ABSTRACT
Conclusion:
Prolonged ICU stay are caused by multiple factors, and palliative care units and home care facilities must be used frequently to make the best use of ICU beds and to prevent prolonged ICU stays, which cause increased mortality and negative financial outcomes.
Results:
The age and gender distributions of the 178 patients were not statistically different between the two groups (p=0.355 and p=0.758, respectively). The group with an ICU stay of ≥30 days had a significantly higher tracheostomy rate (p<0.05) than the group with an ICU stay of <30 days. In this study, percutaneous endoscopic gastrostomy procedures were used more frequently on patients who stayed in the ICU for 30 days or more than on those who stayed for less than 30 days (p=0.000).
Materials and Methods:
The data of 178 patients were analyzed retrospectively. Those with an ICU stay of 7-29 days (n=89) were assigned as “ICU stay day <30 days - Group 1” and those with a stay of 30 days and more (n=89) were assigned as “ICU stay day ≥30 days - Group 2”. The factors related to a prolonged ICU stay were investigated in this study. The data obtained from the hospital data system were compared.
Aim:
A prolonged intensive care unit stay (ICU) is associated with many factors and causes various problems. This study aims to evaluate the clinical characteristics and the factors that led to the stay of the patients treated in the ICUs for 30 days and more.
INTRODUCTION
The care of critically ill patients requires a significant amount of time, resources, and money as well as specialist staff, nursing care, and equipment1,2.
Although some critically ill patients are discharged early from intensive care units (ICUs), others require a prolonged ICU stay, which causes complications such as muscle weakness, pressure ulcers, infections, pulmonary embolism, and delirium. Additionally, a prolonged ICU stay is associated with high hospital-related morbidity and mortality rates and negative long-term consequences3-7. Critically ill patients with a prolonged ICU stay typically require more time and medical resources than usual4. It may be impossible to achieve acceptable mortality rates and quality of life despite the high costs invested in medical care associated with prolonged ICU stays8-10.
The precise definition of a prolonged ICU stay varies in the literature, as several studies have used different stay limits, such as >7, 14, 21, 29, or 30 days11-18. The APACHE-II score, the SOFA score, the need for mechanical ventilation, the need for vasopressors, and multiple organ failure have been identified as the factors that contribute to a prolonged ICU stay15,18-20.
In this study, we aimed to determine the clinical characteristics of patients who were treated in ICUs for 30 days or more and to determine the factors that affected the prolonged duration of their stay.
MATERIALS AND METHODS
Ethical approval of this study was obtained from Tekirdağ Namık Kemal University Non-Interventional Research Ethics Committee (protocol no: 2022.191.10.15, date: 25.10.2022).
In our hospital, there are two 11-bed tertiary ICUs controlled by the department of anesthesiology and reanimation. These ICUs accept surgical and nonsurgical patients and have anesthesiologists available around the clock. There are also other ICUs controlled by other departments such as internal medicine, cardiology, cardiovascular surgery, neurology, and pediatric diseases.
A six-year retrospective study was conducted after receiving approval from the local ethical committee. From May 2017 to September 2022, scans and data collection were performed on 2811 patients who were being monitored in the 11-bed tertiary ICU, which accept patients from the operating theater, wards, emergency services, and outlying center hospitals.
The hospital data system was scanned retrospectively, and 89 patients who spent 30 days or more in the ICU between May 2017 and September 2022 were identified. Additionally, the files of 89 patients who spent between 7 and 30 days in the ICU were randomly scanned and selected from the hospital data system. In other words, this study included 178 patients divided into two groups: Group 1 for ICU stay <30 days and Group 2 for ICU stay ≥30 days. All the data were obtained from the administration registry book of the ICU, hospital electronic patient data system, and patient charts.
This study was conducted according to the ethical principles outlined in the Helsinki Declaration and the guidelines of good clinical practice.
Statistical Analysis
The mean, standard deviation, median, minimum, maximum, frequency, and ratio values were used for the descriptive statistics of the data. The Kolmogorov-Smirnov test was used to measure the distribution of the variables. The Independent Sample t-test and the Mann-Whitney U test were employed to analyze the quantitative independent data. A chi-square test was used to analyze the qualitative independent data, and Fisher’s test was used when the chi-square test conditions were not met. Statistical analysis was carried out with the Statistical Package for the Social Sciences 28.0 software.
