Prolonged Nonsurvival Studies
Most prolonged nonsurvival studies are carried out in anesthetized animals and extend over many hours or days. In a prolonged nonsurvival study of the visual system, the scientific needs of the experiment usually require that stationary images or images whose motion can be controlled accurately and repeatedly be presented to the retina, so self-generated movements by the animal must be minimized. Minimizing self-generated movements is also required in other neuroscience studies. Neuroscientists frequently resolve this problem by administering neuromuscular blocking drugs (NMBDs) that paralyze all voluntary muscles, including the extraocular muscles (Flecknell, 1987; Hildebrand, 1997). Many of the typical indicators of anesthetic depth (such as response to noxious stimuli and changes in respiratory rate) are thus eliminated, and this makes it difficult to assess whether an animal is experiencing pain and/or distress. But that assessment is critical, both for the welfare of the animal and to avoid compromising experimental results (Moberg, 1999). An earlier workshop on anesthesia and paralysis in experimental animals (Anonymous, 1988) considered the special problems associated with this approach, and they are also discussed in Preparation and Maintenance of Higher Mammals During Neuroscience Experiments (NIH, 1991), the predecessor of the present document. Without exception, the scientific need to use NMBDs must be explained in an investigator’s animal-use protocol and approved by the IACUC (NRC, 1996).
Principle V of the US Government Principles (IRAC, 1985) states that “surgical or other painful procedures should not be performed on unanesthetized animals paralyzed by chemical agents,” and the Guide further states (p. 65):
Neuromuscular blocking drugs . . . do not provide relief from pain. They are used to paralyze skeletal muscles while an animal is fully anesthetized. They
might be used in properly ventilated conscious animals for specific types of nonpainful, well-controlled neurophysiologic studies. However, it is imperative that any such proposed use be carefully evaluated by the IACUC to ensure the well-being of the animal because acute stress is believed to be a consequence of paralysis in a conscious state and it is known that humans, if conscious, can experience distress when paralyzed with these drugs (NIH, 1991; NRC, 1992).
The special concerns associated with prolonged nonsurvival experiments were well summarized in Preparation and Maintenance of Higher Mammals During Neuroscience Experiments (NIH, 1991):
The most critical issues in prolonged nonsurvival experiments arise in the context of anesthesia, maintenance of physiological state, and monitoring of the animal’s condition. The choice of anesthetic must jointly satisfy the need of the experimenter to perturb the preparation as little as possible and his/her obligation to ensure that the animal remains free of pain and distress. Maintaining an anesthetized (and often immobilized) animal in sound physiological condition for several days is a considerable challenge and monitoring both the anesthesia and the animal’s general condition requires careful attention to a number of kinds of measurement.
A variety of experimental protocols have been used to minimize the difficulties. The Guide should be interpreted as a flexible document in reviewing protocols of this sort, because procedures may vary with species and among different experimental paradigms.
The chief animal welfare concern associated with anesthetized paralyzed animals is that the behavioral indicators of pain and/or distress are inhibited by NMBDs, and this makes it necessary to use special measures to monitor and regulate anesthesia (Gibbs et al., 1989). Anesthesia must be regulated in such a manner that it exerts either no effect or a minimal and constant effect on the neurophysiologic responses being measured. Of both animal welfare and scientific concern is the problem of monitoring and maintaining the animal’s physiologic state, particularly in experiments that extend over several days (Lipman et al., 1997). Issues can also arise regarding infection, in that it usually is not possible to conduct prolonged nonsurvival neuroscience experiments aseptically and the duration of these experiments is sufficient to allow infections to develop.
A two-step paradigm is used in most prolonged nonsurvival neuroscience experiments. In one kind of study, during an initial step of 2–4 hours duration, the animal is surgically prepared for a subsequent data collection step, which follows immediately and can last for a few hours to several days (NIH, 1991). During the initial surgical preparation step, all procedures are completed under surgical anesthesia without NMBDs, and analgesics may be administered preemptively to augment the anesthetic regimen (see “Anesthesia and Analgesia” in Chapter 3). Other studies employ a variant of the two-step paradigm that involves performing, several days before the nonsurvival recording session, a survival surgery step during which various devices are implanted (such as a cranial pedestal and cham-
ber that are used to secure the animal and the recording device, respectively) (NIH, 1991) (see “Multiple Survival Surgeries” in Chapter 4). The latter approach frequently allows the neuroscientist to perform the initial-survival surgery step with aseptic techniques in a dedicated surgical facility. Furthermore, only minor procedures (such as venipuncture and endotracheal intubation) are needed on the day of the prolonged nonsurvival recording session, and they can be performed with only light anesthesia augmented by analgesics. Following these minor procedures, general anesthesia is provided for the nonsurvival recording session.
