Anesthesia and Perioperative Management of a Pneumonectomized Dog

Introduction

Pneumonectomy is the excision of all of the left- or right-sided lung lobes. Regardless of the indication, pneumonectomy is a major surgical intervention that poses a great challenge to the cardiopulmonary function of the patient. It has been shown that left- or right-sided pneumonectomy is tolerated by normal experimental dogs.^1,2 Although respiratory, cardiac, and gastrointestinal complications can occur, pneumonectomy, whether left- or right-sided, can also be performed in the clinical setting in dogs and cats.^3,4

In dogs, following pneumonectomy, increased pulmonary blood flow and distension in the remaining lung are considered responsible for the occurrence of compensatory changes, which result in an improvement in oxygenation. Initially, residual lung volume, vital capacity, and maximal breathing capacity are reduced to approximately 50% of preoperative levels. However, residual lung volume increases after 3 mo, and total lung capacity increases up to 37% more than expected for a normal single lung.^1,3

Compensatory improvement of oxygen transport after pneumonectomy exceeding 50% of the lung volume occurs as a result of three principal mechanisms: recruitment of physiologic reserves of diffusing capacity, remodeling of the existing alveolar-capillary network, and new or regenerative alveolar-capillary growth.^3,5 The anatomic sources of recruitment seem to include the following: (1) unfolding and/or distension of the alveolar-capillary membrane; (2) opening of new capillaries; (3) more efficient matching of diffusing capacity with respect to pulmonary blood flow or ventilation; and (4) elevation of the hematocrit.²

The secondary changes, which occur in the remaining lung within months after pneumonectomy, seem to effectively compensate for the lost pulmonary function. However, the “price that has to be paid” for the improvement in oxygenation is that decreased compliance and vital capacity together with increased pulmonary vascular resistance and residual lung capacity can potentially result in right ventricular hypertrophy.^3,5,6 Such alterations make the anesthetic management of a pneumonectomized animal particularly challenging. To the best of the authors’ knowledge, there are no reports in the veterinary literature describing the anesthetic management of pneumonectomized animals requiring surgery for reasons either related or unrelated to the respiratory system. The aim of this case report is to describe the anesthetic and perioperative management of a dog with a history of left pneumonectomy requiring multiple surgical interventions.

Case Report

A 7 yr old intact female Hellenic shepherd dog weighing 46.8 kg presented for a mass protruding through the vulva. The animal had a history of left-sided pneumonectomy associated with Aspergillus fumigatus pneumonia approximately 3 yr before presentation with a grade 2 vaginal wall prolapse. Surgical resection of the protruding vaginal wall along with ovariectomy and prophylactic gastropexy were performed.

The preanesthetic clinical, hematologic (complete blood cell count), and serum biochemical examinations did not reveal any abnormalities except decreased breath sounds over the left hemithorax. An arterial blood sample was obtained through a dorsal pedal artery catheter. Arterial blood gas analysis^a before anesthesia showed a pH of 7.36, a partial pressure of CO₂ (PaCO₂) of 41.8 mm Hg, a partial pressure of O₂ (PaO₂) of 89.1 mm Hg, and a bicarbonate concentration (HCO₃⁻) of 23.4 mmol/L. The calculated alveolar-arterial difference in partial pressure of O₂ (P_AO₂-PaO₂) was 8.38 mm Hg with a fraction of inspired O₂ of 0.21 (reference range, <10 mm Hg).⁷ The dog was characterized as an American Society of Anesthesiologists status 3 case, and preoxygenation was performed for approximately 7 min with 100% O₂ delivered through a face mask. After establishing IV access, induction of anesthesia was achieved by the rapid administration of 0.1 mg/kg midazolam^b IV, 5 μg/kg fentanyl^c IV, and 0.5 mg/kg propofol^d IV. Throughout surgery and for the first 2 hr postoperatively, a constant rate of infusion (CRI) of fentanyl in normal saline was administered. The rate administration of the normal saline was 10 mL/kg/hr, which corresponded to a dose rate of 0.1 μg/kg/min for fentanyl. Intubation of the trachea was performed with a 14 mm cuffed endotracheal tube and isoflurane^e in oxygen (1.5 L/min) was delivered through a rebreathing circle system. The dial of the vaporizer was originally set at 2.5%, but within a few minutes it was adjusted to provide the appropriate depth of surgical anesthesia based on clinical assessment (i.e., evaluation of reflexes and muscle tone) and monitoring of end-tidal isoflurane concentrations and cardiovascular and respiratory parameters. In particular, a standard monitoring protocol was applied from within a few minutes after induction until discontinuation of anesthesia. This included electrocardiography (lead II), pulse oximetry, direct arterial blood pressure measurement^f, capnography, measurement of O₂ and isoflurane concentrations in inspired and end-tidal gas, tidal volume, and airway pressure^g. The gas analyzer was calibrated according to the manufacturer's directions. Intraoperatively, heating pads were used to maintain normothermia. Carprofen^h (4 mg/kg) and cefuroximⁱ (20 mg/kg) were administered IV shortly after tracheal intubation.

