Anesthesia

The possibility that the anesthetic contributed to PND was first posed in a study using elderly rats.^[1] This was followed by a host of animal studies, almost all of which were in healthy adult wild-type rats or mice, exposed to a general anesthetic for several hours, usually breathing spontaneously, allowed to emerge, and then subjected to various behavioral testing at intervals after the anesthetic, typically up to a week and rarely more than a month. With the introduction of vulnerable animals and surgery to the studies, it has become clear that the anesthetic may have modulatory effects on outcomes but that the vulnerability of the animal prior to surgery and the inflammatory trauma as a result of surgery play far more significant roles in determining outcomes.

Class of Anesthetic/Route of Administration

The relative ease of delivering anesthesia by inhalation makes these drugs the most commonly used clinically, and also the most commonly used in animal studies. The potency of all the currently used gases (isoflurane, desflurane, and sevoflurane) is very similar in all mammals, making studies in the rodent of reasonable relevance. The use of injectable anesthetics in rodents, on the other hand, is more problematic. In humans, injectable drugs are almost always given intravenously. While this can also be performed in rodents using the tail vein approach, many investigators instead use the intraperitoneal (IP) approach. A commonly used route of administration for a wide variety of medications in small animals, the IP approach is never used in humans. While potentially relevant to clinical practice, IP administration has different pharmacokinetics and possibly different inflammatory and immunological implications. Oral administration of drugs is a common route for rodents, either by gavage or adding to drinking water, but a rarely used route for general anesthetics.

Physiological Control and Monitoring

General anesthetics have many side effects, some of which are potentially lethal, and they occur at drug concentrations very close to those required for the primary effect, anesthesia. These side effects include cardiovascular and respiratory depression and thermal dysregulation, all of which could independently injure (or protect) the brain and contribute to any subsequent cognitive disorders. In clinical anesthesia, these side effects are carefully monitored and mitigated where necessary. This is almost never true in rodent studies, as it is very difficult to control the ventilation or monitor blood gases in animals weighing only 30 grams. Obtaining blood for the measurement of blood gases or chemistry, for example, is often a terminal procedure because of the volume required. Small, rapid-response pulse oximeters that can provide both heart rate and oxygen saturation are becoming available, but these cannot provide a measure of CO₂ or pH, both measures of the adequacy of ventilation. Blood pressure and cardiac output are also difficult to monitor. Unexpected mortality, a good indicator of significant side effects, is rarely reported. Thus, rodent anesthesia is often poorly monitored and therefore significantly contaminated with physiological derangements that can contribute to, or cause, cognitive dysfunction and confound data interpretation. Also the percent oxygen used with an inhaled anesthetic needs to be determined for aged rodents.

Scaling

An issue of considerable importance, but also of ambiguity, is anesthetic exposure scaling. For example, is a 6 hour anesthetic exposure in an animal whose entire life-span is 2 years equivalent to a 250 hour anesthetic in a human that lives 80 years? Is rodent cognitive performance at 1 week postexposure analogous to that in a human at 1 year? At the most fundamental level, atoms, molecules, and cells move and respond with the same kinetics in both human and rodent, but some organs function at much greater speed (e.g., the heart). On the other hand, a rodent’s circadian rhythm is analogous to that of a human and, as mentioned above, anesthetic potency is almost identical. Therefore, it is simply not clear whether adjustments to exposure lengths or recovery periods need to be made to account for life-span differences. Currently, most investigators do not.

Surgery

Anesthesia is rarely given to patients in the absence of a procedure that would normally cause pain and suffering. Such a procedure could itself cause cognitive disturbance, or may alter the effect of the anesthetic. Thus, investigators have begun to include a surgical procedure in their models to more closely simulate what happens to patients. Importantly, it has become clear in recent years that the surgical injury produces activation of the animal’s innate immune system, inducing up-regulation of many inflammatory mediators in the periphery and in the brain in animal models. Moreover, a number of endogenous molecules, called alarmins, are released from damaged cells during injuries such as fractures and many of these have consequences for cognition.^[2–7] The surgical procedures used in modeling this have varied, and each has their strengths and limitations, especially when applied to healthy adult animals. The remainder of this chapter will refer to studies that included surgery or other systemic inflammatory insults primarily in aged animals.

