CJCCN Volume 34, Number 4, Winter 2023

Abstract

Introduction: Delirium is a frequent and important problem in the intensive care unit (ICU), and non-pharmacological means of prevention are limited. The importance of the physical environment in the occurrence of delirium in intensive care has been reported, particularly the presence of windows and daylight. We organized a trial to evaluate if the installation of virtual windows in the form of paintings in rooms without an actual window can limit the occurrence of delirium in ICU patients.

Methods: We conducted a retrospective pre and post cohort study in a surgical ICU of a university-affiliated hospital. Patients residing for more than 48 hours in a windowless room before and after the installation of virtual windows were included in the trial. The primary endpoint was the incidence of a positive screening test for delirium during their time in the ICU. The Intensive Care Delirium Screening Checklist (ICDSC) was used as an objective screening tool to assess the occurrence of delirium.

Results: A total of 400 patients were included in this trial (pre group: n = 200; post group: n = 200). The groups were well balanced except for the score APACHE II who was significantly higher in the post intervention group. The incidence of a positive screening test for delirium was similar in both groups after correction for confounding factors (29% vs 27%; OR 0,906 [0,584-1,402], p=0,656).

Conclusion: The installation of virtual windows did not reduce the incidence of delirium in a surgical intensive care unit.

Keywords

Critical care, delirium, windows, prevention, circadian cycle.

Implications for nursing: Delirium is a frequent problem in the ICU and non-pharmacological measures of prevention are important. The addition of virtual windows to windowless ICU rooms did not reduce the incidence of delirium, although a favorable trend emerged, especially in high-risk subgroups. Further research with higher number of patients and/or more realistic virtual windows is required in the future.

Background

Delirium is a complex and mostly reversible neuropsychiatric syndrome, involving multifactorial pathophysiology. It is manifested by a fluctuating disturbance of attention and consciousness, constituting a change in the functional level of the individual (APA, 2013). It can present as one of three types (hypoactive, hyperactive, or mixed). According to epidemiological studies, about 20% of hospitalized patients will develop delirium during a hospitalization (Bucht et al., 1999) although this condition is not diagnosed in a third of cases (Inouye et al., 1998). In intensive care, the incidence of delirium varies widely in the literature, reaching about 10-50% of patients (Salluh et al, 2010; Roberts et al., 2005; Thomason et al., 2005; Ely et al., 2004; Tomasi et al., 2012; Ouimet et al., 2007). It must be promptly recognized and treated, as this complication is associated with many clinical and economic implications, including increased hospital time, time spent on ventilator, mortality, and even some form of cognitive dysfunction that can persist after discharge from intensive care (Pisani et al., 2009; Shehabi et al., 2010; Lat et al., 2009; Girard et al., 2010b; Pandharipande et al., 2013).

As the effectiveness of treatments for delirium is limited, efforts are gradually turning to prevention methods. Although several pharmacological interventions have been attempted without much success (Girard et al., 2010a; van Ejik et al., 2010; Page et al., 2013), current preventive approaches rely mainly on non-pharmacological approaches. The importance of the physical environment in the occurrence of delirium in intensive care has been reported, particularly the presence of windows and daylight (van Rompaey et al., 2009). These findings remain controversial in medical literature, but they are still at the origin of the architectural guidelines for the construction of intensive care units recommending the installation of windows in all rooms (Thompson et al., 2012).

At the CIUSSS de l’Estrie-CHUS Hôpital Fleurimont, 8 of the 14 surgical intensive care unit beds do not have windows. Faced with the physical impossibility of creating real windows for these rooms, a grant was obtained to commission local artists to create a series of virtual windows in painting, representing natural landscapes (promoting stress reduction more than urban landscapes (Ulrich et al., 1991; Ulrich, 1981)) through window frames. The addition of an opaque shade that can be rolled down over the landscapes has made it possible to recreate a day-night alternation, also potentially beneficial in the prevention and treatment of delirium in intensive care [Fig.1].

