Chapter 41 – Postoperative Admission to the Intensive Care Unit

Chapter 41 Postoperative Admission to the Intensive Care Unit

Bertrand Guidet , Caroline Thomas , Guillaume Leblanc , Ariane Boumendil and Dominique Pateron

The patient is in intensive care but still alive.

Javier Perez de Quellar


Elderly patients account for a large proportion of all surgical procedures. Those who undergo high-risk surgery represent a major share of admissions to intensive care units (ICUs) (Nathanson et al. 2009). These patients often have pre-existing conditions that may lead to perioperative morbidity and mortality.

Postoperative outcome is influenced by the interaction of many different factors, such as the surgical procedure performed, preoperative patient health, intra- and postoperative events, available care options and organizational aspects, e.g., the volume of the performed surgical procedure carried out (Nathanson et al. 2009). Targeting patients who will benefit from the highest postoperative care is a challenging task for anesthetists and surgeons dealing with older patients undergoing major surgery, and appropriate patient triage to postoperative ICU is of paramount importance in cost-effective resource allocation and maximization of results (Valentin 2017).

This chapter will review the characteristics of elderly critical patients, criteria to target them for appropriate ICU admission, possible benefits, complications and outcomes, and will investigate the special needs of older patients admitted to the ICU.

The Elderly Critical Patient

Patients over 65 and 80 years respectively represent 40–50% and 10–20% of all ICU admissions. The level of perioperative care must take into account their physiologic progressive loss in functional reserves, their pathologic preoperative decline in functional status (FS), prior comorbidities and the severity of the acute illness.

Elderly patients are more vulnerable to acute stress due to age-related diminution of physiological reserves (see Chapters 1 and 32), which can make them unable to compensate for the increased metabolic demand accompanying acute illnesses. This imbalance can lead to increased rates of intensive care complications, morbidity and mortality. Indeed, as reported in multiple cohort studies, elderly patients undergoing major non-cardiac or cardiac surgery develop higher rates of multiple organ failure and infective complications than younger patients.

These higher rates lead to increased mortality. For patients 80 years and over, ICU and hospital mortality has been reported to be 10–30% and 25–40%, respectively. However, mortality rate is highly variable, in part depending on the reason for admission, but also reflecting biases or inaccuracy in preadmission evaluation. In many cases in fact, the acceptance or refusal of ICU admission is decided solely on the basis of chronological age, with no consideration for FS. However, a study investigating the outcome of ICU admission with the Mortality-Probability Model, found that, although the presence of serious comorbidities decreased the likelihood of survival to discharge for all age groups, advanced age alone did not preclude successful surgical and ICU interventions (Nathanson et al. 2011).

Elderly admitted to the ICU following elective surgery have a better prognosis than those admitted after emergent surgery. In a retrospective cohort study of 204 ICU survivors (81.7±2.4 years), 57% of patients who had planned surgery survived at one year, compared to 11% of patients who had an unplanned surgical reason for admission and 10% of medical patients (de Rooij et al. 2008). In a study by Bagshaw et al. (2009) 72% of ICU patients over 80 years were discharged home after elective surgery.

The characteristics of ICU stay of elderly patients are different from their younger adult counterparts in terms of length of stay (LOS), therapeutic intensity level and workload. Since elderly patients have a reduced physiological reserve and ability to recover, we might presume that the ICU and hospital LOS is longer than in younger patients. Interestingly, the ICU and hospital LOS for patients over 80 years are generally equivalent or shorter than those observed in younger patients (2,3). (Table 41.1). This is probably explained by a lower therapeutic intensity level for elderly patients, despite similar or higher initial illness severity (Boumendil et al. 2005). Another consideration is the probable impact of an earlier decision to withdraw organ support in critically ill elderly patients. However, these decisions should be made in accordance with the principles of patient self-determination, reasonable expectations and expected quality of life (QoL), instead of merely the chronological age.

