR =
interquartile range, LOS = length-of-stay.

Table 2
Spectrum of cases (n = 361) admitted to the intensive care
unit according to frequency

Cause for ICU admission No. of cases

All trauma 69
General surgical 57
Cardiac disease 52
Septic shock 19
Non-tuberculous respiratory disease 19
Tetanus 15
Upper airway obstruction 14
Uterine rupture 13
Acute soft tissue infections 12
Tuberculous disease 12
Severe malaria 8
End-stage HIV 6
Guillan Barre Syndrome 6
Eclampsia 5
Thyroidectomy 5
Other 49
Total 361

ICU = intensive care unit, HIV = human
immunodeficiency virus.

Table 3
Mortality, age, sex and length-of-stay according to the leading causes
for medical ICU admission

Cause for admission n Median age Median LOS
 (IQR) (IQR)

Cardiac disease * 52 38.0 (29.5-48.3) 3.0 (1.0-7.0)
Septic shock 19 34.0 (30.0-61.5) 1.5 (0.3-3.0)
 respiratory disease 19 35.0 (21.5-47.0) 2.0 (1.0-2.5)
Tetanus 15 35.0 (17.5-39.0) 6.0 (2.0-16.5)
Upper airway obstruction
 ([daggeer]) 14 5.5 (2.0-14.3) 5.5 (1.0-11.0)
Tuberculous disease * 12 30.0 (30.0-36.3) 4.5 (1.0-7.8)

Cause for admission Mortality Female Male
 (%) (%) (%)

Cardiac disease * 48.1 61.5 38.5
Septic shock 57.9 63.2 36.8
 respiratory disease 41.2 42.1 57.9
Tetanus 53.3 13.3 86.7
Upper airway obstruction
 ([dagger]) 42.9 35.7 64.3
Tuberculous disease * 26.7 58.3 41.7

* Congestive cardiac failure, cardiogenic shock, myocardial
infarction. ([dagger]) Croup, pharyngeal abscess, epiglottitis.
([double dagger]) Tuberculous meningitis, tuberculous pericarditis,
disseminated tuberculosis, pulmonary tuberculosis. LOS = length-of-
stay, IQR = interquartile range.

Table 4
Mortality, age, sex and length-of-stay according to the leading
causes. for surgical ICU admission

Cause far admission n Median age Median LOS
 (IQR) (IQR)

All trauma 69 25.0 (18.0-40.0) 4.0 (2.0-8.0)
Severe head injury 50 25.0 (16.3-39.5) 3.0 (1.8-8.3)
Other trauma 18 30.5 (21.3-57.0) 4.0 (3.0-7.0)
General surgery * 57 35 (25.0-49.0) 3.0 (1.0-6.0)
Acute soft tissue infection 12 40.0(32.5-55.0) 1.5(1.0-4.5)

Cause far admission Mortality Female Male (%)
 (%) (%)

All trauma 50.0 23.2 76.8
Severe head injury 52.1 28.0 72.0
Other trauma 47.1 11.1 88.9
General surgery * 49.1 24.6 75.4
Acute soft tissue infection 58.3 33.3 66.7

* Appendicitis, peritonitis, small and large bowel obstruction,
typhoid perforation, acute abdomen, postoperative laparotomy, re-
laparotomy. LOS = length-of-stay, IQR = interquartile range.

Table 5
Causes for paediatric intensive care admissions
(<16 years of age)

Cause for ICU admission No. of children < 16 years

Severe head injury 12
Communicable disease 12
Upper airway obstruction 10
General surgical 6
Other medical 6
Other surgical 4
Obstetric 1
Total 51

ICU = intensive care unit.
COPYRIGHT 2013 Australian Society of Anaesthetists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Original Papers
Author:Smith, Z.A.; Ayele, Y.; McDonald, P.
Publication:Anaesthesia and Intensive Care
Article Type:Report
Geographic Code:6ETHI
Date:May 1, 2013
Previous Article:Has the middle-level anaesthesia manpower training program of the West African College of Surgeons fulfilled its objectives?
Next Article:A survey of patient understanding and expectations of sedation/anaesthesia for colonoscopy.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |