Introduction

Mechanical ventilation (MV) is an essential life‑support intervention used when patients cannot maintain adequate gas exchange independently [1]. Despite its lifesaving benefits, MV is also associated with substantial risks, including ventilator‑associated pneumonia, barotrauma, hemodynamic instability, and prolonged ICU stay that may worsen clinical outcomes in critically ill patients [2–4]. In low- and middle-income countries (LMICs), these risks are often exacerbated by shortages of trained ICU staff, limited equipment, and delays in critical care interventions [5,6].

Globally, the burden of patients requiring mechanical ventilation is increasing, largely due to rising trauma, infectious diseases, and non-communicable diseases (NCDs) [7]. However, in Ethiopia, research on MV outcomes remains limited, with most available studies conducted in few urban tertiary centers and lacking comprehensive assessment of key determinants such as re‑intubation status, mode of ventilation, pre‑intubation clinical status and accessibility of assurances [8–14].

Previous studies in Ethiopia have reported high mortality rates among mechanically ventilated patients, ranging from 41% to 57%, reflecting significant challenges in critical care delivery [2,8–10,14,15]. Yet these studies often lacked standardized definitions, did not sufficiently examine important variables like multi-organ dysfunction syndrome (MODS), and did not investigate prolonged pre‑intubation hospitalization, a potentially modifiable factor.

Furthermore, the Adult Intensive Care Unit (AICU) at Jimma Medical Center (JMC) serves a large catchment population and frequently manages complex critically ill patients, making it an important setting for understanding MV outcomes. To date, no comprehensive study has examined clinical outcomes and associated factors among mechanically ventilated adult patients at JMC over a long period, thus creating a critical evidence gap for improving care service and policy.

Therefore, this study aimed to assess clinical outcomes and identify key determinants of mortality among mechanically ventilated adult patients at Jimma Medical Center from 2019 to 2023. By addressing previously unexamined variables and synthesizing a 5-years ICU data, this study provides evidence that may improve ICU protocols, guide clinical decision‑making, and inform resource allocation.

Methods and materials

Variables

Results

Soco-dimographic and Admission Characteristics among mechanically ventilated patients in the AICU at JMC.

A total of 1,233 patients were admitted and received mechanical ventilation (MV) in the AICU at JMC from January 1, 2019, to December 30, 2023. Out of these, 411 medical records were reviewed. Most patients (67.9%) were aged between 18 and 40 years, with a mean age of 36.9 ± 16.2 years, ranging from 18 to 94 years. More than half of the patients were male, 215 (52.3%) (See Fig 1).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Socio-demographic characteristics among mechanically ventilated patients in the AICU at JMC from 2019 to 2023.

https://doi.org/10.1371/journal.pone.0341812.g001

The most common primary admission diagnoses were respiratory disorders (22%), followed by other infectious (18%) and Neurological Disorders (14%). At ICU admission, 52.3% had a Glasgow Coma Scale (GCS) score below 8, the most predominant initial indication of mechanical ventilation was respiratory failure 195 (47%), followed by GCS < 8 or coma 88 (21%). additionally; comorbidities were documented in 56.4% of patients, with hypertension and chronic heart disease being the most common;(see Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Indication of mechanical ventilators and related characteristics among mechanically ventilated in the AICU at JMC from 2019 to 2023 (n = 411).

https://doi.org/10.1371/journal.pone.0341812.t001

Mechanical ventilation–related Factors among mechanically ventilated patients in the AICU at JMC

Concerning complications, more than half of 252 (61.3%) of patients had developed complications. Of these complications, VAP and bed sores accounted for 55 (21%) and 52 (20%), respectively. More than one-third, 138 (33.6%) of the patients were re-intubated, and ETT block was the predominant reason for the re-intubation, which accounted for 52 (12.5%). Concerning multi-organ dysfunction syndrome (MODS), 316 (76.9%) of the patients had MODS. The most prominent MODS were cardiac (36.7%), renal (25.3%), and hepatic failures (20.8%); (see Table 2).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Complications, re-intubation, and MODS-related factors among mechanically ventilated patients in the AICU at JMC from 2019 to 2023 (n = 411).

