Publicatiedatum: Exoticscon, september 2023

Not One More Guinea Pig: A Guide to Safe Anesthesia
and Surgery

Authors: Eva Stoffels, PhD, and René Remie, PhD.
Affiliation: From Marumoto Veterinary Practice, Hoevenweg 18, 7722 PN Dalfsen, The Netherlands (Stoffels), René Remie Surgical Skills Center (RRSSC), Operetteweg 27, 1323 VK Almere, The Netherlands (Remie).

Abstract: This work describes the authors’ experiences with surgery of pet guinea pigs. To reduce mortality, the principles of good surgical practice must be observed. The preoperative assessment of the patient plays a pivotal role. Both visual observation and quantitative markers (blood lactate and urinary parameters) are of importance. Detailed anesthetic protocols, including monitoring, are provided for typical procedures (cutaneous, dental, abdominal). The most common causes of complications are discussed, and troubleshooting measures are proposed.

Introduction

Anesthesia and surgery of guinea pigs are still considered challenging. Complications are common; in one study, perioperative mortality of about 4% was reported.¹ It was unclear whether this pertained to all situations where anesthesia was required or only to specific types of surgery, but this number can be nevertheless considered surprisingly high.

The author performs two or three surgeries daily, amounting to about 500 in a year. About 70% of these are dental procedures, where the mortality risk is practically zero. The second most frequent procedures are non- visceral soft tissue surgeries (tumor excision, vasectomy), with similarly low complication rates. Finally, only laparotomic surgeries are associated with some risks; however, these can be significantly reduced. In the author’s experience, the overall mortality rate lies well below one percent. In this work, the authors share experiences and successful protocols based on cases from the practice.

Many factors determine the success versus failure of surgery including the patient’s initial condition, the procedure’s invasiveness, the surgeon’s skills, the duration and depth of anesthesia, the choice of drugs, and, eminently the adequate analgesia protocol. The principles of good surgical practice and typical anesthetic protocols for different types of surgery are presented.

Finally, the most common postoperative complications are discussed: the instantaneous (i.e., in the operating theater) and the delayed problems (i.e., in the recovery, up to several days after surgery). The underlying mecha- nisms are elucidated, and troubleshooting protocols are provided.

Patient Assessment

To do or not to do?
Even when carried out optimally, surgery still causes pain and tissue damage. In addition, the inevitable release of catecholamines, cortisol and pain mediators causes postoperative immunosuppression² and makes the patient prone to opportunistic infections. Miracle surgeries do not exist; there will always be adverse effects for the animal to cope with. The problems arise proportionally to the duration and depth of the anesthesia and the degree of tissue damage. Therefore, the surgeon should always perform the risk-benefit analysis and make a prudent decision.

Correct evaluation of the patient has a pivotal impact on the success of the surgery. The American Society of Anesthesiologists (ASA) classification³ (I-V, where I refers to a healthy animal and V to a moribund one) applies to guinea pigs, although assigning the animal to the proper category may sometimes be challenging. Nevertheless, even before a thorough health examination, the surgeon must address the following questions:

• Can the animal survive the surgery?
  Surgery never warrants instant improvement. The animal must overcome the perioperative stress. Patients belonging to ASA IV and V categories are unlikely to do so. In such cases, humane endpoints should be considered. For example, animals with severe respiratory distress or metabolic acidosis should not undergo surgery unless they can be stabilized first.

• Is surgery essential?
  Non-essential (elective) surgeries should be performed solely on ASA I patients. Thus, the surgery should be postponed until the animal is disease-free. For example, an animal with (even slight) dyspnea, digestive problems, or a poor body condition should not be neutered. In addition, old animals (> 4 years) should not undergo elective surgeries because the postoperative immunosuppression may prove fatal (see Complications).

• Does surgery actually solve the problem?
  Answering this requires a correct and complete diagnosis. Multiple issues may present, but it is prudent to tackle only the ones that require intervention. For example, a female with cystic ovaria should not be spayed. Ovarian cysts are a symptom (of chronic stress⁴) rather than a disease. Surgery will definitely not solve the problem. The cause of chronic stress (e.g., cystitis) must be addressed with priority.

• Will surgery improve the quality of life?
  This question addresses the ethical side of veterinary medicine. Many procedures are technically feasible and may even prolong the animal’s life, but they result in severe debilitation, handicap and/or pain. For example, a permanent nasogastric tube, gastric cannulation or even total intravenous nutrition via a jugular catheter are possible. The authors would rather consider humane endpoints if these are palliative measures.

• Are there preexisting diseases?
  A well-considered decision must be taken if the animal is known for previous heart, kidney or liver issues. For example, an animal with end-stage kidney disease should not undergo surgery because the chance of recovery and life expectancy are slim.

• Are there less invasive treatment methods?
  Surgery should not be considered the gold standard but rather the last resort. Depending on the animal’s condition, less invasive methods should be attempted first. For example, small-sized bladder or ureter stones do not always require surgery but can be expelled using certain medications.⁵ Many uroliths in females can be removed using a cystoscopic procedure, thus avoiding laparotomy.

In the following paragraphs, methods of perioperative assessment are described in detail.

Pain has devastating physiologic impacts on guinea pigs and presumably on other prey animals. The type of nociception determines the severity of the physiologic response.

