Impact of Comorbidities and COVID-19 on the Survival Rate of Geriatric Patients Receiving Anesthesia Services: A Prospective Cohort Study in Indonesia

After that, we used a Kaplan-Meier survival analysis to describe the survival rate. If p<0.05, it was displayed in gure form.


Conclusion
Approximately four in ten elderly patients receiving anesthesia died, and the percent increased when the patients had comorbidities and COVID-19.

Background
Chronological age is de ned as the actual number of years a person has lived. With increasing chronological age, physiological functions are expected to decline, which signi cantly contributes to older people's susceptibility to illness. 1 The alterations vary in each individual and may affect multiple organ systems. Changes in the nervous system are in the form of a diminished cognitive function. In the cardiovascular system, it presents as an increased mean arterial pressure and a decreased cardiac output. In the respiratory system, there may be a decrease in arterial oxyhemoglobin. In the digestive system, the emptying process in the gut may decelerate due to reduced hepatic metabolism. In the renal system, older people may have a decreased glomerular ltration rate. 2 As the global economy improves and as nations provide more higher quality healthcare, the year by year life expectancy grows as opposed to the decreased mortality worldwide, especially in Indonesia. [3][4][5] By 2025, the World Health Organization predicts that approximately 1.2 billion people will be 60 years or older around the globe. The Indonesian Population Projection Study in 2013 supported the idea that by 2035, the population aged 60 years and older in Indonesia will increase to nearly 14% of the population. [5][6][7] Consequently, the growing population of geriatric patients means that there will be an increased number of patients receiving medical and operative health services, including anesthesia. Dominguez et al. revealed that people over 65 years of age underwent surgeries more frequently than in patients who were 65 years or younger. 8 Geriatric patients tend to have many more comorbidities due to decreased physiological and cognitive functions. These may impact the perioperative concerns and outcomes. 9 These factors will also affect the pharmacodynamics and pharmacokinetics of the anesthetic drugs, particularly during surgery when many anesthetic drugs are used. The combination of having comorbidities and a weaker drug metabolism ability causes the administration of anesthetics to become more challenging and requires more acute consideration from anesthesiologists, as anesthetics may cause undesirable adverse outcomes both intraoperatively and postoperatively. 10 In the era of the COVID-19 pandemic, elderly patients must be highlighted because of the challenges in this population, including the frailty of the patients, and safety is a global health concern. Elderly individuals are prone to having a high risk of contracting COVID-19, suffering from worse outcomes, and being at risk of neglect from their chronic conditions. 11 In Latin America, COVID-19 infection raises the morbidity and mortality rate to 20% in surgical patients. 12 In planning a good clinical management program in this population, the anesthesiology staff needs to know the clinical characteristics and prognosis of geriatric patients who receive anesthesia services. Unfortunately, in Indonesia, there is still a lack of su cient studies that determine the prevalence and survivability in geriatric patients receiving anesthesia services. This study described the characteristics and analyzed the survival rate of elderly patients who received anesthesia services, especially those with comorbidities and COVID-19.

Study Design
This study is a prospective observational study. This research was conducted in 14 general hospitals in Indonesia, which also act as teaching hospitals for Anesthesiology and Intensive Therapy Residency Education. The data collection was carried out for four months, and every collector was briefed and had standardized training before the data collection.

Sample and Population
We analyzed the patients who underwent surgery and required anesthesia, and the patient cohort included both COVID-19 and nonCOVID-19 patients. We included all patients aged 60 years old or older in the period between February and May 2021. We excluded all patients who did not complete the follow-up between the observation periods.

