Post-ICU Stress Symptoms Associated with Increased Acute Care Service Utilization

By Betty Tran, MD, MSc, Editor

SYNOPSIS: In this prospective, longitudinal study of adult medical-surgical ICU patients, in-hospital substantial acute stress symptoms were associated with a greater risk of rehospitalization within 1 year post-ICU discharge; those with substantial post-traumatic stress disorder symptoms at 3 months post-ICU also had a greater risk of future emergency department visits within the year.

SOURCE: Davydow DS, et al. Psychiatric symptoms and acute care service utilization over the course of the year following medical-surgical ICU admission: A longitudinal investigation. Crit Care Med 2014;42:2473-2481.

Over the past two decades, advances in critical care have resulted in more patients surviving to hospital discharge, but these successes are attenuated by several sequelae of critical illness, including depression and post-traumatic stress disorder (PTSD).¹ Risk factors for and the health care ramifications of these disorders are poorly understood. Given this, Davydow and colleagues aimed to investigate whether PTSD symptoms in the acute (< 1 month) ICU hospitalization period and PTSD and depressive symptoms at 3 months post-ICU were risk factors for future hospitalizations and emergency department (ED) visits.

Over a 12-month period (2010-2011) at Harborview Medical Center (Seattle, WA), 150 adult patients admitted to medical-surgical ICUs were prospectively recruited. The primary exposure variables were the presence of acute stress symptoms, as assessed by the PTSD Checklist-Civilian version (PCL-C) prior to hospital discharge and at 3 months post-ICU, and depressive symptoms at 3 months post-ICU, as ascertained by the Patient Health Questionnaire-9 (PHQ-9).^2,3 Substantial acute stress symptoms were defined as scoring > 3 on at least one intrusive symptom, three avoidant symptoms, and two arousal symptoms on the PCL-C; substantial depressive symptoms were defined as a PHQ-9 score of > 10.

Baseline interviews were conducted a median of 8 days (IQR 5–15) after hospital admission. Primary outcomes were number of hospitalizations and ED visits between discharge and 12 months post-ICU.

After adjustment for baseline patient-related (including history of major depression, alcohol/drug use, Charlson comorbidity score among other variables) and hospitalization-related (including mechanical ventilation duration, admission diagnosis, number of surgeries) characteristics, substantial PTSD symptoms in the hospital were associated with a greater risk of rehospitalizations within 1 year of ICU discharge (relative risk [RR], 3.00; 95% confidence interval [CI], 1.80-4.99); there was also a non-significant trend toward increased risk of future ED visits (RR, 1.94; 95% CI, 0.95-3.98). In addition, substantial PTSD symptoms at 3-months post-ICU were associated with a greater risk of ED visits within the year after ICU discharge (RR, 2.29; 95% CI, 1.09-4.84), even after additional adjustment for acute care service utilization in the 3 months post-ICU discharge. Depressive symptoms at 3 month follow-up were not associated with risk of rehospitalization or additional ED visits between 3 and 12 months post-ICU.

COMMENTARY

This study adds to the growing body of literature describing the post-discharge needs of ICU survivors as substantial. Inarguably, critical care interventions, such as low tidal volume ventilation for acute respiratory distress syndrome and early antibiotics and fluid resuscitation in sepsis, have saved lives. However, in the same way that heart disease and cancer have replaced infection as the leading causes of death, our achievements are tempered by the generation of a growing population of patients with chronic medical problems as a result of their critical illness: psychiatric disorders, profound neuromuscular weakness, endocrinopathy, malnutrition, increased vulnerability to infection, functional disability, and symptom distress. This study highlights the need for focused attention, follow-up, interventions, and research in the post-ICU period. These efforts should not only help individual patients manage their symptoms and comorbidities, but would hopefully have beneficial effects on both preventing further disability and complications, as well as reducing health care costs through decreasing rates of utilization of acute care services such as recurrent hospitalizations and ED visits.

REFERENCES

1. Zimmerman JE, et al. Changes in hospital mortality for United States intensive care unit admissions from 1988 to 2012. Crit Care 2013;17:R81.
2. Weathers FW, et al. The PTSD Checklist-Civilian Version. Boston, MA, The National Center for PTSD, Boston VA Medical Center, 1991.
3. Kroenke K, et al. The PHQ-9: Validity of a brief depression severity measure. J Gen Intern Med 2001;16:606-613.