RESULTS
The patients had a median age of 66 years, with a mean age of 64.8±15.8 years. The patients comprised of 38.8% females and 61.2% males. Group 1 had a mean age of 65.6 years while Group 2 had a mean age of 64 years. Group 1 consisted of 33 female (37.1%) and 56 male (62.9%) patients, Group 2 consisted of 36 female (40.4%) and 53 male (59.6%) patients (Table 1). The age and gender distributions of the patients were not statistically different between the two groups (p=0.355; p=0.758) (Table 1). About 37.1% and 41.6% of the patients with ICU stays of <30 days and ≥30 days, respectively, were accepted from emergency services. Patients accepted from outlying hospitals had the second-highest population. The admission diagnoses of both groups were respiratory failure following CPR and chronic obstructive pulmonary disease, respectively. The three main admission sources and admission diagnoses did not differ between the groups (p>0.05).
Tracheostomy was performed on 28.1% (n=50) of the patients during their ICU stay (Table 1). Additionally, percutaneous endoscopic gastrostomy (PEG) was performed on patients who were unable to eat orally or who were predicted to require tube-feeding for a long time. About 11.8% (n=21) of the patients underwent PEG procedure (Table 1). Furthermore, 32% (n=57) of the patients required vasopressors, 92.7% (n=165) were on mechanical ventilation upon admission, and 28.1% (n=50) were admitted to the ICU after surgery (Table 1).
Group 2 had a significantly higher tracheostomy rate (p<0.05) than Group 1 (Table 1). Group 2 had a significantly longer tracheostomy opening day and a longer number of days from tracheostomy opening until discharge (p<0.05) than Group 1. Group 2 had a significantly higher PEG ratio (p<0.05) than Group 1 (Table 1). Both groups had no significant differences in the day of PEG opening and the number of days from PEG opening until discharge from the ICU (p>0.05; Table 1).
Group 2 had a considerably lower need for inotropic drugs at admission (p<0.05) than Group 1 (Table 1). Additionally, Group 2 had a considerably longer inotrope intake time (p<0.05) than Group 1. The rate of receiving mechanical ventilation support did not differ significantly (p>0.05) between both groups. Group 2 had a significantly longer mechanical ventilation time (p<0.05) than Group 1. Additionally, Group 2 received a significantly higher amount of transfused ES and FFP (p<0.05) than Group 1. The postoperative admission rate and the dialysis history rate did not differ significantly (p>0.05) between both groups (Table 1).
The APACHE-II scores for predicting the mortality rate did not differ significantly (p>0.05) between the groups. Group 1 had a significantly higher exitus rate (p<0.05) than Group 2 (Table 2).
The sodium, potassium, chloride, magnesium, blood urea nitrogen, creatinine, alanine aminotransferase, aspartate aminotransferase, lactate, and C-reactive protein values at admission did not differ significantly (p>0.05) between the groups (Table 3). Group 2 had significantly higher calcium and albumin levels (p<0.05) than Group 1 (Table 3).
DISCUSSION
Advanced life support, mechanic ventilation, and organ support systems lengthen ICU stays for critically ill patients21. This six-year retrospective study aimed to determine the predisposing factors contributing to prolonged ICU stays. We selected a maximum stay limit of 30 days after conducting a literature search. We searched the hospital data system to determine if prolonged ICU stays were influenced by basic laboratory parameters, the need for vasopressors and mechanical ventilation upon admission, the need for tracheostomy and PEG, and related data, such as tracheostomy opening day, the number of tracheostomy days, PEG opening day, the number of days after PEG opening, transfused erythrocyte and fresh frozen plasma units, and renal replacement therapy during intensive care.
