Chemokines are Associated with Delirium after Cardiac Surgery

Abstract

Background

Delirium has been hypothesized to be a central nervous system response to systemic inflammation during a state of blood brain barrier compromise. The purpose of this study was to compare postoperative changes in groups of inflammatory markers in subjects who developed delirium following cardiac surgery and matched controls without delirium.

Methods

Serum samples were drawn from forty-two subjects undergoing cardiac surgery preoperatively and postoperatively at six hours and postoperative day (POD) four. The serum concentrations of 28 inflammatory markers were determined with a microsphere flow cytometer. A priori, inflammatory markers were assigned to five classes of cytokines. A class Z-score was calculated by averaging the standardized, normalized levels of the markers in each class. Beginning on POD 2, subjects underwent a daily delirium assessment.

Results

Twelve subjects with delirium were matched by surgical duration, age, and baseline cognition to twelve subjects without delirium. At the 6-hour timepoint, subjects who went on to develop delirium had higher increases of chemokines compared to matched controls (class Z-score 0.3 ±1.0, p<.05). Among the 5 classes of cytokines, there were no other significant differences between patients with or without delirium at either the 6 hour or POD 4 assessments. Conclusion: After cardiac surgery, chemokine levels were elevated in subjects who developed delirium in the early postoperative period. Since chemokines are capable of distrupting blood-brain barrier integrity in vitro, future studies are needed to define the relationship of these inflammatory mediators to delirium pathogenesis.

Introduction

Delirium, a multifactorial geriatrics syndrome(1), is a common complication of cardiac surgery, occurring in 32–72% of older patients.(2) Postoperative delirium has been associated with increased mortality(3), postoperative complications(3), functional decline(4) and increased cost(5). Despite the high prevalence and negative sequelae of post-operative delirium, its pathophysiology remains unknown.(6, 7) Nevertheless, it has been proposed that systemic inflammation may actually contribute to delirium pathogenesis by compromising blood-brain barrier (BBB) integrity.(8–10)

The magnitude of the operative inflammatory response has been implicated as a risk factor of neurocognitive decline, including delirium, after surgery.(11–13) Normally, the BBB inhibits cytokines and many medications from passing across capillaries into the brain parenchyma.(14) Thus, the brain is relatively protected from the effects of systemic inflammation. Delirium is felt to be a central nervous system manifestation of a systemic disease state that may indeed cross the blood-brain barrier. In many of the situations in which delirium is likely to occur (infections, post-operative states, etc.), BBB integrity may be compromised. Chemokines are locally acting cytokines that enhance migration of inflammatory cells into the brain by compromising the BBB.(15, 16) Once BBB integrity is compromised, the brain becomes more susceptible to the effects of systemic inflammation.(10, 15)

The inflammatory response to surgery involves a complex coordination of cytokines, chemokines, and adaptive physiological responses required to maintain homeostasis. A transient post-operative increase in levels of circulating inflammatory markers (10–100x above baseline) has been hypothesized to result from tissue damage, adrenal stress response, cardiopulmonary bypass and/or anesthesia.(17–19) Inflammatory marker levels peak 6 to 24 hours postoperatively and return to baseline levels over 2 to 4 days.(18, 19) Inflammatory cytokines are produced by activated immune cells and are involved in the early amplification of the inflammatory response. Chemokines rapidly mediate leukocyte movement to sites of inflammation including across the blood brain barrier. Cytokines that promote a T-Helper 1 (TH-1) leukocyte and cytotoxic T-lymphocyte (CTL) response increase cell proliferation and CD8 expression. Cytokines that promote a T-helper 2 (TH-2) response are responsible for B-cell proliferation and humoral immune maturation. As a result of their respective roles, inflammatory cytokines and chemokines should be elevated early, with cytokines promoting TH-1/CTL and TH-2 responses increased during later stages of the post-operative inflammatory response.

The purpose of this study was to determine if a difference exists in the postoperative pattern of change in a priori determined classes of inflammatory markers in matched subjects with and without delirium after cardiac surgery. We hypothesized that as compared to their baseline levels, subjects with delirium would have: 1) increased inflammatory cytokines and chemokines 6 hours postoperatively and 2) increased cytokines that promote TH-1/CTL and TH-2 responses 4 days postoperatively.