https://doi.org/10.1371/journal.pone.0341812.t002

Clinical and Hospital-Related Factors among mechanically ventilated patients in the AICU at JMC

Out of ventilated patient 30.2% of them were developed multiple organ dysfunction syndrome (MODS). Among ventilated patients, cardiopulmonary resuscitation (CPR) was performed for (22.4%) of patients within 24 hours of ventilation. Of the ventilated patients, (62.5%) used Deep Vein Thrombosis (DVT) prophylaxis. The most prominent drug used for DVT prophylaxis was heparin 256 (62.3%). Of the total number of ventilated patients, 238 (58%) used inotropic drugs. The most commonly used inotropic drugs were adrenaline (48.9%), followed by dopamine (23.6%). Regarding sedation status of these patients 41.1% of cases were used sedation with particularly midazolam or fentanyl. The mean and standard deviation of the duration of illness before hospital admission were 17.2 + 15.9 days, and the duration of hospital stay before intubation was 5.3 + 4.8 days. Also, the mean length of stay on mechanical ventilator (MV), in the intensive care unit (ICU), and hospital were 10.1 + 11, and 11 + 12.19 days, respectively; (see Table 3).

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Clinical and Hospital-Related Factors among mechanically ventilated patients in the AICU at JMC from 2019–2023 (n = 411).

https://doi.org/10.1371/journal.pone.0341812.t003

Clinical outcome among mechanically ventilated patients in the adult intensive care unit at Jimma Medical Center from 2019 to 2023

Out of the total number of mechanically ventilated patients, two hundred twenty-two (54%) with 95% CI (49.24, 58.6) died, whereas one hundred eight nine (46%) recovered and were discharged alive; (see Fig 2).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Death outcome among mechanically ventilated patients in the AICU at JMC from 2019 to 20239 (n = 411).

https://doi.org/10.1371/journal.pone.0341812.g002

Factors associated with mortality

Bivairable analysis.

In bivairable analysis, many variables were significantly associated with mortality at p < 0.25 and therefore included in the multivariable model. These variables included, patient age, Pre-intubation hospital stay, re-intubation, presence of comorbidities, development of complications, sedation use, MODS, GCS of patients during admission, serum electrolytes, Creatine level, and CPR done;(see Table 4)

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Result of binary and multivariable logistic regression analysis for associated factors with clinical outcome among mechanically ventilated patients in the adult intensive care unit at Jimma Medical Center from 2019 to 2023 (n = 411).

https://doi.org/10.1371/journal.pone.0341812.t004

Discussion

The number of critically ill patients requiring mechanical ventilation (MV) is rising globally, including in Ethiopia, due to increases in trauma, infectious diseases, and non-communicable diseases (NCDs) [3]. This study, conducted at JMC's AICU from 2019 to 2023, provides significant insights into the socio-demographics, clinical characteristics, management practices, and outcomes of patients who received mechanical ventilation. The findings underscore the critical factors associated with mortality and recovery, offering valuable implications for clinical practice and future research.

In the current study, the overall mortality rate among mechanically ventilated patients was 54% with a 95% CI (49.4, 58.6), This finding is comparable to the findings of other studies conducted in resource-limited settings, such as Egypt (54.7%) [26] and Millennium Medical College in Ethiopia (57.1%) [27]. These similarities are likely due to the comparable demographics and socioeconomic conditions of the patient populations, as well as the similarity in admission diagnoses, particularly respiratory failure, which was the most common indication for MV in both studies.

However, the mortality rate in this study was lower than that reported in Uganda (73.5%) [28] and Kenya (60.7%) [29]. A potential explanation for this discrepancy might be that the facilities at Jimma Medical Center are relatively better equipped, and the critical care professionals in the ICU may be more skilled than those in the rural hospitals in these other countries.