The description of a prolonged ICU stay in the literature varies between 7, 14, 30, and 90 days. For example, Miniksar and Keten22 conducted a retrospective analysis and discovered that 3.11% of patients stayed in ICUs for more than 90 days. Alkali et al.20 set a 14-day maximum limit for prolonged ICU stays and discovered that 401 (40.34%) out of 994 admitted patients had prolonged stays. In another study, it was discovered that 6% of 3257 patients stayed for 14 days or more in the ICU23. There are also studies that set a limit of 30 days or more for prolonged ICU stays and discovered that 1.6% and 4.92% of the patients stayed in the ICU for those periods, respectively24,25. In this study, we discovered that 89 patients (3.16%) out of 2811 patients stayed in the ICU for more than 30 days.
In this study, the age and gender distributions of the patients were not statistically different between the two groups. Çevik and Geyik17 performed a retrospective study and reported that two groups with ICU stays of less than and more than 30 days had a statistically significant difference in age but not in gender. In another retrospective study, it was reported that there was a statistically significant difference in mean age between groups with prolonged ICU stays of more than 14 days; however, Arabi et al.19 reported no significant difference in mean age and gender between groups of patients who stayed in the ICU for less than and more than 7 days21.
Prolonged mechanical ventilation requirements and unsuccessful weaning are the major indications for tracheostomy26. Several studies have reported that tracheostomized patients require a prolonged ICU stay17,19,22,27. This study also demonstrated that the tracheostomized patients stayed in the ICU for relatively long periods. An early or late tracheostomy time may be considered when managing the airway during intensive care. Patients who stayed in the ICU for less than 30 days had a mean tracheostomy opening time of 6.3±6.8 days, while patients who stayed in the ICU for more than 30 days had a mean tracheostomy opening time of 22.3±14.9 days. The patients with an ICU stay of more than 30 days had considerably long tracheostomy opening days, which may have contributed to their prolonged ICU stay.
From another viewpoint, more patients were recommended for tracheostomies as their length of stay increased. Even though the group with an ICU stay of ≥30 days had a relatively high number of days from tracheostomy to discharge (i.e., transfer to another unit or death), patient survival or the absence of a need for an ICU stay depends on several other parameters. Miniksar and Keten22 claimed that tracheostomized patients had much longer ICU stays. Similarly, Çevik and Geyik17 reported that tracheostomy procedures were a predictor of prolonged ICU stays and mortality. In this study, we did not find tracheostomy to be a predictor of prolonged ICU stays or mortality.
Malnutrition risks, nosocomial infections, and multiple organ failure are the complications that extend the ICU stay of patients with already prolonged stays28. PEG is usually used for patients who require long-term enteral nutrition because of inadequate oral intake. PEG tubes are much more comfortable for feeding and provide higher nutritional efficacy than nasogastric tubes29. In this study, PEG procedures were used more frequently on patients who stayed in the ICU for more than 30 days than on those who stayed for less than 30 days. We believe that this occurred because a PEG procedure was indicated for patients who were expected to have a prolonged ICU stay and required long-term enteral nutrition owing to inadequate oral intake.
Patients with hypotension or septic shock require vasopressors or inotropes for normotension upon admission to ensure adequate organ perfusion. Severe hypotension or septic shock is associated with increased mortality30. Previous studies have demonstrated that patients who require inotropes or vasopressors have prolonged ICU stays17,19. Contrarily, we discovered that more patients with ICU stays of <30 days required these drugs upon admission than those with ICU stays of ≥30 days. This finding may be because patients who require vasopressors or inotropes upon admission have a decreased survival rate.
Most patients require mechanical ventilation upon admission or during intensive care31. Zampieri et al.15 discovered that patients who required mechanical ventilation during intensive care needed to remain in the ICU for more than 14 days. Other studies in the literature have shown that the need for mechanical ventilation is effective in prolonging ICU stay19,32. In this study, 92.7% of the patients required mechanical ventilation regardless of how long they stayed in the ICU. There was no significant difference in mechanical ventilation requirements between the two groups. The mechanical ventilation duration is an important parameter because as it increases, mechanical ventilation-related complications, such as ventilator-related pneumonia and barotrauma, occur33. In this study, the patients who stayed in the ICU for more than 30 days had longer mechanical ventilation days. This may be explained by the fact that an increase in a patient’s mechanical ventilation days causes additional complications and makes patient discharge challenging.