Methods

Subject Enrollment

We prospectively enrolled 42 patients undergoing elective or urgent cardiac surgery at an academic medical center. Eligible procedures included coronary artery bypass graft (CABG), valve replacement, and combined CABG–valve surgery. Subjects with preoperative delirium, active substance abuse, psychiatric disease, and aortic procedures were excluded. Subjects provided their written informed consent and the study was approved by the institutional review board.

Anesthetic and Surgical Methods

Operative procedures were completed by three surgeons using the same conventional approach, including induction of general anesthesia, invasive monitoring, midline sternotomy and systemic heparinization. Mild hypothermic cardiopulmonary bypass (CPB) with cold-bloodhyperkalemic cardioplegia was used. All patients received antibiotics preoperatively and up to 48 hours postoperatively.

Measurement of Inflammation

Prior to surgery and six hours after surgery in the intensive care unit (ICU), blood samples were collected from the central venous line. Postoperative day 4 samples were collected peripherally. Blood samples were processed and serum samples were frozen at −80°C until the time of assay.

Samples were analyzed on a Luminex 100 dual-laser, microsphere flow cytometer (Luminex, Inc Austin, TX) using combined Biosource human cytokine 25-plex and a death receptor 3-plex bead kits (Invitrogen, Inc. Carlsbad, CA). Samples were incubated with the beads for 2 hours, washed, incubated with biotinylated detector antibodies for 1 hour, washed, incubated for 30 minutes with a conjugated fluorescent protein, and again washed. For each inflammatory marker measured, a standard curve was developed using four known concentration standards. The fluorescence of each inflammatory marker was converted to a concentration using the standard curve. In accordance with standard practice, samples with undetectable cytokine levels were entered at half of the minimum detection level derived from the standard curve.

A priori and based on the consensus of three experts in the study of inflammation, inflammatory markers were assigned to one of five classes: 1) inflammatory cytokines; 2) cytokines that promote TH-1/CTL responses; 3) cytokines that promote TH-2 responses; 4) chemokines, and 5) lymphatic chemokines. IL-17 and Death Receptor 5 were not assigned to any class and were analyzed independently.

Delirium

A brief delirium assessment (<15 min) was performed preoperatively and daily postoperatively, beginning on day 2. Subjects were not assessed on postoperative days 0 or 1 because of the intensive medical care required after CABG surgery. Delirium was assessed using the diagnostic algorithm of the Confusion Assessment Method (CAM)(20). Prior to its completion, a standardized mental status interview was conducted, including the Mini Mental State Exam (MMSE)(21), digit span, the Delirium Symptom Interview (DSI)(22), and the Memorial Delirium Assessment Scale (MDAS)(23). The MMSE is a screening assessment of mental status. The digit span asks patients to repeat a series of random digits forward and backward and is an assessment of working memory and attention. The DSI is an interview for eliciting 8 key symptoms of delirium. The MDAS is a severity scale for delirium. This combined assessment for delirium has been shown to be highly reliable (κ=0.95)(24)when administered by trained, non-clinician interviewers.

Matching

An analyst unaware of study aims and inflammatory marker results matched subjects with delirium to subjects who did not develop delirium on the basis of surgery duration (±90 minutes), age (±5 years) and baseline MMSE (±3 points), respectively. Because of the small and diverse sample, the matching process was appropriate to allow comparisons of baseline characteristics which might influence the inflammatory response. We used a Student’s t-test to compare the baseline characteristics of the matched controls, to those with delirium, and to the unmatched group.

Statistics

As the distribution of circulating inflammatory markers is generally non-normal, we log normalized the inflammatory marker concentrations. To calculate the postoperative inflammatory response, we subtracted the baseline log normalized concentration from the postoperative log normalized concentration (log[Postoperative]−log[Baseline]). The concentrations among the inflammatory markers were standardized to the mean and standard deviation of the matched non-delirious control group (Marker z-score). We created a class z-score by averaging the marker z-scores of the inflammatory markers within each class. We compared the mean Z-score among the assigned classes using a Student’s t-Test. All statistical calculations were performed using SPSS version 11.5.0 (SPSS, Inc. Chicago, IL).