Conversely, the mortality rate in our study was significantly higher than the rates reported in middle- and high-income countries like Canada (18%) [18], Brazil (34%) [30], and Korea (36%) [31]. Delays in ICU admission, scarcity of resources, patients unable to pay the cost, since 84.7% died of ventilated patients who had no access to insurance in this study, and a lack of standardized ICU protocols in Ethiopia are likely causes of this disparity [14]. The mortality rate in this study was also higher than in studies conducted in southern India (42.1%) [32]. The highest mortalities in this study might be due to late referral to the ICU and admission problems (admitting unsalvageable patients due to lack of standardized protocol for ICU admission in this setting). Additionally, the rate of mortality in this study was higher than in Ayder Hospital in Ethiopia (28.6%) [3]. A possible justification for this discrepancy might be due to the high re-intubation rate (33.6%), and most of the re-intubated patients developed complications (61.3%) in this hospital, which was a risk for death in this study. However, in Ayder Hospital, only 10% of patients had re-intubations, of which only 35.3% had developed complications.

In this study, the mortality rate was higher in patients with respiratory disease (45%), which is similar to a study conducted in western India [33]. The majority of deaths of patients were those who were admitted from emergencies, which was a similar study conducted in western India [33] and Brazil [27]. However, a study conducted in Sub-Saharan Africa reveals that the majority of deaths of patients were those from medical wards (56.2%) [14]. This difference might be because the majority of ICU admissions were from emergencies, and different patients referred to this hospital as emergencies.

Among dead patients, the majority of age was found between 18 and 40 years, which was similar to the studies done in southern India [32], Nigeria [34], and Sub-Saharan Africa [14], but studies conducted in western India [33] and Korea [34] found older patients to be over 60 years old. This might be different due to the most common ICU admission in Ethiopia related to an emergency car accident vulnerable to the young population.

Acute respiratory distress syndrome (ARDS) was the main cause of death for patients who were on mechanical ventilation in this study (43.7%), which is similar to a study conducted in Western India [33]. Other studies done in Southern India [32], Nigeria [34], and sub-Saharan Africa [14] also show that acute respiratory distress syndrome was the main cause of death for those patients who were on mechanical ventilators, which ranged from 38.6% to 86.07%. Nevertheless, the study done in Egypt revealed cardiac disease as the predominant cause of death. This discrepancy might be due to the type of ICU they had—the cardiac ICU, the respiratory ICU—and the different categories of disease settings admitting patients based on specific diseases in each particular ICU.

The study identified several factors significantly associated with increased mortality, including prolonged hospital stay before intubation, re-intubation, the presence of comorbidities, the development of complications, MODS, and sedation.

Patients who stayed in the hospital for more than five days before intubation had three times higher odds of mortality. The mean duration of hospital stays before intubation (5.3 days) and the length of stay on mechanical ventilation (10.1 days) indicate the prolonged and complex care required for these patients. The use of invasive mechanical ventilation (IMV) in 96.4% of cases and endotracheal tube (ETT) as the primary airway access method (91.5%) is different from standard critical care practices [4,35]. This finding suggests that delays in initiating mechanical ventilation may contribute to worse outcomes; possibly discrepancy might be due to the progression of underlying disease and the development of complications [14].

Re-intubation is often a marker of failed initial weaning attempts [31]. It may indicate underlying issues such as unresolved primary illness or new complications like endotracheal tube blockages, accidental extubating, and self-extubating, all of which were common in this study. Patients who underwent re-intubation had a nine-fold increase in the odds of mortality compared to those who were not re-intubated. This result is consistent with findings from studies conducted in Korea [36], India [32], Brazil [30], China [18], and Sub-Saharan Africa [14]. Re-intubation increases the risk of complications, particularly VAP, as it is an invasive procedure that increases the likelihood of infection. The high re-intubation rate at JMC highlights the need for better monitoring of ventilated patients and more robust protocols for preventing unplanned extubation.

The presence of comorbidities and the development of complications were both strongly associated with increased mortality. A significant proportion of ventilated patients in this study had comorbidities, with congestive heart failure (28.8%), hypertension (20%), chronic kidney disease (12%), and diabetes mellitus (10%) being the most common. These comorbidities were directly associated with higher mortality rates. This finding aligns with studies conducted in India [33] and China [18].