To our knowledge, no study exists on the effect of virtual windows in intensive care units.  Previous studies observed the effect of real windows in intensive care on the occurrence of various events, including delirium. Wunsch et al. (2011) evaluated the effect of windows in a very specific population, i.e., patients presenting with subarachnoid haemorrhage in intensive care without specifically investigating the occurrence of delirium. No beneficial effects could be demonstrated. Kohn et al. (2013) studied several clinical outcomes of a large number of patients in intensive care, comparing patients residing in a room with or without a window. Mortality, length of hospitalization and delirium were among the clinical outcomes observed. Again, no statistically significant difference could be demonstrated. However, delirium was not identified using an objective tool such as the Intensive Care Delirium Screening Checklist (ICDSC) or the Confusion Assessment Method ICU (CAM-ICU), but rather by the presence of certain words in the notes of the clinical records suggesting the presence of delirium (e.g. confusion, agitation, disorientation, etc.). The probability of underestimating the true incidence of delirium, including hypoactive delirium, was therefore very high. Indeed, the incidence obtained in this study was only 3.6%. Since many intensive care units do not have real windows in all rooms, investigating the effect that virtual windows, an easy-to-obtain non-pharmacological therapy, can have on the occurrence of delirium in ICU patients is important. The aim of the present study was to compare the cumulative incidence of delirium among critically ill patients admitted to windowless rooms before and after the addition of virtual windows.

Methods

Study design and patient selection

We conducted a retrospective pre and post cohort study in the surgical ICU at the CIUSSS de l’Estrie – CHUS, Hôpital Fleurimont, a 14-bed surgical and trauma unit in a tertiary-care center in Sherbrooke, Quebec, Canada. Patients included in the study are all patients who were admitted to one of eight ICU beds without a true window for a period of at least 48 hours after June 2017 (the post cohort). June 2017 represents the date of installation of the virtual windows. This cohort was compared to a pre cohort of age- and APACHE II score-matched patients who were hospitalized in the same beds, for at least 48 hours, before the installation of the virtual windows (historical controls). Patients in both cohorts had to be older than 18 years at the time of admission and could not have already been included in the study, without any other exclusion criteria.

Primary outcome

The primary outcome of this study was to assess the effect of the installation of virtual windows on the incidence of positive screening for delirium, defined by an ICDSC  4 score obtained at least once during the stay in intensive care. The ICDSC is a validated tool designed for screening for delirium in intensive care. It allows an earlier recognition of this underdiagnosed clinical entity, since many patients have difficulty communicating adequately or are mechanically ventilated. In literature, this scale has a sensitivity of 99% and a specificity of 64% (Bergeron et al., 2001), in addition to being comparable to the other important screening tool, the CAM-ICU (Plaschke et al., 2008). This score takes into account eight factors: impaired state of consciousness, inattention, disorientation, hallucinations, agitation or psychomotor slowdown, inappropriate speech or mood, sleep-wake disorder and fluctuation of symptoms. Each criterion is worth one point and a score greater than or equal to 4 represents a positive test for delirium. In addition, the inter-rater reliability was 94% in the initial study, demonstrating that this screening method is valid and reproducible. In our community, the ICDSC must be completed by nurses on a regular basis, at least once per shift. The score is documented in the electronic medical record (EMR).

Figure 1 – Virtual windows with partially rolled shade

Several factors were also considered to ensure that the two groups were comparable in their innate risk of developing delirium during their stay in surgical intensive care. We extracted risk factors such as chronic alcohol consumption (> 14 drinks per week for women and > 21 drinks per week for men), active smoking (> 5 cigarettes per day) as well as history of stroke and neurocognitive disorders. We also documented severity of disease with the use of Acute Physiology and Chronic Health Evaluation (APACHE) II score for every patient.