Results

Among the 42 patients enrolled, twelve (29%) developed delirium. Table 1 describes the baseline characteristics of the matched control and delirium individuals, as well as the unmatched subjects. There were no significant differences between matched controls and subjects with delirium in the parameters used for matching (age, MMSE, and surgery time). Interestingly, no subject under the age of 60 years developed delirium. Matched controls were more likely to have a diagnosis of hypertension than the subjects with delirium, but all subjects were taking preoperative medications that lower blood pressure. The unmatched subjects were significantly younger (59.9 ± 5.5 years) and had significantly higher preoperative MMSE scores (28.6 ± 1.5) than the matched controls. There was no significant difference in preoperative comorbidity, medications useage, or surgical time among the groups.

Table 1.

Baseline characteristics of the matching procedure

	No Delirium n=12 mean (SD)	Delirium n=12 mean (SD)	Unmatched n=18 mean (SD)
Age (years)	73.9 (8.4)	74.7 (7.0)	59.9 (5.5)*
Male Gender n(%)	9 (75%)	11 (92%)	17 (94%)
Charlson Comobidity Index	2.1 (2.0)	2.9 (1.9)	1.7 (1.6)
Number of Medications	7.9 (2.4)	7.2 (2.7)	6.2 (2.9)*
Body Mass Index (kg/m2)	30.1 (6.1)	26.7 (4.1)	31.3 (5.7)
MMSE	25.3 (3.6)	26.1 (2.9)	28.6 (1.5)*
GDS	2.5 (2.2)	2.5 (1.3)	2.4 (2.1)
History of n(%):
Hypertension	10 (83%)	5 (42%)*	10 (62%)
Diabetes	5 (42%)	4 (33%)	7 (44%)
Hyperlipidemia	8 (67%)	5 (42%)	7 (44%)
WBC (per mL)	8.3 (2.8)	7.4 (2.3)	7.4 (1.6)
Hematocrit (%)	35.9 (5.3)	35.1 (4.4)	36.0 (5.4)
Creatinine (mg/dL)	1.9 (2.6)	1.2 (0.4)	1.1 (0.5)
Medications n(%)
Aspirin	12 (100%)	10 (83%)	16 (89%)
NSAID	0 (0%)	2 (17%)	2 (11%)
Steroid	0 (0%)	1 (8%)	1 (6%)
Beta-blocker	10 (83%)	7 (58%)	14 (78%)
ACE-I/ARB	7 (58%)	6 (50%)	12 (67%)
CCB	3 (25%)	4 (33%)	4 (22%)
Nitrate	3 (25%)	2 (17%)	2 (11%)
Diuretic	4 (33%)	7 (58%)	6 (33%)
Surgery Time (min)	210.3 (53.9)	202.8 (57.3)	221.1 (54.9)
CPB Time (min)	90.4 (38.6)	91.4 (50.4)	94.6 (36.3)

^*p<.05 compared to No Delirium group

Values are shown as mean ± (SD).

There were no differences between those with delirium (n=12) and the matched controls without delirium (n=12). The unmatched group (n=18) was significantly younger and had increased MMSE scores.

ACE-I/ARB angiotensin converting enzyme inhibitor/angiotensin receptor blocker, CCB calcium channel blocker, CPB cardiopulmonary bypass, GDS-Geriatric Depression Scale (15 point scale), MMSE Mini Mental State Exam, NSAID non-steroidal anti-inflammatory drug, SD standard deviation, WBC white blood cell count

Preoperative levels of inflammatory markers are compared in Table 2. Class Z-scores in subjects with delirium did not differ significantly compared to matched controls. Thus, the subjects were well matched for baseline inflammatory levels. Also, based on these data, it does not appear that greater levels of baseline inflammation are a risk factor for delirium.

Table 2.