Table 41.1 Characteristics of the intensive care unit stay for elderly patients compared to younger patients

Studies (dates) Age categories n ICU LOS (days) Hospital LOS (days) Illness Severity Score
Somme <80 184 9.9 N/A 20.5a
(1991–1996) 80–85 137 8.1 19.8
> 85 91 6.6 20.5
Euricus I (personal data) <80 14733 5.2 16.4 31.4b
80 1326 4.7 15.5 42.9
Euricus II (personal data) <80 14740 5 15.6 29 b
>80 2738 5 16.1 42.4
Boumendil <80 1224 5.7 N/A 32.8 b
(1998–1999) >80 233 6.3 45.1
Demoule 20–69 72 6 21.4 39 b
(1995–2001) < 90 36 5.1 27.3 52
Boumendil 65–79 3175 7.7 18.7 23.2c
(1997–2000) >80 3175 6 15.5 23.3
CUB-REA (personal data) <80 16705 7.3 21.6 41.5 b
>80 2940 6.3 17.6 54.2



c SAPS II (without age)

When Should an Elderly Patient be Admitted to ICU?

No specific guidelines are available for ICU admission of elderly patients. In 1994, the European Society of Intensive Care Medicine (ESICM) published guidelines stating that patients with unstable conditions or at risk of severe complications should be admitted to the ICU. However, these recommendations are designed for a general population, were issued more than 20 years ago and mirror a demographic situation totally different from that currently.

Cardiac, vascular, thoracic and other high-risk surgical procedures such as trauma- or neuro-surgery are those most frequently followed by the need for ICU admission. As resource allocation is a source of concern in this area, only patients who would benefit from an ICU stay should be admitted; however, no available effective criteria have yet been defined to correctly identify these patients. Patients are sometimes admitted postoperatively for monitoring only, but it is unclear whether this practice improves postoperative outcome. Extensive admissions could potentially lead to ICU overuse for postoperative patients. The right approach should probably emphasize the impact of ICU admission on long-term mortality, FS and QoL. Another important factor to consider in the decision-making process is whether or not the ICU admission is congruent with the patient’s care wishes.

Principles for Appropriate Triage

A surgical patient may need ICU admission due to known preoperative risk factors (risk associated with the surgical procedure or with the patient’s preoperative health), or due to unpreventable intraoperative events. In the first case, preoperative evaluation usually allows appropriate selection, whereas in the second it is the expected outcome.

The triage process involves many stakeholders, such as patients, their family or surrogate decision-makers and care-team members (surgeons, anesthetists and ICU physicians). Although not specific in estimating the need for postsurgical ICU admission, some predictors of perioperative adverse outcome have been identified, such as advanced age (≥85 years), highly invasive surgical procedure, recent need for hospitalization, emergency surgery, high ASA score and chronic heart failure. However, no validated score is currently available to predict the need for perioperative ICU admission for the elderly.

Ideally, the triage process should be based on criteria that are as objective as possible. However, objective preoperative information about baseline FS and perceived QoL is often lacking. A reassessment should then be performed early in the following days, after collecting more information on the past medical history and the patient’s wishes about life-sustaining therapies. This “ICU trial” strategy aims to avoid refusal of ICU admission on principle.

Perioperative Scoring Systems

In critically ill patients, a number of illness severity scores are used to predict outcome. The most commonly used scores are the Acute Physiology and Chronic Health Evaluation (APACHE), the Simplified Acute Physiology Score (SAPS) and the Mortality Probability Model (MPM). Even though these scores are not specifically designed for surgical or elderly patients, they account for surgical status and patient age. Other scores such as the Sequential Organ Failure Assessment (SOFA) and the Multiple Organ Dysfunction Score (MODS) assess both the presence and severity of organ dysfunction during an ICU stay. While these scoring systems are widely used in the ICU, they are not designed for the triage of patients before ICU admission and they do not take into account specific intraoperative events or complications. Other scores have been specifically designed to predict the morbidity and mortality in the surgical population, but none uses ICU admission as a variable or as an outcome measure. Moreover, all of these scores may not be useful or accurate in assessing outcome for an individual patient.