ABSTRACT & COMMENTARY

A Guide to When and How to Stop CPR

By Betty T. Tran, MD, MSc, Editor

SYNOPSIS: In cardiopulmonary arrest situations, the mnemonic CEASE (Clinical features, Effectiveness, Ask, Stop, Explain), provides a guide for clinicians on how to discontinue resuscitative efforts and effectively communicate with other clinicians and families.

SOURCE: Torke AM, et al. CEASE: A guide for clinicians on how to stop resuscitation efforts. Ann Am Thorac Soc 2015;Feb 9 [Epub ahead of print].

Although health care providers undergo hours of training and recertification to provide resuscitative efforts for patients in cardiopulmonary arrest, few are given guidance in terms of when and how to stop it. On the basis of available clinical evidence and ethical principles, Torke and colleagues aimed to provide a framework by which clinicians can organize their thinking about when to discontinue resuscitative efforts, which includes communicating effectively with families. Their proposed guide is summarized by the mnemonic CEASE (clinical features that predict survival):

• Clinical features that predict survival: Knowledge of a patient’s clinical history is critical, as pre-arrest factors such as age, metastatic cancer, poor functional status, renal insufficiency, hypotension, and non-cardiac diagnosis are associated with poor neurologic and/or survival outcomes.

• Effectiveness of resuscitation efforts: There is no standard decision aid to stop in-hospital resuscitation efforts, but the length of resuscitative efforts and the patient’s physiological response in real time should be assessed as part of the decision to continue or stop resuscitative efforts. For example, initial ventricular fibrillation or pulseless ventricular tachycardia is associated with better outcomes than asystole or pulseless electrical activity, and survival has been reported to be inversely proportional to resuscitation times.

• Ask the other clinicians present for input: Good communication among team members involved in the resuscitative effort is necessary to exchange relevant knowledge in real time; a collaborative, non-hierarchical environment should be the goal.

• Stop resuscitation efforts: It is the responsibility of the code leader to decide when to stop resuscitation efforts if the efforts are unsuccessful or the interventions needed to support circulation are unsustainable. Although this is a clinical judgment based on objective as well as subjective information, it is important to note that it is a decision made by the team leader and not within the purview of the patient’s family members.

• Explain to the family what has happened: At the end of the resuscitative efforts, the care team is obligated to inform family members what occurred, answer questions, and provide emotional support. This should be done compassionately and involve core skills that can be taught.

COMMENTARY

The article by Torke and colleagues provides an organized approach to discontinuing resuscitative efforts in cardiopulmonary arrest situations. For clinicians who work in the critical care environment, the framework presented is intuitive, and done enough times, almost reflexive. In some situations, stopping CPR is not a difficult decision, especially if we know beforehand that it is unlikely to be beneficial (e.g., the patient with terminal illness who is unlikely to survive CPR, the patient admitted with septic shock who is already on maximum vasopressors). Ideally, resuscitative efforts would be avoided in these situations altogether, but this is dependent on various factors during goals of care discussions with patients and families. During the other times, resuscitative efforts may last longer, especially if we believe that the underlying cause can be reversed. Overall, I suspect many of us think through the clinical situation in our heads, view the resuscitative results and discuss with our colleagues in real time, and debrief with family afterwards, all of which occur without having to consciously deliberate the individual steps.

On the other hand, this schema is probably most useful for physicians-in-training and other clinicians who have fewer encounters with critically ill patients and/or cardiopulmonary arrest situations. I have often witnessed residents excitedly lead code resuscitative efforts (with or without referencing their Advanced Cardiovascular Life Support pocket cards, which do not provide an endpoint), only to continue efforts to no end much to the discomfort of nursing and other ancillary staff. This is likely motivated by lack of experience, fear of stopping too soon, and as the authors note, “tremendous momentum to continue [advanced treatment interventions].” Although the CEASE mnemonic is not a decision rule to substitute for clinical judgment, it provides an organized approach to handling resuscitative efforts until more experience is gained.