Preoperative Inflammatory Markers

		Raw (pg/ul)		Normalized		Marker Z-Scores		Class Z-Scores
Class	Cytokine	No Delirium	Delirium	No Delirium	Delirium	No Delirium	Delirium	No Delirium	Delirium	p-value
Group 1: Inflammatory	IL-1β	678 (1019.5)	368.9 (1027.5)	4.6 (2.4)	3.4 (2.0)	0.0 (1.0)	−0.5 (0.9)	0.0 (1.0)	−0.1 (1.1)	0.45
	IL-1Rα	327 (518)	247.8 (495.3)	4.3 (2.0)	4.1 (1.8)	0.0 (1.0)	−0.1 (0.9)
	IL6	295.8 (422.7)	144.1 (179.3)	5.0 (1.2)	4.2 (1.4)	0.0 (1.0)	−0.7 (1.2)
	IFN-α	19.6 (29.1)	45.9 (74.2)	2.4 (1.0)	2.7 (1.4)	0.0 (1.0)	0.4 (1.5)
	TNF-α	18.3 (21.3)	30 (35.5)	2.4 (0.9)	2.7 (1.2)	0.0 (1.0)	0.3 (1.3)
	TNF-R1	1024.6 (906)	831.3 (895.4)	6.5 (1.0)	6.3 (1.0)	0.0 (1.0)	−0.2 (1.0)
	TNF-R2	7470.2 (3056.6)	7441.9 (2125.8)	8.8 (0.4)	8.9 (0.3)	0.0 (1.0)	0.1 (0.7)
Group 2: T-Helper 1	IL-2	532.3 (838.8)	272.4 (715.9)	4.0 (2.6)	3.1 (2.1)	0.0 (1.0)	−0.3 (0.8)	0.0 (1.0)	−0.1 (1.8)	0.79
	IL-2R	87.5 (91.1)	81 (68.2)	4.1 (0.9)	4.0 (0.9)	0.0 (1.0)	0.0 (1.0)
	IL7	65.1 (112.9)	52.2 (83.6)	3.0 (1.5)	2.9 (1.4)	0.0 (1.0)	−0.1 (1.0)
	IL12p40_p70	72 (58.5)	94.4 (91.9)	4.0 (0.8)	4.1 (1.0)	0.0 (1.0)	0.1 (1.3)
	IL15	164.8 (256.9)	179 (255.2)	4.2 (1.4)	4.1 (1.6)	0.0 (1.0)	−0.1 (1.1)
	IFN-γ	26.5 (45.6)	78.4 (141.2)	2.6 (1.0)	3.1 (1.6)	0.0 (1.0)	0.5 (1.6)
	IP-10	1011.4 (598.9)	1315.3 (1152.7)	6.8 (0.5)	6.5 (2.0)	0.0 (1.0)	−0.6 (3.8)
Group 3: T-Helper 2	IL4	25.9 (39.4)	54.9 (102.4)	2.6 (1.0)	2.8 (1.4)	0.0 (1.0)	0.2 (1.4)	0.0 (1.0)	0.1 (1.3)	0.59
	IL5	60.7 (76)	73.1 (127.8)	3.3 (1.4)	3.1 (1.5)	0.0 (1.0)	−0.1 (1.1)
	IL10	109 (107.2)	130.8 (147)	4.3 (1.0)	4.4 (1.0)	0.0 (1.0)	0.1 (1.0)
	IL13	25.7 (21.3)	49.9 (84.6)	3.0 (0.6)	3.2 (1.0)	0.0 (1.0)	0.3 (1.8)
Group 4: Chemokines	MIP-1α	121.5 (197.1)	113.4 (196.5)	3.6 (1.6)	3.4 (1.7)	0.0 (1.0)	−0.1 (1.0)	0.0 (1.0)	−0.2 (1.2)	0.28
	MIP-1β	53.3 (68.8)	94.8 (115)	3.4 (1.1)	3.6 (1.5)	0.0 (1.0)	0.2 (1.4)
	MIG	42 (48.1)	81.9 (114.8)	3.1 (1.2)	3.3 (1.8)	0.0 (1.0)	0.2 (1.4)
	EOTAXIN	593.4 (548)	421.2 (222.1)	6.1 (0.7)	5.9 (0.7)	0.0 (1.0)	−0.4 (1.0)
	RANTES	7241.2 (1521.1)	6445.7 (1427.6)	8.9 (0.2)	8.7 (0.2)	0.0 (1.0)	−0.6 (1.2)
	CCL-2	558 (419.7)	424.7 (257.2)	6.1 (0.6)	5.8 (0.8)	0.0 (1.0)	−0.5 (1.2)
Group 5: Lymphatic Chemokines	IL8	674.9 (1667.5)	146.7 (133.6)	4.7 (1.9)	4.2 (1.6)	0.0 (1.0)	−0.3 (0.9)	0.0 (1.0)	−0.2 (0.8)	0.52
	GM-CSF	60.4 (116.5)	35.8 (50.6)	2.9 (1.4)	2.8 (1.2)	0.0 (1.0)	−0.1 (0.8)
Miscellaneous	IL17	22.7 (29.7)	38.5 (51.3)	2.7 (0.9)	3 (1.1)	0.0 (1.0)	0.4 (1.3)
	DR5	87.7 (76.3)	88.7 (62.4)	3.9 (1.2)	4.1 (1.0)	0.0 (1.0)	0.2 (0.8)