• Somatic pain is the result of integumentary damage, muscle and skeletal issues. The neural signal transmission is uncomplicated, via sensory fibers through the neospinothalamic tract and a single synapse in the thalamus, with little projections onto the brain stem and the limbic system.⁶ Somatic pain is well-localized and rarely accompanied by autonomous responses. Guinea pigs tolerate this type of pain very well – even deep wounds do not alter the animal’s behavior, appetite and digestion. With adequate pain management, the prognosis is usually favorable.

• Visceral pain is the result of stimulation of various nociceptors in the thoracic and abdominal cavity (e.g., pleura, peritoneum, urogenital system). It is transmitted through the archi- and paleospinothalamic tracts, with multiple synapses in the spinal cord and projections onto the brain stem (reticular formation and periaqueductal gray, and indirectly nucleus raphe and locus coeruleus).⁶ This type of pain is not well-localized but rather experienced as general malaise. A remarkable autonomous response (due to catecholamine and glucocorticoid release) results in circulatory, digestive and other (electrolyte, thermoregulation) disturbances. Guinea pigs poorly tolerate visceral nociception. An animal suffering any grade of visceral pain can be automatically considered ASA III or higher.

• Dental pain sensation may be seen as a hybrid of the abovementioned categories. The trigeminal nerve transmits the nociception and follows a multi-synaptic pathway in the spinal cord.⁶ There is a moderate autonomous response. The animals are visibly sick and often develop secondary digestive issues. Sadly, this is the most common type of pain in guinea pigs.

Guinea pigs are often tricky to decipher. Like other prey animals, they tend to hide their pain and discomfort. Simultaneously, they are intolerant to (especially visceral) pain. Thus, animals with obvious pain symptoms are in a much worse condition than the examiner might think. In addition, the communication of pain and discom- fort is quite different from other species: there is (usually) no vocalization, and the behavioral changes (body language, facial grimaces) may be subtle. Several attempts have been made to standardize the facial expressions (e.g., orbital tightening, whisker and ear position) in response to certain pain stimuli in rats, mice and rabbits.⁷ Similar signs can also be used to evaluate pain levels in guinea pigs, but this approach is unreliable as it does not accurately reflect the severity of pain.

Recognition of pain, or more generally, the poor condition of an animal, requires experience and skills. Features indicative of severe (usually visceral) pain and malaise are listed below:

• Body temperature. Critically ill guinea pigs usually have a low body temperature. They feel cold to the touch, and their rectal temperature may be 37 °C or lower. ASA IV or higher.

• Bradycardia. A heart rate lower than 150 bpm in a conscious animal indicates serious problems. Surgery should be postponed until the patient is stabilized.

• Mastication. Healthy guinea pigs masticate involuntarily. The absence of this reflex is a sign of severe pain (ASA IV or higher).

• Anorexia. An animal that eats less than usual but still shows interest in feeding may suffer light to moderate pain, such as pain due to dental issues. An animal that totally refuses to feed is in a critical condition (ASA IV or higher).

• Posture and muscle tone. Guinea pigs are usually alert and exploring. Critically ill animals appear depressed and tend to gaze downwards (Fig 1). There is remarkable muscle relaxation upon handling. This is called the sedative effect of pain, most likely caused by activating a2 adrenergic receptors in the brain stem.⁸ Surgery should be postponed until stabilization.

• Body condition and fur condition. Cachexia and an unkempt coat are always signs of chronic suffering. Animals in pain may lose weight even if they (apparently) feed normally. This is the catabolic effect of pain. They also stop grooming and often display alopecia and other skin conditions (the telogen effluvium effect). These patients (usually ASA II and III) can undergo surgery, but caution is required.

• Stereotypic behavior. Signs such as bruxism, head nod, twitches or tremors usually indicate pain. Not all animals display these signs.

Figure 1. Postural changes due to visceral pain. Above: an animal with enteritis (right) and healthy animals (left). Note that the sick guinea pig does nog lift its head. Below: a moribund animal with end-stage kidney disease.

Quantitative markers

While the visual evaluation usually provides sufficient information, in some instances, one should resort to unbiased, quantitative criteria such as those obtained through blood and urine analysis.

The authors do not advocate routine blood sampling before surgery. It inflicts a substantial amount of stress, while the outcome is usually of little diagnostic value. For example, the electrolyte concentrations (Na+, K+, Cl-) can deviate from the normal range in healthy animals, while they can lie in the normal range in terminal animals. The renal (urea, creatinine) and hepatic (bilirubin, ALT) parameters have some positive predictive value, but they are elevated only in end-stage diseases (ASA IV/V, not suitable for surgery). The same is valid for bicarbonate; it is affected only in terminal situations (e.g., shock), but otherwise, it lies in the normal range. There are a few exceptions that do have a high diagnostic value, which are listed below.

• Lactate. Lactate is the product of anaerobic glucose metabolism; lactate accumulation indicates inadequate oxygen supply to the tissue due to circulatory failure or inadequate ventilation. Blood lactate in healthy guinea pigs is 0.7-3 mmol/L (n = 20, measured using a Roche Accutrend Plus meter). Hyperlactatemia (4-10 mmol/L) occurs in life-threatening disorders (n = 13), such as paralytic ileus/tympany, entero- toxemia, (septic) shock and heart failure. Values above 10 mmol/L are only seen in terminal animals (n = 7). The authors’ cut-off value is about 7 mmol/L. Patients with higher values are unlikely to survive anesthesia and surgery. Lactate can also be measured in urine; the values are about 50% higher than the blood values, but they can be helpful in decision-making.

