Delirium complications in post-general anesthesia : An observational study in China

Purpose: To evaluate emergence agitation and delirium signs in general anesthesia (GA) and postanesthesia care units (PACUs), and associated risk factors. Methods: Adult patients (n = 380) exposed to GA were recruited over a period of 1 year and five months for this study, and were assessed for emergence delirium (ED) using Richmond AgitationSedation Scale (RASS). Confusion Assessment Method for Intensive Care Unit (CAM-ICU) was used to assess delirium signs on admission to, and during stay in PACU at 30 min, 1 h, and at discharge. Signs consistent with delirium were classified as hyperactive or hypoactive based on a positive CAM-ICU assessment and RASS score. Multivariable logistic regression was used to assess potential risk factors for delirium such as age, American Society of Anesthesiologists (ASA) classification, and opioid and benzodiazepine (BZD) exposure during stay in PACU. Results: Emergence delirium (ED) occurred in 69 (18.2 %) patients out of which 41 (59.4 %) were also CAM-ICU +ve on admission to PACU, with 22 (31.9 %), 11 (15.9 %), and 6 (8.7 %) CAM-ICU +ve at 30 min, 1 h, and at discharge from PACU, respectively. Therefore, 28 (7.4 %) patients had ED with no associated signs of delirium. A total of 117 (30.8 %) patients had signs of delirium (CAM-ICU +ve) during admission to PACU, 58.1 % of whom had hypoactive features based on their RASS scores. Signs of delirium during stay in PACU were observed in 64 (16.8 %) patients, while 61 (16.1 %), 28 (7.4 %), and 14 (3.7 %) patients were CAM-ICU +ve at 30 min, 1 h, and at discharge from PACU, respectively. The results of multivariate logistic regression showed that total perioperative opioid administration (fentanyl equivalent) was independently associated with signs of delirium during stay in PACU, after relevant covariate adjustment (p = 0.03). However, age, BZD exposure (midazolam equivalent), and ASA classification did not show significant association with signs of delirium during stay at PACU. The result of sensitivity analysis showed that duration of anesthesia was independently associated with signs of delirium during stay in PACU. Conclusion: The results of this study suggest that in patients undergoing GA, signs of delirium are common in the immediate postoperative period, with incidence highest on arrival at the PACU and decreasing gradually during stay in PACU. Hypoactive features are common and more prominent during stay in PACU, when compared to hyperactive features.


INTRODUCTION
Delirium, also known as acute state of confusion, is a brain disorder characterized by modification of level of consciousness, disorganized thinking and inattentiveness, with signs which could be either hyperactive or hypoactive.Hyperactive signs include restlessness, rapid mood swings, and agitation, while the hypoactive signs include inattentiveness, sluggishness and lethargy.Delirium occurs in about 60 -80 % of patients under mechanical ventilation, and 20 -50 % of patients during their length of stay (LOS) in hospital [1][2][3].Indeed, studies have shown an independent association between brain abnormality and prolonged LOS, health care cost, extended cognitive impairment and increased incidence of mortality [4][5][6][7].
Recent studies have suggested that early postoperative delirium in PACU might be associated with poor prognosis [6].In addition, delirium occurring during stay in PACU might lead to supplementary delirium during postoperative care [7].The diagnosis of delirium is difficult because emergence from GA usually presents with similar signs which include changes in attentiveness, disorganized thinking and altered mental status.It has been hypothesized that signs of delirium are most frequent after GA, but they diminish with time [6].It has also been speculated that delirium signs may persist in a significant number of patients following discharge from PACU, and that hypoactive signs may be common [6].Patientspecific or anesthetic features such as advanced age, drugs and exposure have been shown to be closely associated with increased incidence of delirium [7].The aim of this study was to assess patients for development of ED in the PACU after exposure to GA, and the associated risk factors during stay in the PACU.

Patients and general information
This cross-sectional study received approval from Institutional Review Board of the Second People's Hospital of Huai'an (approval code: RSH112015) and informed consent was acquired from participants.The International ethical guidelines for health-related research involving humans was followed [9].Patient confidentiality was strictly maintained.Training in delirium assessment was given to four PAC nurses by the main investigator with experience in CAMICU [1] delirium assessments.Non-cardiac surgery patients in PAC admission post GA with volatile anesthetics were included.Patients were allocated by nurse in charge (round-robin method) to the PAC nurses.Therefore, that was a convenience sampling study of randomly allocated patients to four nurses.Exclusion criteria included patients with medical history of brain injuries, severe dementias, or neuromuscular disorders.
Non-cardiac surgery patients on volatile GA (n = 380) were recruited over a 1-year and five months period for this study.Patients in the following categories were excluded: (1) non-Chinese-speaking or deaf patients; (2) patients with history of severe dementia, anoxic brain injury or neuromuscular disorders; and (3) patients who did not sign written informed consent with their family members.The study protocol was approved by the Review Board of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Hubei, China.

Data collection
Data collected from the patients included deidentified demographics such as age, gender, race and ASA classification; health history such as comorbid diseases, chronic alcohol or illicit drug use, smoking status, details of anesthetic course such as length of anesthetic exposure and volatile agent used, induction agent and dose.Others were lowest intraoperative vital signs such as temperature, oxygen saturation, systolic and diastolic blood pressure; perioperative medications used such as BZD, opioids and ketamine; PACU vital signs, blood product administration, Aldrete scores and verbal pain scores.