The difficulty involved in assessing whether paralyzed animals are free of pain and/or distress is acknowledged by the Guide, which states (p. 65):
Some classes of drugs—such as sedatives, anxiolytics, and neuromuscular blocking agents—are not analgesic or anesthetic and thus do not relieve pain; however, they might be used in combination with appropriate analgesics and anesthetics. Neuromuscular blocking agents (e.g., pancuronium) are sometimes used to paralyze skeletal muscles during surgery in which general anesthetics have been administered (Klein, 1987). When these agents are used during surgery or in any other painful procedure, many signs of anesthetic depth are eliminated because of the paralysis. However, autonomic nervous system changes (e.g., sudden changes in heart rate and blood pressure) can be indicators of pain related to an inadequate depth of anesthesia.
It follows that adequate anesthesia must be established and verified before the administration of NMBDs and initiation of the data-collection session. The animal should be maintained without NMBDs at a fixed anesthetic level until it is physiologically stable. It will take at least 30 minutes, depending on the duration of action of the anesthetics used during the initial surgical preparation period (if there was one), at a fixed anesthetic level and without change in physiological state to ensure that the animal is stable. That period should be used to establish and validate the physiological signs that will be monitored under paralysis to document that the animal is being maintained in a suitable condition. Experience has shown that care should be taken to ensure that the level of anesthesia established during this initial period is adequate but does not compromise neural responsiveness in the areas under study (NIH, 1991). That is critical because reducing the level of anesthesia after NMBDs have been administered is problematic. Obtaining a performance based assessment of the adequacy of the new anesthesia level may require that NMBDs be withdrawn to assess skeletal muscle response; however, this can entail difficulties because a long period may be needed to restore muscle responses (Hildebrand, 1997). Noninvasive assessment of neuromuscular function with examination of evoked responses of skeletal muscle to peripheral motor nerve stimulation can facilitate monitoring both the induction of paralysis and the recovery from NMBDs (Hildebrand, 1997).
Anesthesia and general physiologic state should be monitored for each animal during each procedure (Mason and Brown, 1997; NRC, 1996). Notations
should include the time, date (if appropriate), drugs or solutions administered, and the name or initials of the person making the entry. A number of physiologic measures are helpful in monitoring animals on NMBDs, including heart rate, electroencephalogram, arterial blood pressure, blood oxygen saturation, urine production and pH, end-tidal CO2 and/or blood gas concentration, rectal temperature, and general autonomic signs of arousal, such as salivation, pupil size, and lacrimation (Hildebrand, 1997; NIH, 1991). Physiologic measures should be documented periodically throughout an experiment. The periodicity of monitoring and documentation should be more frequent during the initial stages of experimentation (such as every 15 minutes) and gradually extended once the animal has been stabilized on NMBDs. The details of the specific physiologic measures to be monitored and the frequency and means of documentation should be described in the research protocol and approved by the IACUC. The use of automated multifunction measuring devices—which maintain a running record of such metabolic measures as blood pressure, blood oxygenation, expiratory CO2, and pulse rate—can greatly facilitate the monitoring of anesthesia and general physiologic state (Vogler, 1997). However, not all devices maintain a historical record, and regular measurements should be recorded under these circumstances. The use of automated devices cannot substitute for direct monitoring of the animal by a human observer (see also Mason and Brown, 1997), and a human observer should be present at all times during a prolonged nonsurvival procedure, as the clinical status of the animal can change quickly and require intervention. Monitoring data should be filed by experiment and animal and kept for at least the duration of the overall project (NIH, 1986).
The issues involved in maintaining an experimental animal in good physiologic condition during a prolonged recording experiment are similar to those involved in other situations that require the long-term maintenance of animals in clinical situations (NIH, 1991). Animals that are paralyzed must be ventilated artificially, and standard veterinary practices should be followed when selecting the gas mixture and anesthetic or analgesic used (Vogler, 1997). In some cases, gaseous anesthetics or analgesics are included in the gas mixture (such as isoflurane and N2O), while in other cases, room air with or without added oxygen is used. The use of a mixture of 50% O2 and 50% N2O may be helpful, in that N2O potentiates intravenous anesthetics (NIH, 1991). However, N2O increases cerebral blood flow, and this side effect may be of special concern to researchers performing intracranial procedures (Drummond et al., 1987). Animals exposed to artificial ventilation for long periods should be hyperinflated (sighed) at regular intervals to help to avoid collapse of the pulmonary alveoli (atelectasis) (Vogler, 1997). Hydration of the inspired gases is also helpful in preventing desiccation of lung tissues. That can be accomplished by passing the air through fluid in a flow-through system or by using a closed-loop system that obviates additional hydration; passive humidification devices are also effective, inexpensive, and appropriate for multi-day use. A straightforward way of monitoring the adequacy of
artificial ventilation is to measure end-tidal CO2 either continuously or at frequent intervals (Vogler, 1997).
