Surgical diseases are massive burden globally with an estimated 300 million surgical procedures undertaken each year. These interventions are by no means a measure of the technical skill of the surgical team, but are also heavily dependent on the quality of the peri- and post-operative care, especially the post-operative care of patients.(1, 2) High-risk patients benefit from more intensive monitoring and organ support, and life-saving measures, in (ICU) which serves as the center of postoperative care.(3, 4)  Prompt access to ICU resources is a key factor in patient safety and early recognition and intervention for physiological changes that can quickly become life-threatening.(5) The first thing that should happen in an ideal health care system is to admit patients to the ICU whenever a need is recognized. But in the real world of tertiary care centers, especially in low and middle-income countries (LMICs), realizing this vision is frequently not possible. A mismatch between the number of intensive care beds and demand often occurs in these settings. (6) This is exacerbated by insufficient numbers of critical care staff and equipment and monitoring, as well as an inefficient patient flow process. (7) This results in the ‘boarding' phenomenon where seriously ill patients stay in post-anesthesia care units (PACUs), emergency department or general ward because a suitable bed is not available in the ICU. This time lag can cause considerable stress for patients and health care professionals and be a risky situation.

There are several physiological effects of a late admission to the ICU. Vulnerable postoperative patient is at significant risk for hemodynamic instability, respiratory compromise, metabolic derangements. (8) Inadequate and delayed implementation of comprehensive intensive monitoring can lead to a "failure to rescue" situation when a treatable complication like occult hypovolemia and incipient sepsis develops into irreversible organ failure. Data from developing countries, however, are limited, because of more severe resource limitations. The global dynamics of bed availability, triage and its effect on patient outcomes are likely to be more complex and/or more harmful in these settings. (9)

As in most developing countries, Pakistan's health care system is beset with tremendous challenges. Public sector tertiary hospitals (usually serving large population base) are often facing extremely low bed occupancy rates in ICUs.(10) Admission to the ICU may be determined by a mix of clinical urgency, the attending physician's decision and bad news – bed availability.(11) This process may result in the triage of patients on the basis of what may or may not be considered best practice, or on the basis of "salvage ability".(12) Based on our hypothesis, we believe that these delays in admission to the ICU correlate with worse postoperative outcomes such as higher mortality and major complications and that these outcome differences are statistically significant when the delay is longer than two hours.

2.1 Study design and duration A retrospective study was conducted in the Intensive Care Department, with the support and collaboration of the Surgical Unit, at a tertiary care hospital in Abbottabad, Pakistan. This hospital is a main referral hospital for Hazara division and other areas. The study was conducted for one year (from 1st June 2023 to 31st June 2024). 2.2. Sample size The study population consisted of all adult patients (age > or = 18) referred to the ICU after elective surgery or emergency surgery. The total number of patients was determined by a convenience sampling technique, depending on the feasibility of data extraction in the required time period and the number of patients likely to be referred to the ICUs. The patients were all screened for admission to the ICU for the duration of the study. Patients included were adult (18 years and older) patients undergoing an elective or emergency surgery, and referred to the ICU after the surgery. Those patients with complete and accessible medical records were included. Exclusion criteria were: patients with incomplete medical records, patients admitted to the ICU for non-surgical reasons, patients transferred to the ICU from another hospital, and patients with do-not-resuscitate (DNR) orders and/or a life expectancy of less than 24 hours from an underlying terminal disease. 2.3. Data Collection Procedure Trained medical officers of intensive care unit collected data from hospital record maintained in hospital medical record room. To achieve this uniformity a standardized data abstraction form was drawn up. Patients in the two cohorts for comparison were pre-defined: Timely Admission Group (Exposed): Admitted patients in the Intensive Care Unit within less than 2 hours after the referral order. Delayed Admission Group (Unexposed): patients admitted to the ICU > 2 hours after the referral order. This time frame of two hours was chosen because of previous studies and clinical consensus that substantial physiological changes may occur within this time frame. 2.4. Variables and Measurements Baseline information consisted of age, gender, BMI, ASA status, surgical specialty, surgical priority, primary diagnosis, and type of procedure. Illness severity was determined by the APACHE II score which reflects the worst physiological measurements within the first 24 hours in the ICU. The main outcome was in-hospital mortality, an all-cause death. Secondary outcomes were major postoperative complications: ARDS, septic shock, myocardial infarction, and stroke. Also recorded were the length of stay (LOS) in days at the ICU. 2.5. Statistical Analysis The data were analyzed with SPSS 26.0 (IBM Corp., USA). Continuous variables were displayed as mean ± SD or median (IQR) and categorical variables were presented as frequencies and percentages. As appropriate, independent t-test, Mann–Whitney U test, Chi-square test, or Fisher's exact test was used. Variables with p < 0.2 in univariate analysis were included in multivariate logistic regression to identify independent predictors of in-hospital mortality. Results with p value < 0.05 were regarded as statistically significant. 2.6. Ethical Considerations The ethical approval was taken from the hospital Administration with all codal formalities.