Statistical analysis

We estimated that with 400 participants, we were able to detect a 10% difference in the incidence of delirium between the two groups with an 80% power at a 0.05 threshold for significance, and an estimated incidence of delirium of 25%. We believe that it is realistic to achieve with a 10% reduction in the incidence of delirium with a simple non-pharmacological intervention. Indeed, this estimate is supported by a study published by Rosa et al. (2017) which assessed the difference in the occurrence of delirium in a group of patients with restrictive visiting hours (< 4.5h/day) compared to another group with longer visiting hours (12h/day). They managed to demonstrate a significant decrease of 10.9% (9.6% vs 20.5%).

In order to compare the two groups, the analysis of dichotomous variables was done by the Chi-square or exact Fisher tests. The analysis of continuous variables was done using the Student t-test for normally distributed variables or by the Mann-Whitney U-test for non-normally distributed variables. A p score of less than 0.05 was considered statistically significant.

Regarding the primary objective, a chi-square test was used for the direct association between the group and the presence of delirium. A univariate and multivariate Poisson regression models were used to compare outcome occurrence.  Relative risk (RR) and adjusted RR (aRR) are presented with associated 95% confidence intervals.  SPSS version 28.0 (IBM Corp., Armonk, NY) and R version 4.2.1 (Foundation for Statistical Computing, Vienna, Austria) were used for statistical analysis.

Ethics

The study was evaluated and approved by the local Research Ethics Committee without the need for informed consent because of the retrospective nature of the study, as well as the risk-free intervention considered as standard of care in both groups.

Results

Patients

The installation of the virtual windows in the intensive care unit was completed in April 2017. The pre-installation cohort of virtual windows included 200 patients from May to December 2016. The post-installation cohort of virtual windows included 200 patients from June to December 2017. All patients without exception who resided for more than 48 hours in the 8 windowless rooms were included in the study.

The demographic, clinical characteristics and co-morbidities related to delirium occurrence of patients are represented in Table 1. The groups were well balanced. The mean age was 64.4 in the “pre” cohort and 63.5 in the “post” cohort. Men represented 66.8% of the included patients.  There were 22% of smokers, 8.3% who consumed alcohol significantly, 1.5% with a history of neurocognitive disorders, 7.8% with a history of stroke. The APACHE II score was significantly lower in the pre-well-balanced group with a median of 20 [13-26] compared to the post-group with a median of 22 [16-28] (p = 0.002).

Table 1

Baseline Patient Characteristics

Baseline characteristic	Pre group (n =200)	Post group (n = 200)	p
Age – years (mean ± standard deviation)	64.4 ± 14,2	63.5 ± 14,3	0.519
Male – n (%)	130 (65)	137 (68)	0.460
Chronic alcohol consumption – n (%)	20 (10,1)	13 (6,5)	0.198
Active smoking – n (%)	47 (23,6)	41 (20,5)	0.453
History of stroke – n (%)	16 (8,0)	15 (7,5)	0.852
Presence of major neurocognitive disorders – n (%)	2 (1)	4 (2)	0.685
Deaths – n (%)	12 (6)	14 (7)	0.685
Median APACHE II score – [IQR]	20.1[13-26]	22.0 [16-28]	0.002

Primary outcome

A total of 28% of patients enrolled in the study had a screening test considered positive for delirium. The cohort after implantation of the windows demonstrated a 7% relative risk reduction of developing delirium, although this change was not statistically significant (incidence of delirium 29% in the “pre” cohort versus 27% in the “post” cohort with a risk ratio (RR) of 0.93 ([95% CI].  0.68 to 1.28], p = 0.656). When adjusted for APACHE II score, the relative risk reduction remained not statistically significant at 13% (RR 0.87 ([95% CI].  0.63 to 1.19], p = 0.392) [Table 2]. Multivariate regression analysis including other confounding factors yielded similar results.

Subgroup analysis [Figure 2] revealed no group with statistically significant benefits from the addition of virtual windows, but higher-risk patients showed a trend towards a greater reduction in delirium.  For patients with age greater than 65, RR was 0.83 (95% CI 0.57-1.22, p = 0.35) and for patients with a history of stroke, RR was 0.46 (95% CI 0.14-1.45, p = 0.18).