Parameters
The variables included the patient characteristics, comorbidities, and COVID-19 status. The patient characteristics included age, sex, body mass index, comorbidities based on the Charlson's Comorbidity Index (CCI) score, the COVID-19 status, the American Society of Anesthesiologist (ASA) Classi cation, emergency or elective surgery, type of surgery, anesthesia techniques, sedation techniques, adverse outcomes in or outside the operating room, postoperative ICU care, cognitive function, and quality of life.
The Charlson Comorbidity Index (CCI) score was divided into four categories: 0, 1-2, 3-4, and >5. 13 The adverse outcomes that occurred in or outside the operating room were divided into intraoperative adverse outcomes and adverse outcomes in the recovery room, ICU room, and ward. The cognitive functions of the patients were measured with the Mini-Mental State Exam (MMSE) score. The MMSE score consists of ve domains: orientation, registration, attention and calculation, recall, and language. 14 The quality of life was measured with the World Health Organization Quality of Life (WHOQOL)-BREF. The WHOQOL-BREF score has four domains: physical health, psychological, social relationships, and environment. 15 For the MMSE score and WHOQOL-BREF, we calculated the average score of each participant.

Outcome measure
The output variables in this study included the mortality within 30 days after surgery. The participants were followed prospectively for 30 days, and whether the participants had died or not and the survival rate were recorded.

Statistical Analysis
The statistical tests were performed in the SPSS Statistics software program. Numerical variables with a normal distribution are shown as the mean ± deviation or otherwise as the median (minimum-maximum). Categorical variables are displayed in the form of a number (percentage). Next, we summarized the data in the form of tables and narratives. After that, we used a Kaplan-Meier survival analysis to describe the survival rate. If p<0.05, it was displayed in gure form.

Ethical Feasibility
Overall, this study was granted an approval by the local ethics committees and institutional review boards Medical and Health Research Ethics Committee (MHREC), Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada − Dr. Sardjito General Hospital with approval number KE/FK/1381/EC/2020. Written informed consent from all subjects, a legal surrogate or the parent was waived by the ethics committee.

Results
A total of 2091 patients were recruited into this study, 470 patients were excluded due to incomplete data, and 1621 patients were analyzed (see Figure 1). The average age of the patients was 67.1 ± 6.2 years, and most of them were male. On average, the patients had an ideal body mass index, and ninety percent of the patients had an elective surgery, with the most of them having an ASA classi cation of II. The most common type of surgery was oncological (17.2%), with general anesthesia being the most commonly used technique. Sedation techniques were most commonly used for therapy (see Table 1). Based on the place where adverse outcomes occurred, hypotension was observed in the intraoperative and recovery rooms (12.7% and 4.0%), a prolonged length of stay was observed in the ICU (18.1%), and pain was observed in the ward (3.6%). Thirteen percent of the geriatric patients after surgery were admitted to the ICU (see Table 1).
The cognitive function was measured with the MMSE score. Before surgery, the mean MMSE score was 25.9 ± 3.5, and the mean had increased to 26.3 ± 3.5 thirty days later. Therefore, the quality of life measured with the WHOQOL-BREF had an average score before surgery of 86.7 ± 63.1 and had a decreased mean of 68.9 ± 19.2 thirty days later.
The 30-day mortality rate was 4.4%, and the mean time of death was 15.3 ± 8.8 days after surgery (see Table 1). The most common comorbidity was hypertension, which was found in approximately 30% of the patients. The patients with CCI scores of 3-4 had a higher mortality rate (15.3%). The percent of patients with COVID-19 who died was 26.6%. The comorbidities with the highest mortality were dementia, rheumatologic disease, liver disease, previous myocardial infarction, and diabetes with chronic complications ( Table 2).  Figure 3).