In Ethiopia, the high burden of chronic diseases is often undiagnosed or poorly managed in the general population, especially in rural areas [14]. This increases the vulnerability of patients requiring mechanical ventilation, as they are more prone to complications like multi-organ dysfunction syndrome (MODS), which was seen in 76.9% of the patients in this study. These findings highlight the importance of early identification and management of comorbid conditions and vigilant monitoring for complications during mechanical ventilation.

In this study, 61.3% of patients developed complications during ventilation, with ventilator-associated pneumonia (VAP), hospital-acquired pneumonia (HAP), electrolyte imbalances, and bedsores being the most common. These complications significantly contributed to the high mortality rate, consistent with studies from Egypt [5] and Sub-Saharan Africa [14]. However, this result was higher than in Brazil [30,37], and China [38]. The high rate of complications can be attributed to suboptimal infection control practices and a shortage of critical care nurses in the ICU at JMC. This is a common issue in LMICS where healthcare workers may lack the necessary training in ICU protocols and infection prevention [39].

The development of MODS was associated with a three-fold increase in mortality risk. Multi-organ dysfunction syndrome (MODS) was observed in 76.9% of patients, with renal failure (36.7%) and cardiac failure (25.3%) being the most common. This is consistent with findings in other low-middle-income countries [9,14,26]. However, the rates were higher than in developed nations such as Brazil [30] and India [38], highlighting the severity of illness in patients admitted to the ICU at JMC. The development of MODS is often a consequence of delayed ICU admission and inadequate management of complications, which is common in resource-limited settings like Ethiopia. The possible justification for this might be the lack of specialized healthcare workers such as respiratory therapists, and limited diagnostic facilities often result in delayed treatment for complications like MODS. Investments in specialized training and diagnostic capabilities are essential for improving outcomes.

According to this study, sedation use was shown to be an important factor significantly associated with mortality in mechanically ventilated patients. Sedation use was found to decrease the odds of mortality by 63.2%, as it prevented agitation and self-extubating, which could lead to traumatic injuries and respiratory failure [22,40]. However, excessive or prolonged sedation has been linked to higher mortality in other studies [41]. The judicious use of sedation, guided by protocols such as the Richmond Agitation-Sedation Scale (RASS), is crucial in LMICs to balance the risks and benefits of sedation in critically ill patients [42]. Continuous monitoring and appropriate titration of sedation could reduce the complications associated with mechanical ventilation [31]. This finding is supported by previous studies in Sub-Saharan Africa [14]. However, studies from other regions, such as Egypt, have shown that prolonged sedation can increase the risk of mortality due to delayed extubating [5]. The possible justification for this controversial practice is likely at JMC, where the controlled use of sedation is guided by the Richmond Agitation-Sedation Scale, which helps healthcare professionals maintain an optimal sedation level that prevents agitation without delaying extubating.

The strengths of this study provide a comprehensive analysis of the clinical outcomes of mechanically ventilated patients in the AICU over five years, making it the first of its kind in the southwest region of Ethiopia. By examining a diverse cohort of patients across different units, the research enhances the generalizability of the findings.

Despite, this, study has some limitations that should be acknowledged. Firstly, the retrospective design means that direct intervention or real-time observation of patients was not possible, which may have limited the collection of some important variables relevant to the study. Additionally, those who had requested against medical advice could not be tracked with poor prognoses until they passed away or recovered, making it unable to assess all outcomes of ventilated patients.

Conclusion and recommendations

Acknowledgments

First of all, we would like to express our sincere appreciation to Jimma University for giving us this chance to participate in the research activity. We would like to acknowledge Jimma University Medical Center, Shanan Gibe General Hospital, and their respective staff for their support while undertaking this study. Also, we are grateful to acknowledge our study subject who voluntarily participated in this study for providing the necessary information, and the data collectors for collecting the data carefully.