Orthopedic. Because orthopedic surgery is exceedingly common in the geriatric populations, some investigators have developed orthopedic models though thus far they have not been used in aged animals. Most commonly, the tibia is fractured under anesthesia in a mouse, and then repaired with an intramedullary rod, simulating a common hip fracture. The advantages are the consistency and relevance, but the disadvantage is that pain and tissue damage may limit the animal’s mobility. Cognition and motor ability can be difficult to separate in conventional rodent maze tasks only a few days after surgery, while other cognitive tests, such as fear conditioning, may not be similarly affected.

Abdominal. Abdominal procedures are also common in the elderly population, including colectomy, herniorrhaphy, small bowel resection, prostatectomy, etc. In the rodent, several aseptic abdominal procedures have been reported in aged wild-type rodents and Alzheimer transgenic models, ranging from a simple laparotomy^{[2, 4, 8]} to partial hepatectomy,^{[9, 10]} splenectomy,^{[9, 11]} nephrectomy^[12] to cecal ligation and excision (without puncture).^{[13, 14]} As in the orthopedic procedures, postoperative pain may limit movement-dependent behavioral assays, especially if tested soon after the surgery. Analgesics, however, introduce new confounders that have to date not been considered to influence outcome.

Models of Systemic Inflammation (Sepsis and LPS)

Surgery constitutes a significant inflammatory trauma and shares many features with the systemic inflammatory responses observed during systemic infection and sepsis. Indeed a number of endogenous molecules released during tissue injury (alarmins such as high mobility group box 1, HMGB1) activate the same immune system receptors as those activated by gram-negative bacterial endotoxin (lipopolysaccharide, LPS) and lead to release of a similar repertoire of inflammatory mediators. As such, studies of systemic inflammation induced by LPS or sepsis can and have^[15] provided information on cognitive impairments relevant to PND.

Moreover, not all surgery is clean, and sepsis syndrome among the elderly receiving operative care is not uncommon. Sepsis is also associated with significant morbidity and mortality, as well as subsequent cognitive problems in humans, especially the elderly,^{[16, 17]} and in rodent sepsis survivors.^[18–20] One model of sepsis in the rodent consists of ligating and puncturing the cecum (CLP), to lead to abscess formation and fecal contamination of the abdomen. This “CLP” surgery leads to a pronounced sepsis syndrome, the survival from which depends on a number of operative (number of punctures) and postoperative (fluids, antibiotics, warming) manipulations. Mortality can be very high, and thus this is rightly considered a model of septic shock rather than a model of surgery. Although an extreme one, it is not an irrelevant model of surgery but the microbial contribution to innate immune responses considerably outweighs the surgical/traumatic contribution. A limitation of this surgery can be the generation of a “survivor-effect,” whereby the animals/humans that do not die, either did not receive the same insult or become resistant in some way to the insult and its consequences, and go on to do as well or better than the controls.^{[21, 22]}

In order to simplify the sepsis syndrome to the upstream initiators, some investigators administer purified LPS, either IV or IP, to induce a transient and robust systemic inflammation or “sickness behavior” with neuroinflammation and changes in cognition.^[23–26] This procedure is done in the absence of anesthesia, which may be able to help tease out the effects of the anesthetic. Whether this sickness behavior constitutes delirium or PND is not clear, although it does not always produce cognitive impairment.^[27] However, these challenges do activate innate immunity and neuroinflammation and overlap significantly with inflammatory activation occurring during aseptic surgery.

Animal Models of Preexisting Vulnerabilities

A general theme to this chapter on animals, as well as the design of clinical studies, is that the enrichment of the study population for vulnerabilities (age being one) enhances the probability of detecting relevant effects so that mitigation strategies can be tested (Table 6.1). Not at all patients and animals suffer cognitive disorders following anesthesia and surgery; there are clearly preexisting vulnerabilities that interact with the stress of anesthesia and surgery. This interaction could be very specific, so that knowing the vulnerability may provide clues as to the pathophysiology by which anesthesia and surgery produce acute cognitive dysfunction and possibly even accelerate decline. Age is a key vulnerability factor, as we mentioned above, but this may encompass multiple distinct risk factors. One key feature in the aged human population is the occurrence of cognitive impairment at baseline (i.e., preoperatively) and surgery in aged rodents can produce deficits that are exaggerated as compared to deficits produced in young healthy animals.