Table 2

 Incidence of Testing Positive for Delirium

Results	Pre group (n = 200)	Post group (n = 200)	RR /aRR (IC 95%)	p
Positive screening test for delirium – n (%)	58 (29)	54 (27)	0.93 (0.68, 1.28)	0.656
Positive screening for delirium adjusted for APACHE II score			0.87 (0.63, 1.19)	0.396

Figure 2 – Sub-group analysis

Discussion

The present single center before and after study was unable to demonstrate a statistically significant difference in the incidence of delirium in intensive care when installing virtual windows. Our study demonstrates an interesting trend, especially in patients at higher risk for delirium, namely patients over 65 or with a past history of stroke. Patients with a pre-existing history of major neurocognitive disorders were underrepresented in our cohort, precluding any conclusions on the effects of our intervention on this population. Underrepresentation may be the effect of a selection bias, being a surgical ICU population with mostly high-risk elective surgery patients.

Our study assesses a very relevant clinical question, considering an increasingly older population with several risk factors for delirium. Easy to implement, cheap and risk-free non-pharmacological interventions for preventing delirium should be sought and studied.

One of the important limitations of our study is its retrospective design. This implies a risk of historical bias, as prevention or treatment methods may have changed between the pre- and post-intervention groups. Indeed, after April 2017, a weaning protocol for sedation and mechanical ventilation was introduced in the intensive care unit of the CIUSSS de l’Estrie CHUS. This intervention validated in the literature as a method of preventing delirium may therefore have interfered with the results obtained. However, given the low adherence to this protocol by medical teams throughout the duration of the analysis, the risk of interference with the results was considered minimal. In addition, it was not possible to correct for all factors that may influence the incidence of delirium. For example, total exposure to benzodiazepines, opioids or sedatives during intensive care hospitalization was not assessed, an important factor in the development of delirium. It is likely that this variable was similar across groups as practices in this regard varied little over the study period. Finally, generalization to a larger population may be more difficult since this is a single-center study, involving only surgical intensive care patients.

In conclusion, the installation of virtual windows has not been shown to be effective in reducing the incidence of delirium in surgical intensive care. Non-pharmacological interventions for the prevention of delirium always seem to be a promising opportunity to counter the health and economic impacts of delirium in intensive care. However, further studies will be needed to demonstrate their effectiveness against this major issue.

Authors’ Names and Affiliations

Olivier Gobeil MD, Faculté de médicine et des sciences de la santé (FMSS), Université de Sherbrooke,Québec,Canada

Véronique Savard RN, Centre Hospitalier Universitaire de Sherbrooke, Québec, Canada

Michael Mayette MD, Faculté de médicine et des sciences de la santé (FMSS), Université de Sherbrooke, Québec, Canada, Centre Hospitalier Universitaire de Sherbrooke, Québec, Canada, Centre de Recherche Clinique du Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Québec, Canada

Correspondence may be addressed to Michael Mayette MD

michael.mayette@usherbrooke.ca

Funding and Conflict of Interest

Funding for this study was provided by the Department of Medicine, Section of General Internal Medicine of the Faculté de Médecine et des Sciences de la Santé at the Université de Sherbrooke; funding for the creation of the paintings of windows was provided by the Fonds Brigitte-Perreault, a nonprofit organization dedicated to the humanization of care. The authors declare no potential conflict of interest, financial, commercial, or otherwise.

Acknowledgments

The authors would like to acknowledge the work of the three artists who created our “virtual windows”: Véronique Savard, René Brouillet and Anne-Marie Leclair. We also thank the Fonds Brigitte-Perreault for supporting this work. Finally, we thank Samuel Lemaire-Paquette for his work on statistical analysis.

References

American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders, Fifth Edition. Am Psychiatric Assoc, 21(21), 591-643.

Bergeron, N., Dubois, M. J., Dumont, M., Dial, S., & Skrobik, Y. (2001). Intensive Care Delirium Screening Checklist: evaluation of a new screening tool. Intensive Care Medicine, 27(5), 859-864.