Blood transfusions are associated with increased morbidity, increased mortality, and complications such as prolonged ICU stays, increased mechanical ventilation requirements, and even multiple organ failure. Health care costs have increased, and patients who require intensive care are unable to receive it because there are not enough beds available21,34,35. Tobi and Amadasun36, Halawi et al.37, and Lipschitz38 reported that a prolonged ICU stay was a predisposing factor for blood transfusions. Similarly, in this study, elderly patients and those with prolonged ICU stays received significantly more transfusions than patients with ICU stays of <30 days.
In this study, the two groups had no significant differences in admission after surgery, renal replacement therapy, APACHE-II scores, or predicted mortality rates. In a Japanese multicenter retrospective cohort study, patients with prolonged ICU stays had a significantly higher APACHE-II score of 23 than those with short ICU stays (APACHE-II score: 13)39. Various studies have also discovered a relationship between high APACHE-II scores and prolonged ICU stays18,40,41. However, Kıray et al.21 observed no differences in terms of APACHE-II scores between patients with prolonged and short ICU stays. Santana Cabrera et al.25 discovered that patients with surgical admission had prolonged stays, but Toptas et al.42 reported that patients with no surgical admission had prolonged ICU stays. Çevik and Geyik17 stated that renal replacement therapy did not affect the length of stay, but Santana Cabrera et al.25 stated that patients who received renal replacement therapy for long periods had long ICU stays. According to the literature, renal replacement therapy may prolong ICU stay because it prevents acute complications associated with acute renal failure43,44.
Although we discovered that the two groups had no difference in APACHE-II scores, the group with a prolonged ICU stay had a significantly higher mortality rate than the group with a short ICU stay. This result may be due to several factors, such as the opinion that critically ill patients who require high doses of vasopressors and are unresponsive to conventional treatments may not survive for a long time and may therefore stay in the ICU for less than 30 days. However, even if the patient’s status is quite stable at the time of admission, other factors, such as unsuccessful weaning, the need for prolonged mechanical ventilation, the reluctance of patient’s relatives to provide their consent for procedures such as tracheostomies and PEGs, the reluctance to transport tracheostomized patients home with home ventilators, and the unavailability of ward and palliative beds to discharge the patient, may make the patient vulnerable to other complications such as nosocomial infections, ventilator-associated pneumonia, and increased mortality. There are also similar opinions in the literature15.
Electrolyte imbalance is a common problem in critically ill patients upon admission45. In this study, despite investigating the electrolyte, renal, and hepatic parameters at the time of admission, we could not find a relationship between basic laboratory parameters and a prolonged ICU stay, except for calcium and albumin levels. This is because the length of ICU stays increases as the calcium and albumin levels increase. Toptas et al.42 discovered that patients’ length of ICU stay increases when their urea, creatinine, and sodium levels increase. Miniksar and Keten22 also reported that hypomagnesemia was significantly related to a prolonged ICU stay.
Study Limitations
In this single-center study, our case number was low compared to similar studies in the literature. The deficiencies in various scoring systems and their records in the hospital information system from past to present have been one of the limiting steps of our study.
CONCLUSION
In this six-year retrospective study, we aimed to determine the predictors of prolonged ICU stays. We discovered that patients who stayed in the ICU for more than 30 days had a relatively high mortality rate. Additionally, tracheostomy and PEG procedures were found to be related to prolonged ICU stays in our study. The absence of a need for vasopressors was also found to be associated with prolonged ICU stays. Prolonged ICU stays are caused by multiple factors, and palliative care units and home care facilities must be used frequently to make the best use of ICU beds and prevent prolonged ICU stays, which cause increased mortality and negative financial outcomes.
Ethics
Ethics Committee Approval: Ethical approval of this study was obtained from Tekirdağ Namık Kemal University Non-Interventional Research Ethics Committee (protocol no: 2022.191.10.15, date: 25.10.2022).
Informed Consent: Retrospective study.
Peer-review: Externally and internally peer-reviewed.
Authorship Contributions
Surgical and Medical Practices: O.B., Concept: O.B., Design: O.B., Data Collection or Processing: O.B., A.G., Analysis or Interpretation: A.Ş., A.G., C.A., Literature Search: A.Ş., C.A., Writing: O.B.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study received no financial support.