Table 6.1 Animal Models of Anesthesia and Surgery with Behavioral Testing for PND

Animal Model	Preexisting Vulnerability	Event		Intervention	Post-op Cognitive Test/Time	Results of Cognitive Testing	Reference
		Anesthesia	Surgery
Mice BALB/c	Age 23–25 mos	Ketamine and xylazine, i.p.	Laparotomy	–	MWM 24 hours	Not impaired: Surgery	[2]
Rats Sprague-Dawley	Age 20–24 mos	Isoflurane	Partial hepatectomy	–	MWM 1, 3, 7 days	Impaired: Surgery	[42]
Mice C57BL/6	Age 14 mos	Chloral hydrate, i.p.	Partial splenectomy	Celastrol, i.p.	MWM 1, 3, 7 days	Impaired: Surgery Not impaired: Surgery + Celastrol	[9]
Mice C57BL/6	Age 16 mos	Isoflurane	Partial hepatectomy	–	MWM 31–35 days	Improved cognition: Surgery	[43]
Rats F344/BN	Age 24 mos	Halothane	Laparotomy	hIL-1RA (intracisternal /i.p.)	FC days 4, 12	Impaired: Surgery + hIL-1RA (i.p. injection) Not impaired: Surgery + hIL-1RA (intracisternal injection)	[4]
Rats Sprague-Dawley	Age 22–23 mos	Isoflurane	Partial Splenectomy	–	MWM 1, 3, 7 days	Impaired: Surgery	[44]
Rats LCR, HCR	Metabolic syndrome, Age 4–5 mos	Isoflurane	Tibial fracture with fixation	–	FC 1–7 days MWM 5 mos	Impaired: Surgery + LCR > Surgery + HCR	[45]
Rats Wistar	Age 18–20 mos	Sevoflurane	Laparotomy + upper mesenteric artery clamped	–	MWM, Y-Maze, OF, EPM, FC days 7, 35	Impaired: Surgery (affective behavior only) Not impaired: Surgery (cognition)	[29]
Rats Sprague-Dawley	Age 20 mos	Isoflurane	Partial hepatectomy	Ulinastatin, i.v.	MWM 3, 7, 15 days	Impaired: Surgery Not impaired: Surgery + Ulinastatin	[46]
Mice C57BL/6	Age 14 mos	Isoflurane	Partial hepatectomy	–	Y-maze 1, 3, 7	Impaired: Surgery	[47]
Rats LCR, HCR	Metabolic syndrome, Age 4 mos	Isoflurane	Tibial fracture with fixation	–	FC 3 days	Impaired: Surgery + LCR > Surgery + HCR	[48]
Mice 3xTgAD (App/tau/PS1)	AD transgenic Age 5–11 mos	Desflurane	Abdominal (cecal ligation & excision)	–	MWM 2 weeks; 14 weeks	Impaired: Surgery	[13]
Rats Sprague-Dawley	Age 22 mos	Fentanyl + droperidol, i.p.	Splenectomy	–	Y-maze 1, 3, 9 days	Impaired: Surgery	[11]
Rats Fischer 344	Age 20 mos	Propofol Isoflurane	Right carotid exposure	–	BM FC 6–18 days	Impaired: Surgery + Propofol; Surgery + Isoflurane	[37]
Rats Sprague-Dawley	Age 18–22 mos	Chloral hydrate	Splenectomy	–	MWM 1, 3, 5, 7 days	Impaired: Surgery	[49]
Mice C57BL/6	Age 19 mos	Sodium pentobarbital, i.p.	Partial hepatectomy	Minocycline, i.p.	MWM 31–35 days	Impaired: Surgery Not impaired: Surgery + Minocycline	[50]
Rats Wistar	Age 24–25 mos	Isoflurane	Laparotomy (incision only)	Ketoprofen, SQ	RAM 3–14 days	Impaired: Surgery Not impaired: Surgery + Ketoprofen, Surgery + Morphine	[51]
Rats Sprague-Dawley	Age 18 mos	Sevoflurane	Partial hepatectomy	–	MWM 1, 3, 7 days	Impaired: Surgery	[10]