Table 2 describes the process leading to inflammatory marker comparisons. Raw scores were log normalized because of data skew. Normalized scores were standardized to a mean of 0 and SD of 1 (Z-Scores) in the population without delirium (Marker Z-scores). Thus, the Marker Z-Scores for the ‘No Delirium’ group are 0 with a SD of 1. This step allows the comparison of the individual inflammatory markers and the effect of inflammatory class (Class Z-score). The Class Z-score is the average of the Marker Z-scores in the designated class. At baseline there were no significant Class Z-score differences between those with delirium and matched controls

Table 3 depicts the postoperative change in circulating inflammatory markers. At 6-hours postoperatively, circulating levels of chemokines was significantly increased (Marker Z-score >0) in participants who went on to develop delirium than matched controls without delirium (p<.05). However, the inflammatory cytokine group was not elevated. On the 4^th postoperative day, the levels of cytokines that promote TH-1/CTL and TH-2 responses tended to be lower (Class Z-scores <0) in subjects who developed delirium, yet this difference was not statistically significant.

Table 3.

Postoperative Change in Inflammatory Markers at 6 hours (ICU) and 4 days (POD 4)

		ICU					POD 4
		Marker Z-Scores		Class Z-scores			Marker Z-Scores		Class Z-scores
Class		No Delirium	Delirium	No Delirium	Delirium	p- value	No Delirium	Delirium	No Delirium	Delirium	p- value
Group 1: Inflammatory	IL-1β	0.0 (1.0)	−0.5 (0.8)	0.0 (1.0)	−0.1 (1.0)	0.57	0.0 (1.0)	0.1 (0.7)	0.0 (1.0)	0.0 (0.9)	0.75
	IL-1Rα	0.0 (1.0)	−0.1 (0.9)				0.0 (1.0)	−0.2 (0.8)
	IL6	0.0 (1.0)	0.5 (0.9)				0.0 (1.0)	0.2 (1.0)
	IFN-α	0.0 (1.0)	−0.4 (1.4)				0.0 (1.0)	−0.5 (1.0)
	TNF-α	0.0 (1.0)	−0.5 (1.0)				0.0 (1.0)	−0.5 (0.8)
	TNF-R1	0.0 (1.0)	0.3 (0.9)				0.0 (1.0)	0.2 (1.1)
	TNF-R2	0.0 (1.0)	0.1 (0.7)				0.0 (1.0)	0.4 (0.8)
Group 2: T-Helper 1	IL-2	0.0 (1.0)	−0.2 (0.6)	0.0 (1.0)	−0.2 (1.2)	0.34	0.0 (1.0)	0.0 (0.8)	0.0 (1.0)	−0.3 (1.2)	0.10
	IL-2R	0.0 (1.0)	−0.7 (0.8)				0.0 (1.0)	−0.2 (0.8)
	IL7	0.0 (1.0)	−0.2 (0.8)				0.0 (1.0)	−0.1 (0.8)
	IL12p40_p70	0.0 (1.0)	−0.3 (0.9)				0.0 (1.0)	−0.6 (0.9)
	IL15	0.0 (1.0)	−0.0 (1.0)				0.0 (1.0)	−0.3 (0.7)
	IFN-γ	0.0 (1.0)	−0.4 (1.6)				0.0 (1.0)	−0.4 (1.1)
	IP-10	0.0 (1.0)	0.8 (2.0)				0.0 (1.0)	−0.3 (2.4)
Group 3: T-Helper 2	IL4	0.0 (1.0)	−0.4 (1.6)	0.0 (1.0)	−0.1 (1.1)	0.80	0.0 (1.0)	−0.4 (1.1)	0.0 (1.0)	−0.3 (0.9)	0.08
	IL5	0.0 (1.0)	0.1 (0.9)				0.0 (1.0)	−0.1 (0.8)
	IL10	0.0 (1.0)	0.4 (0.8)				0.0 (1.0)	−0.3 (0.8)
	IL13	0.0 (1.0)	−0.3 (0.7)				0.0 (1.0)	−0.5 (0.9)
Group 4: Chemokines	MIP-1α	0.0 (1.0)	−0.1 (1.0)	0.0 (1.0)	0.3 (1.0)*	0.04	0.0 (1.0)	−0.3 (0.9)	0.0 (1.0)	0.0 (1.1)	0.68
	MIP-1β	0.0 (1.0)	−0.1 (0.9)				0.0 (1.0)	−0.5 (1.0)
	MIG	0.0 (1.0)	0.5 (1.5)				0.0 (1.0)	−0.4 (1.1)
	EOTAXIN	0.0 (1.0)	0.7 (0.6)*				0.0 (1.0)	0.8 (0.8)*
	RANTES	0.0 (1.0)	0.3 (0.6)				0.0 (1.0)	0.0 (1.2)
	CCL-2	0.0 (1.0)	0.8 (0.8)*				0.0 (1.0)	0.5 (0.9)
Group 5: Lymphatic Chemokines	IL8	0.0 (1.0)	0.4 (0.8)	0.0 (1.0)	0.2 (0.8)	0.44	0.0 (1.0)	0.1 (1.3)	0.0 (1.0)	0.1 (1.1)	0.93
	GM-CSF	0.0 (1.0)	0.0(0.8)				0.0 (1.0)	0.1 (0.9)
Miscellaneous	IL17	0.0 (1.0)	−0.6 (1.1)				0.0 (1.0)	−0.5 (0.8)
	DR5	0.0 (1.0)	−0.2 (0.7)				0.0 (1.0)	−0.4 (0.8)