In a recent study of 275 patients aged 65 years and over, undergoing intermediate or high-risk elective surgery, a multidimensional frailty score model (composed of the Charlson Comorbidity Index, dependence in activities of daily living, dementia, risk of delirium, short mid-arm circumference and malnutrition) predicted mortality rates more accurately than the ASA classification (84% sensitivity and 69% specificity). High-risk patients (defined as a multidimensional frailty score >5) showed increased postoperative mortality risk and longer hospital LOS (Kim et al. 2014).

Perioperative Complications

Although advanced age has not been clearly identified as a risk factor for postoperative major cardiac complications or death, elderly patients present a higher risk for adverse cardiovascular events, such as dysrhythmia and myocardial infarction. They also have a greater risk of pulmonary complications (atelectasis, respiratory failure, pneumonia), acute kidney injury and delirium. Moreover, intraoperative events (hemorrhage, myocardial ischemia, shock) may impact on postoperative outcome and affect the need for ICU admission. Emergency surgery, surgical duration and extremes of vital signs (high or low blood pressure, tachycardia) are all associated with adverse postoperative outcomes.

In a retrospective observational study on patients aged 80 years and over, undergoing emergency procedures (n = 178), the most common complications were infection (21%), cardiovascular (18%) and neurological (18%). In a retrospective study to identify risk factors for mortality in patients aged 80 years and older, admitted to ICU after surgery (n = 255), multivariate analysis showed the need for vasopressors in the first 48 hours after surgery to be the strongest predictor of hospital mortality (Ford et al. 2007).

Benefits and Outcomes

The increased rates of ICU complications, morbidity and mortality make the benefit of ICU admission uncertain. Routine postoperative ICU admission for elderly patients after high-risk surgery may permit earlier recognition and treatment of complications, thereby reducing morbidity and mortality. However, no randomized clinical trial has been conducted so far on this issue. Moreover, epidemiological and observational studies failed to provide clear evidence of benefit (Boumendil et al. 2011, Sprung et al. 2012). However, the interpretation of these studies is limited by a great variation (from 8% to 40%) in the percentage of elderly patients admitted to ICU, due to differences in medical practices, local policies and the availability of ICU beds. In a prospective cohort study (ICE-CUB 1), only 25% of patients with a critical condition were referred to the ICU by the emergency department physicians. Independent factors associated with the absence of ICU referral were advanced age, active cancer, low severity of acute illness and low score on the Activities of Daily Living scale (Garrouste-Orgeas et al. 2009).

ICU mortality rates are highly variable among different centers and countries, probably due to multiple factors, such as variable admission criteria, different clinical practices and differences in resources availability. Most studies found that age significantly impacts on overall mortality in critically ill patients; however, after adjusting for prior comorbidities, patient origin, initial diagnosis and severity of acute illness, the impact of age on mortality was significantly reduced. Moreover, hospital mortality in elderly patients discharged alive from the ICU was two times higher. Several hypotheses may explain this excess of mortality after an ICU stay: low physiological reserve and frailty, premature discharge from ICU or discharge to an inappropriate setting.

Few studies compared the elderly ICU mortality rate to a reference population. A single-center study of 299 patients over 80 years of age estimated that the two-year mortality of elderly patients admitted to the ICU was more than twice than that observed in the general population (Roch et al. 2011). Important determinants of elderly ICU mortality are difficulties in performing daily activities, mechanical ventilation, emergency admission, non-operative source of admission and a higher age-adjusted APACHE II score (Stein et al. 2009, Ryan et al. 2008).

Only gold members can continue reading. Log In or Register to continue

Jan 16, 2021 | Posted by in ANESTHESIA | Comments Off on Chapter 41 – Postoperative Admission to the Intensive Care Unit
Premium Wordpress Themes by UFO Themes