Bucht, G., Gustafson, Y., & Sandberg, O. (1999). Epidemiology of delirium. Dementia and Geriatric Cognitive Disorders, 10(5), 315-318.

Ely, E. W., Shintani, A., Truman, B., Speroff, T., Gordon, S. M., Harrell Jr, F. E., Inouye, S. K., Bernard, G. R., & Dittus, R. S. (2004). Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit. JAMA, 291(14), 1753-1762.

Girard, T. D., Pandharipande, P. P., Carson, S. S., Schmidt, G. A., Wright, P. E.,

Canonico, A. E., Pun, B. T., Thompson, J. L., Shintani, A. K., Meltzer, H. Y., Bernard, G. R., Dittus, R. S., & Ely, E. W. (2010a). Feasibility, efficacy, and safety of antipsychotics for intensive care unit delirium: the MIND randomized, placebo-controlled trial. Critical Care Medicine, 38(2), 428.

Girard, T. D., Jackson, J. C., Pandharipande, P. P., Pun, B. T., Thompson, J. L.,

Shintani, A. K., Gordon, S. M., Canonico, A. E., Dittus, R. S., Bernard, G. R., & Ely, E. W. (2010b). Delirium as a predictor of long-term cognitive impairment in survivors of critical illness. Critical Care Medicine, 38(7), 1513.

Inouye, S. K. (1998). Delirium in hospitalized older patients: Recognition and risk factors. Journal of Geriatric Psychiatry and Neurology, 11(3), 118-125.

Kohn, R., Harhay, M. O., Cooney, E., Small, D. S., & Halpern, S. D. (2013). Do windows or natural views affect outcomes or costs among patients in ICUs?. Critical Care Medicine, 41(7), 1645-1655

Lat, I., McMillian, W., Taylor, S., Janzen, J. M., Papadopoulos, S., Korth, L.,  

Nold, J., Agarwal, S., Azocar, R., & Burke, P. (2009). The impact of

delirium on clinical outcomes in mechanically ventilated surgical and

trauma patients. Critical Care Medicine, 37(6), 1898-1905.

Ouimet, S., Kavanagh, B. P., Gottfried, S. B., & Skrobik, Y. (2007). Incidence,

risk factors and consequences of ICU delirium. Intensive Care Medicine,  33(1), 66-73.

Page, V. J., Ely, E. W., Gates, S., Zhao, X. B., Alce, T., Shintani, A., Jackson, J.,

Perkins, G. D., & McAuley, D. F. (2013). Effect of intravenous haloperidol on the duration of delirium and coma in critically ill patients (Hope-ICU): A randomised, double-blind, placebo-controlled trial. The Lancet Respiratory Medicine, 1(7), 515-523.

Pandharipande, P. P., Girard, T. D., Jackson, J. C., Morandi, A., Thompson, J.

L., Pun, B. T., Brummel, N. E., Hughes, C. G., Vasilevskis, E. E. Shintani,

A. K., Moons, K. G., Geevarghese, S. K., Canonico, A., Hopkins, R. O.,

Bernard, G. R., Dittus, R. S., & Ely, E. W. (2013). Long-term cognitive

impairment after critical illness. New England Journal of Medicine,

369(14), 1306-1316.

Pisani, M. A., Kong, S. Y. J., Kasl, S. V., Murphy, T. E., Araujo, K. L., & Van

Ness, P. H. (2009). Days of delirium are associated with 1-year mortality in

an older intensive care unit population. American Journal of Respiratory

and Critical Care Medicine,180(11), 1092-1097.

Plaschke, K., Von Haken, R., Scholz, M., Engelhardt, R., Brobeil, A., Martin, E.,

& Weigand, M. A. (2008). Comparison of the confusion assessment            method for the intensive care unit (CAM-ICU) with the Intensive Care

Delirium Screening Checklist (ICDSC) for delirium in critical care patients

gives high agreement rate(s). Intensive Care Medicine, 34(3), 431-436.