^*p<.05;

The postoperative inflammatory marker concentration (ICU and POD4) was divided by the preoperative concentration (fold-change) and log normalized. To obtain the ‘Change Z-Score’, the population was standardized with a mean of 0 and SD of 1 relative to the control group without delirium. We present the Change Z-score for each inflammatory marker and ‘Class Change Z-Score’ which represents the mean of the inflammatory marker class. Overall, circulating cytokine increased after surgery at both times. Among participants with delirium, there was a greater increase in chemokine class levels in the ICU and a trend toward decreased T-Helper 1 and T-Helper 2 class responses at POD4.

Discussion

Delirium after surgery has been hypothesized to occur as a result of the inflammatory response.(9) However, the basic pathophysiology is not understood. This study utilized new technology in order to examine peripheral inflammatory marker responses to cardiac surgery in patients with delirium and in matched controls without delirium. Based on the temporal nature of typical inflammatory responses, we hypothesized that inflammatory cytokines and chemokines would be initially elevated because of their role in the control of subsequent inflammatory responses and that cytokines known to promote TH-1/CTL and TH-2 responses would be elevated on postoperative day 4. At both times, we postulated that delirium would be associated with a greater increase compared to those without delirium. Unexpectedly, our findings indicate a much more restricted inflammatory response. In the immediate postoperative periods, patients who later developed delirium had increased chemokine levels, but levels of classical inflammatory cytokines were similar between those who developed delirium and matched controls. At 4 days after surgery, patients with delirium tended to demonstrate smaller increases of cytokines that promote the TH-1/CTL and TH-2 response.

Chemokines have not been traditionally included in delirium research, but recent studies have added biological plausibility to a hypothesis implicating these inflammatory mediators in the earliest events contributing to delirium. While our study was not designed to detect changes in levels of individual chemokines, elevations in inflammatory cytokines (particularly IL-6) have been linked to postoperative delirium in other studies.(25, 26) Chemokines are known to promote leukocyte migration into the CNS(27) and in the case of Chemokine C-C motif ligand 2 (CCL2) mediate blood brain barrier disruption in the context of ischemic injury.(15) Moreover, at least three important risk factors for delirium: aging(28, 29), brain injury(29) and anticholinergic medications(30) have been associated with greater basal and induced CCL-2 activity. Our observed elevations in systemic chemokine levels are quite transient, with no detectable differences between individuals with and without delirium by the 4^th postoperative day.