• Creatine phosphokinase (CPK). The typical values are below 200 IU/L. Elevated CPK indicates rhabdo- myolysis, which in turn suggests a severe systemic disease. Animals with CPK > 1000 IU/L are unlikely to survive the surgery.

• Blood or urine lactate is particularly valuable in preoperative assessment. Small amounts of blood can be collected from the ear vein, as described by Nemeth et al.⁹ This is relatively uncomplicated and does not excessively stress the animal.

• Urinalysis should be performed routinely because it yields crucial information about the patient’s con- dition. Urine can be obtained by spontaneous micturition or by applying gentle pressure to the caudal abdomen. The following values are most informative:

– Urinary pH. This is probably the most critical parameter in patient evaluation. Normal guinea pig urine is strongly alkaline (pH 8-9). Lowered values indicate acidosis and are only seen in critical patients. Interestingly, respiratory acidosis (e.g., due to a lower respiratory tract disease) yields only moderate pH reduction (7-8). Acidic urine (pH 5-6.5) always indicates life-threatening metabolic acidosis. Such values are usually accompanied by strongly elevated blood or urine lactate. The guinea pig may still appear alert, but if left untreated, it will die within maximally 48 hours due to multiple organ failure. The cause of metabolic acidosis (enterotoxemia, septic shock, cardiogenic shock etc.) should be established and treated if possible. Guinea pigs with metabolic acidosis must not undergo surgery as the survival rates are negligible.

– Hemoglobin. This must be differentiated from hematuria, as is common in cystitis or urolithiasis.sup>5 In terminal patients, urine may contain hemoglobin due to intravascular hemolysis; the corresponding pH is usually strongly acidic (pH 5). The blood samples from such animals are hemolytic. Stabilization can be attempted but is seldom successful. Euthanasia should be considered.

– Specific gravity (SG). Normal values, measured with a refractometer, should not exceed 1.020 g/ml. Since SG reflects the hydration state of the animal, it should be known before administering any fluids. Fluid replacement in animals that are actually not dehydrated can lead to fatal lung and brain edema. If SG is higher than 1.020 g/ml, fluids warmed up to the body temperature (38 °C) may be administered at 20 ml/kg through the subcutaneous route in the shoulder region. This is the easiest and safest method of administration of large volumes of liquids to guinea pigs. The resorption from this site is relatively fast (less than 5 minutes). The IV route (cephalic vein) is also possible, but catheter insertion requires sedation. The IV fluids must be infused very slowly to avoid overloading the circulation. Usually, 0.9% NaCl, NaCl with glucose (blood glucose should be determined before) or lactated Ringer’s are used. In cases of metabolic acidosis, these solutions should be avoided because of their low pH. Instead, Plasmalyte (pH 7.5) with acid-base correction can be used (see Complications).

– Nitrogen species. Nitrate is an important marker molecule. In the urine of healthy guinea pigs, it does not exceed 300 mg/L; however, it is markedly elevated (500-900 mg/L) in animals fed a high-nitrate diet (greens). These animals usually suffer from (chronic) cystitis and urolithiasis. In animals fed a low-nitrate diet, elevated values result from increased inducible nitric oxide synthase (iNOS) activity; nitrate is a nitric oxide metabolite.10 In- creased iNOS activity is a sign of an inflammatory process; values higher than 1000 mg/L are alarming. Animals with high urinary nitrate can still undergo surgery but are prone to complications.

After the assessment, and if necessary, stabilize the patient. To do this, the surgeon must choose an adequate protocol.

Protocols

Various techniques of laboratory animal anesthesia have been described in the literature.¹¹ Good anesthetic practice comprises unconsciousness, adequate analgesia, muscle relaxation and autonomous block (in certain types of surgery). The stages of anesthesia (anesthetic depth), initially described by Guedel,¹² are depicted in Figure 2.

After testing the available methods including inhalation versus injectable anesthesia, various (pre) medications, intravenous versus intraperitoneal or subcutaneous administration routes, epidural versus infiltration nerve block, the author (ES) has selected a relatively simple protocol that includes injectable drugs and inhalation anesthesia. Depending on the invasiveness of the procedure, three variants (i.e., light, regular, and heavy) can be distinguished. Even in heavy protocols, the anesthetic depth should never exceed stage 3.2, as explained in Figure 2.

Figure 2. Stages of anesthesia after Guedel.¹² Stages 3.1 or (maximally) 3.2 can be considered safe. Source: Guedel AE. Inhalation Anesthesia: A fundamental Guide. New York, NY
The Macmillan Co.; 1937:14.

General preparatory measures include:

• Unlimited access to high-quality feed and water. Guinea pigs must not be fasted before surgery, because this will increase the rate of gastrointestinal complications and significantly reduce survival. Anorexic animals should be fed about 1 hour before anesthesia and given parenteral fluids if necessary (see above). Just before induction, the mouth of the animal may be lavaged to remove food remains.

• Preemptive administration of analgesics and antibiotics. It is crucial to increase the threshold of sensory nerve firing even before the actual tissue damage has occurred. This will help to reduce postoperative hyperalgesia.¹³ NSAIDs such as meloxicam (1 mg/kg SC) are usually sufficient. The author (ES) also advocates prophylactic use of antibiotics (e.g., enrofloxacin, 10-20 mg/kg SC) because this suppresses the intestinal pathogens and may moderate or eliminate post-surgical wound infections. Especially in unclean procedures, such as male neutering or vasectomy, dental procedures or abscess treatment, peri-operative use of antibiotics has undeniable benefits.