Rats Sprague-Dawley	Age 20 mos	Isoflurane	Laparotomy	Candesartan	MWM 2–7 days	Impaired: Surgery Not impaired: Surgery + Candesartan	[52]
Mice ADTg (APP/PS1) C57BL/6J	AD transgenic Age 9 mos, ADTg 18 mos, WT	Bupivacaine	Laparotomy (incision only)	–	FC 24h, 72h, 7, 30, 60 days MWM 1–7 days	Impaired: Surgery ADTg; Surgery WT	[8]
Mice C57BL/6	Age 12 mos	Ketamine and xylazine, i.p.	Nephrectomy	RNase A, SQ	MWM 1, 3, 7 days	Impaired: Surgery Not impaired: Surgery; Surgery + RNase A	[12]
Rats LCR, HCR	Metabolic syndrome, Age 4 mos	Isoflurane	Tibial fracture with fixation	–	FC 1week MWM 3 mos	Impaired: Surgery (LCR + HCR) 1 week Surgery LCR 3 mos Not impaired: HCR 3 mos; Anesthesia only	[54]
Rats Wistar	Age 25 mos	Sevoflurane	Laparotomy + upper mesenteric artery clamped	–	MWM, NOR OF 9–14 days	Impaired: Surgery	[34]
Rats Wistar	Age 18 mos, Prior infection	Sevoflurane	Laparotomy + upper mesenteric artery clamped	–	OF NOR MWM 9–14 days	Impaired: Surgery + prior infection	[54]
Mice C57BL/6	Age 16 mos	Isoflurane Pentobarbital sodium (ICV implants)	Laparotomy ICV cannula implantation	SAHA, ICV injection	OF FC STM LTM 5–7 days	Impaired: Surgery Not impaired: Surgery + SAHA	[55]
Rats Wistar	Age 24–25 mos	Isoflurane	Laparotomy	Environmental enrichment (PPE)	NOR 7 days	Impaired: Surgery Not impaired: Surgery + PPE	[56]
Rats Sprague-Dawley	Age 18–20 mos	Isoflurane	Partial hepatectomy	–	Y-maze 1–2 days	Impaired: Surgery	[57]
Rats Sprague-Dawley	Age 20 mos	Sevoflurane	Partial hepatectomy	–	MWM 8–12 days	Impaired: Surgery	[58]
Rats Sprague-Dawley	Age 16 mos	Isoflurane	Splenectomy	Lithium, i.p.	MWM 1 day	Impaired: Surgery Not impaired: Surgery + Lithium	[33]
Rats Sprague-Dawley	Age 24 mos	Pelltobarbitalum Natricum	Laparotomy	R‑7050, intracisternal injection	MWM 1, 3, 5 days	Impaired: Surgery Not impaired: Surgery + R‑7050	[59]
Mice BALB/c	Age 20–22 mos	Isoflurane	Splenectomy	Dexmedetomidine, i.p.	Y-maze 1, 3 days	Impaired: Surgery Not impaired: Surgery + Dexmedetomidine	[38]
Mice C57BL/6	Age 16 mos	Isoflurane	Laparotomy	Apocynin	OF FC 6–7 days	Impaired: Surgery Not impaired: Surgery + Apocynin	[60]
Mice C57BL/6	Age 14 mos	Sodium pentobarbital, i.p.	Partial hepatectomy ICV cannula implantation	IL17A antibody, ICV injection	MWM 1, 3, 7 days T-maze 1, 3, 7 days	Impaired: Surgery Not impaired: Surgery + IL17A antibody	[61]
Mice C57BL/6	Age 20 mos	Chloral hydrate, i.p.	Laparotomy	Mifepristone Roscovitine	MWM 5–13 days EPM 5 days	Impaired: Surgery Not impaired: Surgery + Roscovitine; Surgery + Mifepristone	[62]
Mice C57BL/6	Age 18 mos	Isoflurane	Appendectomy	–	MWM 3, 28 days	Impaired: Isoflurane; Surgery + Isoflurane	[63]