The cause of the non-significant trend toward lower TH-1/CTL and TH-2 responses in individuals with delirium at postoperative day 4 is unclear. We propose three hypotheses for this finding: a) a lack of reserve capacity to mount an inflammatory response; b) most of these cytokine elevations are very modest (exceptions: IL-4, IL-10) and these substances exerted predominantly local autocrine effects and were being consumed locally(9); and c) stress mediators, such as glucocorticoids, are likely to be elevated in the setting of delirium(31) and tend to suppress T-Helper 1 immunity, while promoting T-Helper 2 immunity.(32) Thus, as we learn more about the nature of the relationships among different cytokines, we can better understand the cognitive effects of cytokines.

The inflammatory response is a highly complex and dynamic inter-related process where any one cytokine can modulate a pro-inflammatory and/or anti-inflammatory response depending on multiple clinical, physiologic and immune considerations. Traditionally, studies of inflammation and geriatric syndromes measured individual inflammatory markers using enzyme linked immunosorbent assays (ELISA) to measure a single protein in a blood sample.(33, 34) While reliable, ELISAs are time, labor and sample consuming, resulting in increased cost and inefficient use of available blood sample. The dual-channel microsphere flow cytometer is a recent advance that allows assessment of multiple cytokines simultaneously using a blood sample (100 microliters) comparable to that required for an individual ELISA. As with all new technologies, some cautionary notes have been raised. First, a reliable uniplex assay cannot be merely added into a reliable multiplex array without additional validation(35). Studies have demonstrated good correlations, but often poor concurrence of quantitative values between multiplex kits made by different manufacturers(36), as well as between multiplex kits and ELISA measurements(35). Because inflammatory markers are likely to be dependent on one another, grouping markers into classes for analysis increases statistical power and reduces the likelihood of finding a spurious association as a result of multiple testing. Nevertheless, carefully performed multiplex assays offer investigators opportunities for evaluating varied elements of inflammatory responses in the same individual over time.

Our study’s approach to examine peripheral cytokine panels after cardiac surgery has both strengths and limitations. Its major advantage is the utilization of the microsphere flow cytometer which allowed the measurement of 28 cytokines simultaneously. Thus, we were able to acquire a broad picture of changes in many cytokines and group them according to their roles within the overall inflammation response. However, the number of inflammatory markers measured increases the number of subjects needed to definitively draw conclusions about the relationship between delirium and inflammation. In this study, we focused on class effects because the large number of inflammatory markers limited the power to make definitive statements regarding any individual cytokine. While our study would clearly benefit from more subjects with and without delirium, this type of analysis is commonly used in the gene microarray literature, where the expression of thousands of different genes may be measured in a single subject to identify specific areas of interest for future research.

Our study is limited by the number of subjects. The matched analysis provided a baseline level of risk adjustment. Inflammation and delirium can be affected by cognitive performance, age, and the surgical procedure. By matching for these factors, we were able to control some of the baseline risk. In future studies with more patients, it would be helpful to match by gender, comorbidity, and atherosclerosis burden which can affect inflammation.(9)

Since delirium is a disorder of the central nervous system, our assumption that serum levels of inflammatory markers parallel those in the CNS inflammatory response cannot be definitively proven at this time. However, we did find associations with the inflammatory response and delirium, suggesting that they may indeed be related. Moreover, since delirium may be associated with breakdown of the blood-brain barrier, serum levels may indeed be correlated with CNS levels, although this requires further evaluation.

In conclusion, using new technology to examine a large panel of cytokines, we found that early postoperative increased elevations of chemokines were associated with the development of delirium after cardiac surgery. While more research is needed to definitively establish a causative link, our evidence suggests that delirium may be at least in part mediated via chemokines, as these potent immune mediators attract inflammatory cells to the site of brain injury, may cause breakdown of the blood brain barrier, and may ultimately suppress T-cell-mediated immunity in patients who develop delirium in the post-operative period. These results represent a potential pathophysiologic mechanism of delirium after surgery. Moreover, once replicated in a larger sample that will allow analysis of individual chemokines or cytokines, our findings may also offer a first potential target for mechanism-driven interventions for this common and morbid syndrome.