Assessment and definitions
Face valid definitions for ED and PACU delirium signs were determined based on literature review and expert opinion.Emergence delirium was defined as agitation after discontinuation of the inhaled anesthetic based on RASS score of +1 to +4.Patients were assessed for delirium signs using CAM-ICU performed at the point of admission to PACU at 30 min, 1 h, and at discharge from the PACU.Delirium signs were deemed presentif patients were CAM-ICU positive at any time point.Hyperactive delirium signs were defined as RASS scores of +1 to +4 for agitated patients accompanied by a positive CAM-ICU.Hypoactive delirium signs were defined as RASS scores in the range of -3 to 0 for somnolent or calm patients, accompanied by a positive CAM-ICU.Postoperative delirium was defined as delirium that continued beyond the PACU or occurred in the hospital ward or ICU.

Results of multiple regression analysis
Results of multivariate logistic regression showed that total perioperative opioid administration (fentanyl equivalent) was independently associated with signs of delirium during stay in PACU after relevant covariate adjustment (p = 0.03).However, age, BZD exposure (midazolam equivalent), and ASA classification did not show significant association with delirium signs amid PACU stay.Result of sensitivity analysis showed that duration of anesthesia was independently associated with signs of delirium amid PACU stay.These results are shown in Table 3.The results showed that about 18.2 % of patients had ED after GA, and a higher incidence of signs of delirium was observed on admission to PACU with the patients exhibiting hypoactive features.While signs of delirium were significantly diminished during stay in PACU, 3.7 % of patients had ongoing delirium signs during discharge from PACU regardless of whether they met the criteria for discharge or not.This suggests that such patients might not be counted among those presenting signs of ED.
The number of vital surgeries performed globally per annum is higher than 230 million, and PACU delirium is suggestive of post-operative delirium.This means that a considerable number of patients who leave the PACU are at increased risk of delirium and delirium-related complications [4,7].In this study, total perioperative opioid administration was independently associated with signs of delirium amid PACU stay after relevant covariate adjustments.However, age, BZD exposure, and ASA classification did not show significant association with signs of delirium during stay in PACU.Results of sensitivity analysis showed that duration of anesthesia was independently associated with delirium signs amid PACU stay.Studies involving the assessment of unusual emergence pattern from GA have reported ED prevalence of nearly 5 % [14,15].It is not clear if these signs on admission to PACU are indicative of ED or are simply typical signs of emergence from GA.Unfortunately, there is no typical definition of complete emergence from GA.In this study, signs of delirium were prolonged in 7.4 % of patients 1 h after admission to PACU, and in 3.7 % of patients at point of discharge from PACU.These results are in agreement with those previously reported [10].Consideration should be centered on early detection and management of ED, since at the point of discharge from PACU, ED is usually associated with post-operative delirium and decline in cognitive function [6,7].In addition, patients with signs of delirium at point of discharge from PACU may have a type of ongoing delirium rather than the easily reversible sedative-related delirium which may even be worse [4].
The incidence of signs of delirium observed in this study is less than the 45 % reported in a previous study [13].In the study, elderly patients were assessed for post-operative delirium following complete recovery from anesthesia, and it was observed that the patients (>70 years) were predictably at increased risk of ED.Studies have shown that CAM-ICU is more specific than sensitivity analysis when used to assess ED in less seriously ill patients, particularly during their stay in PACU [4,13].Therefore, it appears that ED was under-diagnosed in this study.Although neuro-psychiatric assessment using Diagnostic and Statistical Manual of Mental Disorders (DSM) is usually considered as standard, it is capital-intensive and time-consuming [11].
In this study, CAM-ICU was used because of its simplicity and feasibility [4].Results of this study showed that pain scores at the point of admission into and discharge from PACU were minimal and comparable clinically in patients with and without delirium, an indication that pain might be a factor in the pathogenesis of delirium.Previous studies have reported a steady relationship between opioid analgesics and perioperative delirium, with a few of them suggesting increased risk [2,4,16] and some proposing no relationship [17], while others suggested a decline in incidence of delirium [5].While it is important to control pain in perioperative procedures, it is also important to note that over-zealous drug administration may induce ED.Studies have shown that postoperative delirium is closely associated with preoperative states, degree of surgical injuries, and sedative/analgesic drug exposure [18][19][20][21].
Age has also been shown to be associated with ED [15].In this study, patients had median age of 56 years, with a few < 40 or > 65 years in age, an indication that age may be a major factor in the pathogenesis of ED.In addition, 94 % of patients fell under 2 or 3-ASA classification.There was no connection between BZD usage and delirium signs.These results are in agreement with those previously reported [16,22].While ED might be ancillary to pain, most of the patients that had it were CAM-ICU +ve, and were on admission in the PACU.There was Trop J Pharm February 2019; 18(2): 389 no significant difference in pain scores between patients that had of delirium and those that did not present the signs during admission to PACU.The patients had indications of brain disorders such as disorganized thinking or inattention, an indication that hyperactive features which occurred in a large number of the patients were not induced by pain.These results are in agreement with the reports of previous studies [23,24].

Limitations of the study
The likely limitations of this study include: (1) possibility of confounding factors in regression analysis; (2) small sample size; (3) inability to establish if patients had postoperative delirium or mild cognitive impairment before exposure to GA; (4) inability to assess pain score in patients during the emergence time, and (5) the use of a single race/population.

CONCLUSION
The results of this study suggest that in patients undergoing GA, signs of delirium are common in the immediate postoperative period, with the incidence highest on arrival at PACU and decreasing gradually during their stay in PACU.Hypoactive features are common and more prominent during patients' stay in PACU, compared to hyperactive features.

Table 2 :
Summary of outcomes