Animals are typically given an osmotically balanced fluid and metabolites, such as lactated Ringers solution with dextrose, intravenously (DiBartola, 2000; Haskins and Eisele, 1997). When data collection goes on for several days, supplementation of that solution with potassium and/or amino acids may be desirable. For shorter experiments (under 48 hours), periodic subcutaneous administration of fluid may be sufficient. In all cases, the total volume administered should be adequate to make up for what is lost through the skin and lungs and should be sufficient to maintain renal function. The role of the kidneys in maintaining pH and osmotic balance is critical, and normal renal function can be particularly important in preventing the imbalances that may occur when animals are subjected to prolonged artificial ventilation. Monitoring urine output may be helpful in some situations to ensuring adequate renal function and hydration.
During long experiments, rearranging an animal’s limbs and body and massaging the large muscle masses regularly can help to prevent the edema and venous pooling that occur in the absence of muscle tone and movement. Providing regular doses of antibiotics, vitamins, and anti-inflammatory agents may help to keep an animal in a stable condition and prevent infection (NIH, 1991). Core body temperature should be monitored throughout the period of paralysis, and supplemental heat should be provided as needed with forced air or circulating water heating pads (e.g., Vogler, 1997).
Evaluating the need for aseptic technique in prolonged nonsurvival experiments requires the professional judgment of the investigator, veterinarian, and IACUC case by case. In general, the need for asepsis will depend on the duration of the experiment and the extent to which it involves the exposure of tissues or body cavities. As stated in APHIS/AC Policy 3, “nonsurvival surgeries require neither aseptic techniques nor dedicated facilities if the subjects are not anesthetized long enough to show evidence of infection.” Any procedure that lasts longer than 12 hours and involves exposed tissues or body cavities presents a significant opportunity for infection to occur, and the risk increases with the length of the procedure (Knecht et al., 1987; McCurnin and Jones, 1985).
Failure to use aseptic procedures in prolonged nonsurvival experiments increases the possibility that research data will be compromised and increases the risk of premature death due to sepsis. Either of those outcomes could entail the use of additional animals, which would be inappropriate for both ethical and regulatory reasons (see U.S. Government Principle III). It would also violate the requirements of the Guide and the AWRs for the provision of adequate veterinary care. The use of aseptic procedures also helps to ensure that students learn proper surgical technique and strengthens an institution’s ability to respond to public inquiries about the use of animals in research.
When a preparatory survival surgical procedure is conducted to implant devices, such as a pedestal and chamber, it should be performed under aseptic
conditions (AWR 2.31(d)(1)(ix); NRC, 1996). Most implanted devices and their carriers can be disinfected, but it might not be possible for some sensitive or delicate equipment, such as some types of microelectrodes (for further information on this topic see “Animal Care and Use Concerns Associated with Introduction of Probes into Neural Tissue” in Chapter 4). Whenever possible, it is advisable to sterilize or disinfect devices before their insertion into neural tissue. Because a typical neuroscience laboratory contains many other items, such as recording equipment, that cannot be sterilized, the full application of aseptic technique during a prolonged nonsurvival experiment is usually impossible. One approach to that problem is to use appropriate aseptic technique to create and maintain a local sterile field that includes any openings into major body cavities that are made during a prolonged nonsurvival session.
Institutions should develop policies and guidelines to assist investigators in adapting aseptic surgical procedures to the laboratory setting. Topics that should be considered in preparing guidelines include preparation of the laboratory room, with particular attention to the site where surgery and recording will take place (for example, taking into account the relative locations of supply and exhaust ventilation ducts with respect to airborne contamination of the surgical field); preparation of the animal; preparation of the surgeon and any other experimenters who will come into proximity to the animal; instrument preparation; intraoperative monitoring; and training (APHIS/AC Policy 3; NRC, 1996, pp. 78–79). Neuroscientists can assist veterinarians and IACUCs in developing performance-based standards for monitoring the occurrence of deleterious effects by providing postmortem tissue specimens for histopathologic analysis.