Rats Wistar	Age 20–24 mos	Isoflurane	Laparotomy	Insulin, intranasal + pentobarbital anesthesia	NOR PO day 14	Impaired: Surgery Not impaired: Surgery + intranasal insulin	[64]
Rats Wistar	Age 24–25 mos	Isoflurane	Laparotomy	Pregabalin, i.p. (pre-op vs. PO)	NOR OF 14 days	Impaired: Surgery; Surgery + Pregabalin (PO) Not impaired: Surgery + Pregabalin (pre-op)	[65]
Mice C57BL/6	Age 12–14 mos	Chloral hydrate, i.p.	Laparotomy	Deferoxamine, i.p.	FC 3, 7, 14 days MWM 3, 7, 14 days	Impaired: Surgery Not impaired: Surgery + Deferoxamine	[66]
Rats Sprague-Dawley	Age 18 mos	Chloral hydrate, i.p.	Laparotomy	Deferoxamine	MWM 1, 3, 7 days	Impaired: Surgery Not impaired: Surgery + Deferoxamine	[67]
Rats Sprague-Dawley	Age 19–22 mos	Sevoflurane	Partial hepatectomy	anti-HMGB1	BM 1–4 days	Impaired: Surgery Not impaired: Surgery + anti-HMGB1	[68]
Rats Wistar	Age 18 mos	Pentobarbital sodium, i.p.	Laparotomy + splenectomy	Dexmedetomidine	MWM 1, 3, 7 days	Impaired: Surgery Not impaired: Surgery + Dexmedetomidine	[69]
Mice C57BL/6	Age 20 mos	Isoflurane	Laparotomy	Berberine	OF FC 6–7 days	Impaired: Surgery Not impaired: Surgery + Berberine	[70]
Rats LCR, HCR	Metabolic syndrome, Age 4 mos	Isoflurane	Tibial fracture with fixation	Preoperative exercise (PE)	FC 1week MWM 3 mos	Impaired: Surgery + PE (LCR + HCR) 1 week; Surgery LCR 3 mos Not impaired: Surgery + PE (LCR + HCR) 3 mos	[71]
Rats Sprague-Dawley	Age 21–23 mos	Isoflurane	Partial hepatectomy	Electroacupuncture Minocycline, i.p.	MWM 3, 7 days	Impaired: Surgery Not impaired: Surgery + Electroacupuncture; Surgery + Minocycline	[72]
Rats Sprague-Dawley	Age 18–20 mos	Isoflurane	Partial hepatectomy	IL-4, intracisternal injection	MWM 1, 3, 7 days	Impaired: Surgery Not impaired: Surgery + IL-4	[73]
Rats Sprague-Dawley	Age 18 mos	Chloral hydrate, i.p.	Partial hepatectomy	Electroacupuncture	MWM 1, 3, 7 days	Impaired: Surgery Not impaired: Surgery + Electroacupuncture	[74]
Mice 3xTgAD (App/tau/PS1)	AD transgenic Age 11 mos	Propofol	Abdominal (cecal ligation & excision)	–	MWM 3 weeks; 15 weeks	Not impaired: Anesthesia + surgery	[14]
Mice FAD5X	AD transgenic Age 5 mos	Isoflurane Desflurane	Laparotomy	Vitamin K2	BM 3, 7, 14, 21, 28, 35, 42 days	Impaired: Surgery + Isoflurane (day 14) Not impaired: Surgery + Desflurane; Surgery + Isoflurane (all days except 14)	[75]
Rats Sprague-Dawley	Age 18 mos	Isoflurane	Splenectomy	Nimodipine, Hypertonic Saline	MWM 1, 3, 7 days	Impaired: Surgery Not impaired: Surgery + Nimodipine + Hypertonic Saline	[76]
Mice C57BL/6J	Age 14 mos	Bupivacaine, SQ	Laparotomy (incision only)	Edaravone	MWM 1–7 days T-maze 1, 3, 7 days	Impaired: Surgery Not impaired: Surgery + Edaravone	[77]
Mice Kunming	Age 16 mos	Chloral hydrate, i.p.	Partial hepatectomy	Hydrogen-rich saline	MWM 8, 10, 14 days	Impaired: Surgery Not impaired: Surgery + Hydrogen-rich saline	[78]