Discussion
In this study, the average age of the patients was 67 years old, which is lower than the estimated life expectancy in Indonesia, which is approximately 72 years according to the World Health Organization. 16 Age-related diseases such as cerebrovascular disease, coronary heart disease, and diabetes are signi cant causes of death in geriatric patients in Indonesia. 17 The increasing number of aging demographies has caused a signi cant obstacle to anesthetic services because as the number of older people increases, more surgical cases will be performed. Unfortunately, the problem is exacerbated by limited healthcare facilities.
Our study shows that the most common surgeries conducted are oncological, orthopedic, and digestive .3%, 28.2%, and 9.4% of geriatric patients, respectively. This nding indicates that general anesthesia is routine for anesthesia services. However, it also indicates that the provider from the anesthesia service was limited in implementing the concept of balancing anesthesia based on the anesthetic safety.
Traditional preoperative risk assessments may often neglect geriatric-related syndromes, which can increase the morbidity and mortality in older patients. 20 According to Nayak and Varghese 20 , patients that are categorized as ASA IV and ASA V have worse morbidity and mortality than patients categorized as ASA I but that there is no difference in the outcomes compared to patients categorized as ASA II and ASA III. Meanwhile, the outcomes in patients categorized as ASA I, ASA II, and ASA III were not signi cantly different after surgery. 20 This result is in line with this study, which shows that patients with a higher ASA score have an increased mortality proportionally. This result also applies to the ASA score in patients in an emergency setting.
The Charlson's Comorbidity Index is associated with all-cause mortality after discharge. 21 Initially, the CCI was performed to classify comorbid conditions in patients and to determine the effects of the comorbidities on the risk of 1-year and 10-year mortality rates. 13 Today, the CCI is used for several outcomes in surgery or in intensive care settings. 21,22 The CCI may predict surgical outcomes such as 30day mortality, 1-year mortality, 30-day readmission, complications, and failure to rescue. An increasing CCI score can increase the mortality by 16.9% 21 months after surgery. 21 In our study, we classi ed the patients' scores into four groups: group 1 (score 0), group 2 (score 1-2), group 3 (score 3-4), and group 4 (score >5). An increase in the group is associated with an increased mortality rate, but the mortality rate decreases from group 3 to group 4. Huang et al. 23 found a decreasing mortality rate in type 2 diabetic neuropathy patients with scores of 3-4 to 5 (56.7-22.3%), but the survival rate decreased with an increasing CCI score. Several studies have shown that an increasing mortality is associated with patients with increasing CCI scores for 30-day mortality. However, longer follow-up periods, such as periods of 1 year and 10 years, are more commonly used to identify the risk for mortality in postoperative patients. [24][25][26][27] If we extend our study follow-up time, the mortality rate could increase, which could align with an increasing CCI group.
In our study, a higher CCI score did not indicate a higher mortality rate. For example, dementia and rheumatologic disease (score 1) have a higher mortality rate than moderate or severe liver disease (score 3). 13 Many studies that have used the CCI to predict postoperative mortality in the short term have used an average CCI score rather than just using a single score. 25,26,28 Many studies have explained how the CCI should be interpreted using average or categorical data, but a single component of the CCI score is rarely used to predict the outcomes. 13,22,[25][26][27] The CCI score is used not only to predict the mortality risk but also to predict a prolonged length of hospital stay, and it can predict an increased number of complications that may occur or patient readmission after surgery. 21,29 The postoperative care begins at the end of surgery, from the discontinuation of the anesthetic agent to recovery, from the operating room to the ward, and from the ward to hospital discharge. 30 Several complications can occur, such as respiratory depression, hypoxia, hypotension, hypertension, myocardial infarction, dysrhythmias, hypothermia, postoperative pain, and PONV. 30 In this study, we found that the most common incidents in the ward after surgery were postoperative pain, followed by hypotension, PONV, hemorrhage, and oxygen desaturation. Oxygen desaturation was the most signi cant cause of mortality. According to Almayrac and Sache 30 , the incidence of respiratory distress was 70% in the ward versus 23% in the operating room.