• Other premedications. Benzodiazepines, notably midazolam (1 mg/kg SC), have anxiolytic properties, cause anterograde amnesia and produce excellent muscle relaxation.¹⁴ Preoperative administration of midazolam facilitates isoflurane induction. Midazolam also acts as an appetite stimulant postoperatively. Therefore, it is highly recommended for lengthy procedures (> 20 minutes) because it reduces the risk of secondary gastrointestinal disorders resulting from anorexia. It does increase the interval between the cessation of the general anesthesia and the time the animal fully regains consciousness. In critical patients, this may have adverse effects. If desired, midazolam can be antagonized with flumazenil.

Induction
After administering the drugs of choice, the animal is placed on a heated pad or surface (Fig 3) to assure a con- stant body temperature (about 38 °C). Unfortunately, general anesthesia impairs the thermoregulatory center, making the animals extremely prone to hypothermia. Hypothermia should be avoided at any cost because it leads to bradycardia and cardiac arrest.

Artificial tears can be applied to prevent cornea desiccation. Unlike in rabbits, rats or chinchillas, this step is not essential in guinea pigs, which do produce tears during isoflurane inhalation. Most guinea pigs keep their eyes closed while anesthetized with isoflurane.

Anesthesia is induced with isoflurane at 5% while the oxygen flow is maintained at 1.5 L/min. The gasses are administered through a nose mask (guinea pigs are obligate nasal breathers). The authors believe intubation is not beneficial unless the procedure requires mechanical ventilation. The specific anatomy of the nasopharynx, especially the very tight palatal ostium (Fig 4), limits access to the larynx.¹⁵ Several intubation methods have been described,¹⁶ but they usually require relatively deep anesthesia (with concomitantly higher complication rates, see Fig 2) and may cause trauma to the vulnerable tissue. Airway obstruction is seldom an issue in guinea pigs. These animals cannot vomit and usually do not regurgitate unless a2 adrenergic agonists (medetomidine, xylazine) have been administered. The patency of the airway (absence of congestion) must always be checked prior to surgery (see Patient Assessment).

The duration of induction with 5% isoflurane determines the eventual depth of anesthesia. Only brief immobilization is sufficient for light procedures (e.g., dental corrections). Keeping the animal in dorsal recumbency should be possible, usually achieved after 5 to 10 seconds inhalation. For standard surgical anesthesia, inhalation is sustained until the pedal reflex, and the eyelid reflex disappears (about 15-20 s); withdrawal reflexes in response to deep pain sensation (pinching the paw) may still be present. After prolonged inhalation (> 30 seconds), the breathing pattern may change from costal to abdominal, and the breathing rate may slow down; the anesthesia is already too deep (stage 3.3 or higher in Fig 2), and the isoflurane concentration must be reduced immediately.

The optimal maintenance dose of isoflurane is 1.5%. Anesthetic depth can be regulated with reasonable accu- racy, but the working levels should never be higher than 2-2.5% because of the risk of respiratory depression. Isoflurane anesthesia is suitable for more lengthy procedures because, at 1.5%, it does not lead to tissue hypoxia and lactate accumulation (Fig 4).

Figure 3. (A) A guinea pig on a heated pad, (B) fixed using a flexible holder. (C) In this case, a mammary tumor was excised.

Figure 4. Blood lactate as a function of the exposure to 1.5% isoflurane in 1.5 L/min oxygen in three healthy animals.

Monitoring and troubleshooting
Generally, the deeper the anesthesia, the more careful patient monitoring is required. The most used options are pulse oximetry, ECG and capnography.

Pulse oximetry provides a very accurate and sensitive method of evaluating the animal’s ventilation: it simultane- ously records the oxygen saturation and pulse waveform. The sensor can be attached to the carpus or tarsus. Ears and tongue are not reliable, because of inadequate signal intensity. A problem arises in animals with pigmented skin; the technique is still applicable, but the signal quality deteriorates. If no waveform can be obtained, other monitoring techniques should be used. For example, capnography is possible without intubating the animal; however, a tightly fitting mask with an extra port is necessary. ECG is less suitable because it is not very sensi- tive to myocardial hypoxia. Besides, old or stressed animals may display aberrant ECGs, which are difficult to interpret.17 ECG can still be used to detect the pulse in anesthetized animals, but it is not the first choice for monitoring rodents.

Optimally, the pulse oximeter displays a robust and stable signal, with a saturation between 90 and 100%. In dark pigmented animals, 80-90% is still acceptable. The waveform should be monophasic, and the rate should be between 160-220 bpm. Higher rates (230 bpm and more) suggest that the animal is merely immobilized but not fully anesthetized; this may be sufficient for minor procedures.

The most common problem is a drop in oxygen saturation to less than 80%. Most of the time, this usually occurs after induction and may be transient; however, measures should be taken if the animal fails to restore the normal saturation within minutes. For example, if the procedure is elective, one should consider abandoning it. Essential surgeries still must be carried out; doxapram (5-10 mg/kg SC or IP) can be used in such cases. Doxapram is an excellent respiratory stimulant which should always be available in the operating theater. The animal’s head and thorax can be slightly tilted to alleviate the pressure on the diaphragm from the intestines.