This study is a retrospective cohort study that involved 150 patients. The overall mean age was 54.6 ± 16.2 years and the sex ratio was 58.7% in favor of male. The initial demographic data (Table 1) did not differ between the timely admission and delayed admission groups. However, there was heterogeneity in the clinical characteristics. The delayed group (DG) on admission was more physiologically unstable with a greater proportion of emergency surgery (54.7% vs. 36.0%, p=0.012) and mean APACHE II score (24.2 vs. 19.6, p=0.001). The delayed admission group had significantly higher rate of overall in-hospital mortality compared to the timely admission group (26.7% vs. 9.3%, p=0.006) (Table 2). As presented in Table 3, the death rate was higher than three times the rate in the patients who were admitted to the ICU later than scheduled compared to patients who were on time. To determine independent risk factors for mortality, multivariate logistic regression was carried out (Table 3). After adjusting for this, the OR for delayed admission to the ICU for the outcome of in-hospital mortality was 3.4 (95% CI: 1.4-8.2; p=0.007) indicating that delayed admission to the ICU was a strong and independent predictor of mortality in this setting. The APACHE II score also was found to be an independent predictor of mortality (Adjusted OR: 1.1; 95% CI: 1.01-1.2; p=0.043). This is also reflected in the Kaplan-Meier survival curve (Figure 1) where the cumulative survival probability of the delayed admission group was significantly lower than patients in other groups at 30 days follow up.

Table 1: Baseline Demographic and Clinical Characteristics of the Study Cohort

Data is presented as Mean ± SD, Median (IQR) or n (%). Statistically significant differences in text (p<0.05) are indicated by bold text. ASA:

Table 2: Univariate Analysis of Postoperative Outcomes

Data are given as n (%) or Median (IQR). Data that are significantly different from each other are presented in bold (p<0.05).

APACHE II, Emergency and ASA with p<0.2 value in univariate analysis were included in the model. Statistically significant differences (p<0.05) are shown in bold text.