In 1950, anesthesiologists began to document changes in the cognitive function of patients after surgery and anesthesia. These changes include delirium and postoperative cognitive dysfunction (POCD). 31 The postoperative delirium can be measured with the preoperative and postoperative MMSE. An increase in age causes an increase in the postoperative delirium events. The cutoff MMSE score of predicted postoperative delirium is much lower in patients aged 80 or younger than in those patients who are older than 80 years old. It has been proposed that the younger the geriatric patients are, the more severe the impairment that is needed to cause postoperative delirium. 32 Our study found that the MMSE score after surgery (26.3 ± 3.5) was higher than before surgery (25.9 ± 3.5). This result may be because our participants were much younger, 67.1 ± 6.2 (<80 years old). Liu et al. explained in their systematic review that the administration of phencyclidine hydrochloride might increase the incidence of POCD in geriatric patients, based on its dose during anesthesia. Giving phencyclidine hydrochloride at a minimal dose has a similar incidence of POCD as the administration of scopolamine. Furthermore, a dose of 0.01 mg kg -1 of phencyclidine hydrochloride could diminish the cognitive function of patients after surgery to a signi cant degree. They also claimed that the patients who underwent general anesthesia for orthopedic surgeries tended to experience POCD 30 days after the surgery compared with the other types of anesthesia (39% vs. 4%). Prolonged anesthesia, de ned as anesthesia that takes more than 5 hours, could increase the incidence of POCD in elderly patients compared to those patients who had anesthetic events that were less than 5 hours. 33 As people reach 70 years of age or older, the morbidity and death due to surgery also increase. The preoperative risk factors are signi cantly increased with an increasing age and certainly increase the morbidity and mortality. The renal system, cardiovascular system, and respiratory system are signi cantly associated with morbidity and mortality. 34 According to the study conducted by Turrentine et al. 34 , geriatric patients who underwent surgery had a 28% morbidity rate and 2.3% mortality rate. Morbidity and mortality occurred more frequently in people older than 80 years of age. In addition, this study found that 4.4% of patients had died 30 days after surgery.
We found that several factors impacted the 30-day survival mortality rate in our study. Some of these factors include the COVID-19 status and several comorbidities. This study aligns with a previous study, which showed that the COVID-19 status was also associated with mortality, with a 23% mortality rate among hospitalized geriatric patients with COVID-19. 35 The six comorbidities that affected the mortality rate in this study were in concordance with a previous study that showed that cerebrovascular disease, hemiplegia, and tumor metastases have signi cantly caused death in emergency abdominal surgery patients. 21 We found several limitations in our study. First, there were obstacles during the preoperative data collection, such as incomplete or missing records. Second, most hospitals were general hospitals with adequate anesthesiologists, and these hospitals served as education hospitals where they hosted residents and higher-level consultants. Nevertheless, apart from the limitations mentioned above, our results may su ciently represent the general geriatric patients who receive anesthetic services in Indonesia, especially during the era of the COVID-19 pandemic.

Conclusion
The ndings of this study provide important information regarding the comorbidities that are important for the treatment of geriatric patients receiving anesthesia services. In addition, the impact of COVID-19 on elderly patients showed that these patients had an increased mortality rate and a decreased survival rate in Indonesia. Gadjah Mada -Dr. Sardjito General Hospital (KE/FK/1381/EC/2020). All methods were carried out following National Guidelines on ethical standards and procedure for research with a human being.

Consent for publication
Every subject had consent for publication using our institutional consent form.

Availability of data and materials
Not applicable

Competing interests
The authors declare they have no competing interests.

Funding
This study was supported by The Indonesian College of Anesthesiology and Intensive Therapy (KATI) and The Indonesian Society of Anesthesiologists and Intensive Therapy (PERDATIN).
Authors' contributions NMR, SC, DS, MIL, TGAS, and HN contributed to modelling the design of the study, analyzed and interpreted data, and wrote the manuscript. BW, BP, APP, I, NA, KMS, RAH, ZKJ, AHN, M and P contributed to the concept of the study, collected the data from the patients, and interpreted the data. All members of the authorship group read and approved the nal manuscript.