Sudden changes in the pulse frequency and waveform are often indicative of problems. An increase in the heart rate suggests that the animal feels pain. Additional local anesthetics should be administered, and/or the isoflurane concentration can be temporarily increased. A decrease of the pulse frequency to less than 150 bpm accompanied by an irregular and weak waveform is an emergency situation, usually caused by a baroreflex fol- lowing excessive catecholamine release (severe visceral pain) or hypovolemia (bleeding). Besides doxapram and lidocaine administration, atropine should be slowly titrated to effect in increments of 0.2 mg/kg (usually IP in laparotomic procedures). If hypovolemia due to blood loss is the cause, fluids should be administered (usu- ally IP). The isoflurane concentration can be reduced to 0.5-1%, and more local anesthetics can be applied. The procedure should be finished as fast as possible.

Local anesthetics
Tissue damage is inherent to surgery. During tissue handling, inflammatory mediators (prostaglandins, cytokines) are released, triggering the nociceptive fibers to fire.¹³ During general anesthesia, pain mediators and neurotransmitters accumulate, but pain transmission is temporarily blocked at the level of the brain stem (reticular formation). After termination of anesthesia, intense pain is experienced: even minor stimuli elicit a strong pain response (Fig 5). Postoperative pain slows the recovery and may cause the animal to auto-mutilate (e.g., chew on the sutures). Local anesthetics, especially lidocaine, provide a solution to this problem.¹⁸ Lidocaine has very low toxicity in guinea pigs; it can be used for local infiltration at doses as high as 20 mg/kg. Adding epinephrine (e.g., xylocaine 2% with 5 μg/ml epinephrine) increases the duration of action of the lidocaine and may be suitable for most surgeries. Alternatively, longer acting anesthetics such as bupivacaine can be used (with caution).

Local anesthetics can be titrated to effect – additional amounts can always be administered during surgery, such as onto the incision site, into the periodontal ligament in dental procedures, or intraperitoneally during laparotomy. Lidocaine can also be used for epidural or paravertebral nerve blocks. However, this is much more difficult to titrate and adds complexity to the protocol, with no clear benefits.

Using local anesthetics allows us to keep the depth of anesthesia within safe limits, which results in much lower complication rates. Its advantages are enormous and should never be omitted in any surgery.

Figure 5. A schematic depiction of pain stimulus-pain response dependence. Tissue injury causes the curve to shift to the left; as a result, non-noxious stimuli are perceived as painful. (Source: Wikipedia).

Light, regular and heavy protocols

• Light. Short procedures (< 15 minutes) with minor tissue damage require no elaborate protocols or careful monitoring. There is no need to administer midazolam. The animal may even remain partly conscious (i.e., some reaction to pain stimuli may still be present, being a major advantage in dental procedures, which are often explorative). For example, while correcting the clinical crowns is painless because of their rudimentary innervation,19 the actual sources of pain (e.g., wounds, abscesses, infected periodontal pockets, loose or fractured teeth) can be easily identified in this way. At the slightest pain reaction from the animal, lidocaine should be administered (to effect, 2-3 mg) directly into the periodontal space. The same light protocol can be applied to minor cutaneous surgery, like removing small skin tumors or opening abscesses other than odontogenic abscesses.

• Regular. Moderately invasive procedures that may cause substantial (somatic) pain require surgical anesthesia and monitoring. Therefore, the use of local anesthesia is always essential as part of a balanced, multi-modal analgesia/anesthesia plan when extensive tissue damage is expected, tramadol (20 mg/kg SC) should be added to the preoperative cocktail. Tramadol is relatively well-tolerated; unlike other opioids, it does not cause gastrointestinal complications. This is most likely due to its SSRI (selective serotonin reuptake inhibitor) properties.²⁰ Typical examples of regular procedures are male neutering or vasectomy, removing large tumors (e.g., mastectomy) and treating odontogenic abscesses.

• Heavy. This category contains all laparotomic procedures because they inevitably inflict visceral pain. The full preoperative cocktail should be administered, and preferably metamizole (100 mg/kg SC) should be given instead of meloxicam. Metamizole is an NSAID that is more effective in targeting visceral pain than meloxicam.²¹ In the authors’ opinion, administering midazolam to critical patients is not beneficial even if its action is reversed with flumazenil. Maropitant (2 mg/kg SC) can be added. Examples include urinary tract surgery including cystotomy or ureterotomy. During all laparotomies, special care must not be taken to traumatize the viscera (intestine), as this will cause postoperative ileus (see Complications).

Good Surgical Practice

The principles of good surgical practice were introduced in the XIX century by an eminent American surgeon, William Stewart Halsted. These principles can still be considered the gold standard in modern surgery.²² The whole concept can be summarized in one sentence: the tissues should be treated respectfully. Therefore, in practice, the following must be observed:

• Strict asepsis. Thorough disinfection of the incision site can be performed using chlorhexidine solution – alcoholic on the skin, water-based on vulnerable surfaces such as mucosae. Povidone-iodide can be used as well, but not in combination with chlorhexidine. Contamination of the wound with iodine will impair wound healing. The authors prepare the sterile operation site using Press’n Seal cling film.²³ This provides an antibacterial and hydrophobic barrier that sticks to the skin. After disinfection of the film (alcoholic chlorhexidine), the adhesive spray is applied, and a sterile drape is fixed on the top (Fig 6). This double barrier provides good protection. Clamps should never be used because they cause painful wounds. Alternatively, other sterile clear drapes are commercially available.