DISCUSSION This is a retrospective cohort study which offers robust evidence in a resource constrained tertiary care setting in Pakistan that delayed admission to the intensive care unit is an important factor in determining the adverse postoperative outcomes. Our results showed a strong and independent relationship between delays in admission to the ICU more than 2 hours and higher in-hospital mortality, higher rates of major complications and longer lengths of stay in the ICU and hospital. These findings align with the overall international literature, which has been consistently pointing to the risks associated with lack of beds in the ICU and an inadequate patient flow.(13) The high mortality rate in the delayed group (26.7%) should alert the public to the risks of surgical patients having to wait for critical care. This rate is similar to or higher than other similar studies in other developing countries.(14) This increased mortality is due to multiple pathophysiological mechanisms. The post-operative period after significant surgeries is a very stressful period of time for the body. Patients can be affected by the negative influences of anesthetic, surgical trauma and pre-existing comorbidities. In the absence of the intensive monitoring and rapid treatment that occurs in the ICU, vital signs may not be monitored, and early signs of deterioration, such as reduction in urine output or an increase in lactate, go undetected. This can cause a chain of problems to occur. Acute kidney injury (AKI) for example is a complication of unrecognized hypovolemia and occurs in 20% of the delayed group.(15) Interestingly, delayed admission was an independent predictor of mortality even after controlling for the greater severity of illness (APACHE II) in this group, which is of critical clinical significance. It implies that there is an inherent risk in the delay other than just the patient's bodily condition. (16)It is acknowledged that the delay is more likely to occur in sicker patients, who we think are more likely to be delayed as a result of triage processes (waiting for a bed), however our multivariate analysis indicated that the delay was an independent factor and therefore delayed patients were more likely to have a worse outcome. This may be because of the negative aspects of the "boarding" itself; the "boarding" time - spent in the PACU or ward, may have less nurse/patient ratios and not be able to monitor the patient in real time as in the ICU, and so failure to rescue may occur with a higher frequency. Moreover, the delayed group had significantly longer ICU and hospital stays with significant implications for the use of healthcare resources. The additional two days of stay in the ICU, on average, observed in our study can contribute to a vicious circle of capacity strain and result in even more admission of subsequent patients. This "ICU bottleneck" is a frequent problem in tertiary hospitals and could mean cancellation or postponement of elective surgeries, adding to the strain on patients and the healthcare system. This is compounded by the presence of a number of systemic shortcomings in the healthcare landscape in Pakistan, and many LMICs. There is a huge demand for ICU services but there's not enough beds for them, and it's further complicated by lack of sufficient trained intensivists, nurses, and advanced equipment.(17) Admission to the ICU is sometimes made on an "educated guess" by the junior physician and the availability of a bed, not on a well-defined, evidence-based triage criteria.(18) However, since there are no structured step-down units or high dependency units (HDUs), the patients who do not qualify for an ICU bed but are at a high risk of deterioration are usually treated in general wards where they receive suboptimal monitoring.

This tertiary care center found that admission to the ICU more than 2 hours after operation was both strongly and independently associated with an over three fold greater risk of postoperative mortality. Patients who have a delayed treatment also have significantly more rates of major complications such as sepsis, acute kidney injury, and respiratory failure, and their length of stay in the ICU and hospital is longer. The results in this study underscore the importance of prompt access to intensive care in surgical outcomes and patient safety. These negative outcomes are caused by the lack of ICU beds and low efficiency of patient flow in resource-limited settings. Immediate systemic changes are necessary to ensure that critical care is now delivered in a timely fashion and not by chance.

Study Limitations

There are a number of limitations to this study: it was retrospective and single centered, and pre-existing medical records may be inaccurately documented and there is the possibility of unmeasured confounding. The choice of two hours for defining ‘delay' is somewhat arbitrary and may not fully represent the range of potential risk of different delays. Secondly, limited sample size (150 patients) may have limited statistical power for detection of differences in less common secondary outcomes and for detailed subgroup analyses. The results are also not generalizable to other tertiary care hospitals in Abbottabad, which may have different resources, patient populations or surgical case mixes. We also did not have the ability to measure some variables that may have affected patient outcomes, including the clinical judgment of the referring physician or the ability to use specific life-saving interventions in the waiting period.

Recommendations

Healthcare organizations should establish separate bed management teams and adopt standardized early warning scoring systems for quick triage and prioritization of patients who are at the highest risk of postoperative deterioration and require admission to the ICU. The creation of step-down or high dependency units can help to reduce the number of patients in the intensive care unit by offering a level of care that is not as intensive as that provided in the intensive care unit. Hospital management should focus on augmenting the number of ICUs, and on the critically high demand for trained intensivists and critical care nurses, recruitment and training should be targeted. Continuous audit systems should be set up to track the time required for referral to the ICU, delay in admission to the ICU, and patient outcomes which can be used to plan and implement iterative quality improvement plans. Finally, there is the need for national levels of admission criteria for ICUs and how patients are moved, so that there is equitable and timely access to critical care for all tertiary care facilities.