Roberts, B., Rickard, C. M., Rajbhandari, D., Turner, G., Clarke, J., Hill, D.,

Tauschke, C., Chaboyer, W., & Parsons, R. (2005). Multicentre study of

delirium in ICU patients using a simple screening tool. Australian Critical

Care, 18(1), 6-16.

Rosa, R. G., Tonietto, T. F., da Silva, D. B., Gutierres, F. A., Ascoli, A. M.,

Madeira, L. C., Rutzen, W., Falavigna, M., Robinson, C. C., Salluh, J. I.,

Cavalcanti, A. B., Azevedo, L. C., Cremonese, R. V., Haack, T. R.,

Eugênio, C. S., Dornelles, A., Bessel, M., Teles, J. M. M., Skrobik, Y., &

Teixeira, C. (2017). Effectiveness and safety of an extended ICU visitation

model for delirium prevention: a before and after study. Critical Care  

Medicine, 45(10), 1660-1667.

Salluh, J. I., Soares, M., Teles, J. M., Ceraso, D., Raimondi, N., Nava, V. S.,

Blasquez, P., Ugarte, S., Ibanez-Guzman, C., Centeno, J.V., & Laca, M.

(2010). Delirium epidemiology in critical care (DECCA): An international

study. Critical Care, 14(6), 1-7.

Shehabi, Y., Riker, R. R., Bokesch, P. M., Wisemandle, W., Shintani, A., & Ely, E. W. (2010). Delirium duration and mortality in lightly sedated, mechanically ventilated intensive care patients. Critical Care Medicine, 38(12), 2311-2318.

Thomason, J. W., Shintani, A., Peterson, J. F., Pun, B. T., Jackson, J. C., & Ely,

E. (2005). Intensive care unit delirium is an independent predictor of

longer hospital stay: a prospective analysis of 261 non-ventilated patients.

Critical Care, 9(4), 1-7.

Thompson, D. R., Hamilton, D. K., Cadenhead, C. D., Swoboda, S. M.,

Schwindel, S. M., Anderson, D. C., Schmitz, E. V., Andre, A. C., Axon, D.

C., Harrell, J. W., Harvey, M. A., Howard, A., Kaufman, D. C., & Petersen,

C. (2012). Guidelines for intensive care unit design. Critical Care

Medicine,40(5), 1586-1600.

Tomasi, C. D., Grandi, C., Salluh, J., Soares, M., Giombelli, V. R., Cascaes, S.,

Macedo, R. C., de Souza Constantino, L., Biff, D., Ritter, C., & Dal Pizzol,

F. (2012). Comparison of CAM-ICU and ICDSC for the detection of

delirium in critically ill patients focusing on relevant clinical outcomes.

Journal of Critical Care, 27(2), 212-217.

Van Eijk, M. M., Roes, K. C., Honing, M. L., Kuiper, M. A., Karakus, A., van der Jagt, M.,Spronk, P. E., van Gool, W. A., van der Mast, R. C., Kesecioglu, J., & Slooter, A. J. (2010). Effect of rivastigmine as an adjunct to usual care with haloperidol on duration of delirium and mortality in critically ill patients: A multicentre, double-blind, placebo-controlled randomised trial. The Lancet, 376(9755), 1829-1837.

Van Rompaey, B., Elseviers, M. M., Schuurmans, M. J., Shortridge-Baggett, L.

M., Truijen, S., & Bossaert, L. (2009). Risk factors for delirium in intensive

care patients: A prospective cohort study. Critical Care, 13(3), 1-12.

Ulrich, R. S., Simons, R. F., Losito, B. D., Fiorito, E., Miles, M. A., & Zelson, M.

(1991). Stress recovery during exposure to natural and urban

environments. Journal of Environmental Psychology, 11(3), 201-230.

Ulrich, R. S. (1981). Natural versus urban scenes: Some psychophysiological

effects. Environment and Behavior, 13(5), 523-556.

Wunsch, H., Gershengorn, H., Mayer, S. A., & Claassen, J. (2011). The effect of

window rooms on critically ill patients with subarachnoid hemorrhage

admitted to intensive care. Critical Care, 15(2), 1-10.