• Hemostasis. Blood loss during surgery should be prevented. This is achieved by localization of major blood vessels and careful dissection to avoid incidentally damaging them. Certain surgeries carry a risk of excessive bleeding. Mastectomy is especially notorious because mammary gland tumors contain both large blood vessels and fine networks of capillaries. This can result in massive as well as diffuse bleeding. Doppler ultrasound examination should be done before surgery to identify vascular structures to minimize blood loss. After identifying the location of blood vessels, about 2-5 milliliters of 0.9% NaCl (with lidocaine) can be injected around the site. In this artificially created edema, blood vessels are visible during dissection and can be ligated before damage occurs. Blood loss during an average surgery is less than 0.3 milliliters. During excision of large tumors it should be less than 2 milliliters (2% of the blood volume). Bloodless surgery makes excessive administration of fluids redundant because the circulating volume is unaffected.

• Minimizing tissue damage. The surgeon has to avoid excessive grasping of the tissue with instruments. This particularly applies to guinea pig viscera, which do not tolerate handling. In laparotomic procedures, externalization of viscera should be avoided. For example, cystotomy should be performed through a small (< 2 cm) incision in the caudal linea alba, so that only the bladder is elevated out of the abdominal cavity (Fig 6). If internal organs must be externalized, it is imperative to keep the serosa moist using warm saline.

• Anatomic reconstruction of tissues. The anatomic structures (e.g., subcutis, muscles) should be restored to their original position. Tissues should be sutured layer-by-layer, the blood supply must be preserved, and tension on the sutures should be avoided. Good apposition of wound edges is crucial for optimal recovery and wound healing. Usually, the standard running (continuous) pattern with good edge apposition provides the best healing. It is superior to intracutaneous sutures because these provide less stability and may result in dehiscence. Be careful not to introduce foreign tissue into the peritoneum when sutur- ing internal organs because this will elicit a strong inflammatory response. For example, after closing a hollow organ (bladder), the second inverting suture (Cushing) must be applied so that no mucosa or muscularis of the bladder is exposed (Fig 6). Failing to observe this will result in postoperative pain and if the animal survives, intra-abdominal adhesions.

• Reduction of dead space. The space between structures (e.g., tissue layers) that are not attached to each other can fill up with blood, serum or exudate. This creates the perfect environment for bacterial growth and may lead to wound dehiscence. The same is valid for necrotic tissue left at the operation site. Dead space should be avoided by applying anchoring sutures between the layers. For example, while closing the abdominal wall (in three layers), the subcutis should be anchored to the muscular layer and the skin to the subcutis. Any necrotic tissue should be carefully removed.

The surgery should be completed in less than 30 minutes for the best result. Prolonged exposure to anesthesia (> 60 minutes) significantly increases the complication rates.

Figure 6. Left: the patient is covered with Press’n Seal plastic wrap. Right: after disinfection of the plastic, a sterile drape is placed on top of it. In this case, a large urolith was removed. Note that only the bladder was externalized through the incision and that an inverting suture pattern was applied.

Aftercare

Complications

In the operating theater
These complications are infrequent; the author (ES) experienced less than ten such cases in about ten years. In most cases, the animal’s life was terminated deliberately (intraoperative euthanasia) because surgery could not solve the problem, and a humane endpoint decision was made.

The most common cause of death in the operating theater is cardiorespiratory depression. It manifests as a drastic drop in oxygen saturation and heart rate, followed by ventricular escape beats (extreme bradycardia with 50-100 bmp) and cardiac arrest. Reasons for this can be:

– A sudden catecholamine release (nociception/pain that the animal feels despite anesthesia) triggers a baroreflex. This can be anticipated and prevented by adequate monitoring and applying local anesthesia (see Monitoring and troubleshooting).

– The use of a2 adrenergic agonists, especially in patients belonging to ASA III or higher, is likely to trig- ger a baroreflex.8 Besides, a2 agonists cause nausea and regurgitation and increase the risk of airway obstruction in non-intubated animals. For these reasons, the author (ES) never uses these drugs except for euthanasia.

– Hypovolemia due to excessive blood loss.

– Hypothermia. Anesthetized animals have impaired thermoregulation; a low body temperature potentiates endogenous acetylcholine²⁵ and may lead to bradycardia.

– Cardiac arrest due to preexisting cardiovascular diseases. This is usually due to the inadequate preopera- tive assessment of the patient.

Most of the abovementioned causes are due to procedural flaws. However, thorough patient assessment, adequate monitoring and observing the rules of good surgical practice allow the elimination of casualties in the operation theater.

In recovery
The general rule reads: the more profound the anesthesia, the more prolonged and more invasive the procedure, and the worse the condition of the animal before surgery, the higher the risk of complications in recovery. Typical postoperative issues are described below:

– The patient does not fully regain consciousness and dies within several hours. This is usually due to an incorrect preoperative assessment: unstable patients (ASA IV/V) should not undergo surgery unless preoperative stabilization is possible. Too deep anesthesia (> stage 3.3, Fig 2) and using a2 agonists contribute to the risk. Alternatively, this can result from procedural flaws. Examples include inadequate suturing of internal organs, which leads to leakages (such as uroperitoneum), or septic shock due to a body cavity contamination with infectious material (abscess, fecal contamination). Treatment is unrewarding, and euthanasia should be considered.