Mice C57BL/6	Age 14 mos	Chloral hydrate, i.p.	Carotid artery exposure	Rolipram	MWM FC 3–12 days	Impaired: Surgery Not impaired: Surgery + Rolipram	[79]
Rats Sprague-Dawley	Age 18 mos	Fentanyl + Droperidol, i.p.	Partial hepatectomy	Nicotine	MWM 1, 3 days	Impaired: Surgery Not impaired: Surgery + Nicotine	[80]
Mice C57BL/6	Age 12–14 mos	Isoflurane	Tibial fracture with fixation	–	OF FC 3 days	Impaired: Surgery	[81]
Mice ADTg (APP/PS1) (ADTg-CypD KO)	AD transgenic, Cyclophilin D KO	Sevoflurane	Laparotomy (incision only)	–	FC 1–7 days	Impaired: Female ADTg Not impaired: Male ADTg; Female ADTg + CypD KO	[82]
Mice C57BL/6	Age 20 mos	Isoflurane	Laparotomy	Pioglitazone, GW9662	NOR OF PAT 4–7 days	Impaired: Surgery; Surgery + GW9662 Not impaired: Surgery + Pioglitazone	[83]
Rats Sprague-Dawley	Age 20–24 mos	Sevoflurane	Laparotomy	Compound C	MWM 2–6 days	Impaired: Surgery Improved: Surgery + Compound C	[84]
Mice C57BL/6	Age 18 mos	Isoflurane	Splenectomy	Glycyrrhizin	MWM 14–18 days	Impaired: Surgery Not impaired: Surgery + Glycyrrhizin	[85]
Rats Sprague-Dawley	Age 18–19 mos	Chloral hydrate, i.p.	Laparotomy	Minocycline	MWM 1, 3, 7 days	Impaired: Surgery; Surgery + Minocycline	[86]
Mice C57BL/6	Age 13–14 mos	Sevoflurane	Laparotomy	High-fat diet	MWM 2–9 days	Impaired: Surgery; Surgery + high-fat diet; High-fat controls	[87]
Mice C57BL/6	Age 16 mos	Isoflurane Pentobarbital sodium ( ICV implants)	Laparotomy ICV cannula implantation	L-798,106 ICV injection	OF FC 6 days	Impaired: Surgery Not impaired: Surgery + L-798,106	[88]

Note: Papers above included (1) preexisting vulnerability in the animal model (age, Alzheimer transgenic, metabolic syndrome), (2) wild-type mice at least 12 months of age, (3) type of anesthesia, (4) type of surgery performed, (5) cognitive testing conducted. PO = postoperative; pre-op = preoperative; ADTg = Alzheimer transgenic; LCR = low-capacity runner; HCR = high-capacity runner; OF = open field; PAT = passive avoidance test; MWM = Morris water maze; FC = fear conditioning; NOR = novel object recognition; OF = open field; RAM = radial arm maze, EPM = elevated plus maze; BM = Barnes maze; ICV = intracerebroventricular; SQ = subcutaneous; i.p. = intraperitoneal; i.v. = intravenous.

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

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Chapter 1 – Emergence Delirium Chapter 2 – Postoperative Delirium Chapter 4 – Postoperative Cognitive Improvement Chapter 12 – Biomarkers of Postoperative Cognitive Dysfunction: Finding the Signal amid the Noise Chapter 11 – Cognitive Testing for Perioperative Neurocognitive Disorder Chapter 14 – Preoperative Testing to Identify Vulnerable Subgroups