– Simple postoperative ileus. The animals may appear awake and alert but refuse to feed and display signs of visceral pain (see Pain recognition). The fecal output is significantly reduced, and sometimes tympany is present. This type of complication is most often seen in non-visceral procedures carried out using a2 agonists, even if the antidote (atipamezole) has been administered. The mechanism of gastrointestinal stasis is most likely the adrenergic-induced disruption of intestinal circulation,⁸ which leads to temporary hypoxia. If there are no other complications (e.g., dysbiosis, metabolic acidosis), the patients can be treated with high doses of prokinetics (metoclopramide 1 mg/kg SC q3-4h, cisapride 1 mg/kg PO q3-4h), visceral analgesics (see Light, regular and heavy protocols), fluid replacement and TLC. Assisted feeding can be attempted, but only in the absence of tympany. If treated adequately, the condition usually resolves within several days.

– Complicated postoperative ileus. This life-threatening condition is usually a sequalae to complex lapa- rotomic procedures that involve gut handling. The use of a2 agonists exacerbates the problem. The proposed mechanism is the inflammatory reaction to any injury (e.g., serosa laceration, blunt trauma, or desiccation) to the intestinal wall. An important mediator in this process is nitric oxide (NO),26 resulting from the activation of inducible nitric oxide synthase (iNOS) by (among others) inflamma- tory cytokines and substance P. Nitric oxide is an inhibitory neurotransmitter in the myenteric plexus; activation of iNOS leads to paralytic ileus. The affected animals are soporose. There is no fecal output; metabolic acidosis (urinary pH < 7) and dehydration are often present. Untreated, the animals die within 24 to 48 hours. Although this condition may seem uncurable, there are still several treatment options. First, radical visceral analgesia should be applied: flunixin (5 mg/kg SC q12h),²⁴ preferably with fluid replacement therapy. Opioids should be avoided. Fluids with physiologic pH of 7.4 should be infused at 20 ml/kg q6h SC or IV (slowly). To correct metabolic acidosis, sodium bicarbonate is added to the fluid, at 1 mEq for every 0.5 of pH unit below 7 (e.g., 1 mEq for 6.5, 4 mEq for 5). Adding 3-5 mg lidocaine (without epinephrin) to the solution reduces the pain at the injection site and speeds up fluid absorption from the subcutis. The latter is due to vasodilatory properties of lidocaine.²⁷ Prokinetics are of crucial importance. Alizapride²⁸ (a relatively new dopamine agonist) is the drug of choice in cases of severe impairment of gut motility. Its efficacy in guinea pigs is higher than that of metoclopramide. The author (ES) has successfully used it in severe cases of paralytic ileus at 10-25 mg/kg SC q4-6h. Alizapride can be combined with cisapride (1 mg/kg PO q4-6h).

– Another novel treatment option is based on the inhibition of iNOS.²⁹ Aminoguanidine is a selective iNOS inhibitor; it can be administered preoperatively in anticipation of complications or postoperatively (20-40 mg/kg PO q12h). The combination of radical analgesia, prokinetics, and iNOS inhibition reduces mortality by 50-70%. Antibiotics should be given to prevent secondary dysbiosis.

– Dysbiosis. Perioperative immunosuppression increases the risk of infections. The intestinal flora of clinically healthy guinea pigs contains small numbers of (facultative) pathogens: coliforms, clostridia, Enterococcus spp., yeasts and protozoa. Shifts in gut microbiota composition can occur up to 72 h after surgery. The animals develop paralytic ileus (see above) or diarrhea and deteriorate quickly. Feces can be examined under a microscope for the presence of yeasts and protozoa. There is usually no time for bacterial culture and antibiogram. Therefore, broad-spectrum antibiotics (with activity against anaer- obes) should be administered. For example, metronidazole (50 mg/kg q12h) is administered together with enrofloxacin (10 mg/kg q12h). If necessary, antifungal and antiprotozoal medications should be given. Adequate visceral analgesia and fluid replacement therapy are essential. The risk of postoperative dysbiosis increases with the increasing age of the patient. For this reason, older animals (4 years and more) should not undergo elective surgeries.

– Wound dehiscence and infection. This is a rare occurrence in guinea pigs: when the rules of good surgi- cal practice are observed, there is practically no risk. Wound dehiscence due to auto-mutilation occurs mainly after the administration of ketamine. This NMDA antagonist causes neurologic disturbances³⁰ (e.g., agitation, hallucinations), so animals are more likely to gnaw on the sutures. Besides, if no local anesthetics have been applied, the animals that recover from general anesthesia face hyperalgesia. Too tight sutures resulting in disturbed blood supply and tissue hypoxia may also contribute to auto-mutilation. If the wounds are not critical (e.g., cutaneous), topical antimicrobial preparations and local anesthet- ics can be applied, and the wounds can be left to heal by secondary intention. Deep wounds with large amounts of necrotic material must be debrided surgically.

Conclusions

Anesthesia and surgery of guinea pigs can be safely carried out if specific rules are observed. The key to success is meticulous prescreening of the patient, followed by rigorous analgesia and minimalization of tissue trauma. Most complications can be entirely prevented, and even in emergencies, life-saving solutions are available.

References

1. Lennox AM. Mitigating risk factors for anesthetic death in exotic animals: what can we learn from canine and feline standards? Proc ExoticsCon. 2022;259-266.
2. Lachmann G, Von Haefen C, Kurth J, et al. Innate immunity recovers earlier than acquired immunity during severe postoperative immunosuppression. Int J Med Sci. 2018;15:1-9.
3. Portier K, Ida KK. The ASA Physical Status Classification: what is the evidence for recommending its use in veterinary anesthesia? A systematic review. Front Vet Sci. 2018;5:204.
4. Stoffels E. Quantitative stress assessment in guinea pigs. Proc ExoticsCon. 2023; In press.
5. Stoffels E. Lower urinary tract disease in guinea pigs as it relates to nutrition. Proc ExoticsCon. 2022; 267-283.
6. Neuroscience Online: An Open-Access Neuroscience Electronic Textbook. University of Texas. Available at: https://nba.uth.tmc.edu/neuroscience/. Accessed March 23, 2023.
7. National Centre for the Replacement, Refinement and Reduction of Animals in Research. Available at: https://www.nc3rs.org.uk/3rs-resources/grimace-scales. Accessed March 23, 2023.
8. Sinclair MD. A review of the physiological effects of alpha2-agonists related to the clinical use of medeto- midine in small animal practice. Can Vet J. 2003;44:885-897.
9. Nemeth M, Pschernig E, Wallner B, et al. Non-invasive cortisol measurements as indicators of physiologi- cal stress responses in guinea pigs. Peer J. 2016;4:e1590.
10. Lundberg J, Weitzberg E, Gladwin M. The nitrate-nitrite-nitric oxide pathway in physiology and therapeu- tics. Nat Rev Drug Discov. 2008;7:156-167.
11. Flecknell PA. Laboratory Animal Anesthesia. London, UK: Academic Press; 2023.
12. Guedel AE. Inhalation Anesthesia: A Fundamental Guide. New York, NY: The Macmillan Co; 1937:14.
13. 13. Dahl JB, Steen Møiniche S. Pre-emptive analgesia. Br Med Bull. 2005;71:13-27.
14. Sohn HM, Na HS, Lim D, et al. Immediate retrograde amnesia induced by midazolam: a prospective, non- randomized cohort study. Int J Clin Pract. 2021;75:e14402.
15. Adner M, Canning BJ, Meurs H, et al. Back to the future: re-establishing guinea pig in vivo asthma models. Clin Sci. 2020;134:1219-1242.
16. Nambiar MP, Gordon RK, Moran TS, et al. A simple method for accurate endotracheal placement of an intubation tube in guinea pigs to assess lung injury following chemical exposure. Toxicol Mech Methods. 2007;17:385-92.
17. Merentie M, Lipponen JA, Hedman M, et al. Mouse ECG findings in aging, with conduction system af- fecting drugs and in cardiac pathologies: development and validation of ECG analysis algorithm in mice. Physiol Rep. 2015;3:e12639.
18. Malamed SF. Local Anesthesia. St. Louis, MO: Elsevier; 2019.
19. Jayawardena CK, Takano Y. Nerve-epithelium association in the periodontal ligament of guinea pig teeth. Arch Oral Biol. 2006;51:587-595.
20. Sansone RA, Sansone LA. Tramadol: seizures, serotonin syndrome, and co-administered antidepressants. Psychiatry. 2009;6:17-21.
21. Ventura-Martinez R, Mares-Sánchez JJ, Avilés-Herrera J, et al. Antinociceptive synergy between metamizole
    and hesperidin in a model of visceral pain in mice. Arch Med Res. 2021;52:389-396.
22. Hunt GB. BSAVA Manual of Canine and Feline Surgical Principles. Gloucester, UK: BSAVA Publications;
    2012;21:264-276.
23. Emmer KM, Celeste NA, Bidot WA, et al. Evaluation of the sterility of Press’n Seal cling film for use in
    rodent surgery. J Am Assoc Lab Anim Sci. 2019;58:235-239.
24.  Lemonnier LC, Thorin C, Meurice A, et al. Comparison of flunixin meglumine, meloxicam and ketoprofen on mild visceral post-operative pain in horses. Animals. 2022;12:526.
25. Nagai M, Toda H, Iriki M. Hypothermia enhances acetylcholine-induced contraction of isolated rat ileum. Jpn J Physiol. 1987;37:545-549.
26. Moojen TM, Van Gulik TM, Hoek FJ, et al. Possible role of nitric oxide in postoperative ileus: a comparative study. Neurogastroenterol Motil. 1999;11:403-408.
27. Newton DJ, McLeod GA, Khan F, et al. Mechanisms influencing the vasoactive effects of lidocaine in human skin. Anaesthesia. 2007;62:146-150.
28. DrugBank Online. Alizapride. Available at: https://go.drugbank.com/drugs/DB01425. Accessed June 1, 2023.
29. Yanagida H, Sanders KM, Ward SM. Inactivation of inducible nitric oxide synthase protects intestinal pacemaker cells from postoperative damage. J Physiol. 2007;582:755-765.
30. Wu JL, Haberman RP, Gallagher M, et al. Probing for conditioned hallucinations through neural activation in a ketamine mouse model of schizophrenia. Neurosci Bull. 2020;36:937-941.