A transient apnea of short duration was noted shortly after commencement of isoflurane administration, and IPPV was applied for a period of a few minutes, taking care not to deliver tidal volumes >10 mL/kg. A second arterial blood sample was obtained from the dorsal pedal artery at this point, and an epidural injection was performed in the lumbosacral junction using 2% lidocaine^j (1.7 mg/kg) and 0.5% bupivacaine^k (0.42 mg/kg). The results of this second arterial blood gas analysis revealed respiratory acidosis. The pH was 7.29, PaCO₂ was 51.1 mm Hg, PaO₂ was 566 mm Hg, HCO₃⁻ was 23.9 mmol/L, P_AO₂-PaO₂ was 68.87 with a fraction of inspired O₂ of 0.98 (reference range, <100 mm Hg)⁷. Volume-controlled IPPV was started using a Siesta i TS^l ventilator when the dog was moved to the operating room after surgical preparation. The ventilator was set at a rate of 25 breaths/min, a tidal volume of 300 mL (6.4 mL/kg), and a positive end-expiratory pressure (PEEP) of 4 cm H₂O (corresponding values for the spontaneously breathing animal were approximately 18 breaths/min and a tidal volume of 230 mL [4.9 mL/kg]). A few minutes later, the end-tidal partial pressure of CO₂ (P_ECO₂), determined via capnography, was 40–45 mm Hg. This was considered satisfactory and no further adjustments in the ventilator settings were made. No signs of the dog spontaneously breathing against the ventilator were observed.

During resection of the vaginal wall in sternal recumbency, and despite a reduction in the percentage of inspired isoflurane so that the end-tidal isoflurane concentration was 0.8–0.9%, approximately 20 min after the epidural injection, hypotension was observed (mean arterial pressure was <60 mm Hg) with occasional readings of mean arterial pressure of 50–55 mmHg. Treatment consisted of administration of lactated Ringer's solution (LRS) at a rate of 10 mL/kg/hr (in addition to the 10 mL/kg/hr of normal saline used as carrier fluid for the fentanyl CRI). This effectively increased mean arterial pressure to 60–70 mm Hg within approximately 10 min. Overall, 250 mL LRS was infused for 30 min.

During aseptic preparation for the midline celiotomy with the dog in dorsal recumbency, the ET_ISO was gradually increased to 1.4%. Nevertheless, increases in heart rate (from 60 beats/min during resection of the vaginal wall to approximately 90 beats/min) and mean arterial blood pressure (from approximately 65 mm Hg to 90–110 mm Hg) were observed upon intra-abdominal manipulations. The ET_ISO was further increased to 1.6–1.8%, resulting in blunting of the autonomic responses within a few minutes. The remaining period of surgery (during the gastropexy) was uneventful, and 10 min before the estimated time of end of surgery, the ventilation rate was decreased to 8 breaths/min for 2–3 min, then ventilation was discontinued. Spontaneous breathing was observed (respiratory rate was 15 breaths/min, P_ECO₂ was 48 mm Hg, and percentage of saturated hemoglobin obtained from pulse oximetry was 98%) and was deemed adequate. The administration of isoflurane was discontinued upon completion of surgery and after a central IV line (via jugular puncture) and a nasal catheter were placed. The total duration of surgery was 110 min, and the duration of anesthesia was 160 min.

The dog was transferred to the intensive care unit for recovery. The fentanyl CRI was continued for the first 2 hr after surgery. Another arterial blood sample was obtained approximately 10–15 min after discontinuation of anesthesia and before administration of supplemental oxygen was started via the nasal catheter (at a rate of 1 L/min). The pH was 7.28, PaCO₂ was 50.3 mm Hg, PaO₂ was 93 mm Hg, and HCO₃⁻ was 23.8 mmol/L. The dog recovered relatively quickly (i.e., head support 15 min after discontinuation of isoflurane, attempts to ambulate after 95 min) and very smoothly while under the influence of the fentanyl CRI. To extend postoperative analgesia, two doses of intramuscular morphine^m (0.2 mg/kg) were administered at 2 hr and 7 hr postoperatively. Another dose of cefuroxim (20 mg/kg) was also administered postoperatively IV . Central venous pressure (CVP), measured 90 min postoperatively, was −2 cm H₂O; thus, LRS was administered at a rate of 500 mL/hr. The CVP increased to 2 cm H₂O within 2 hr, and the rate of LRS administration was decreased to a maintenance rate (50 mL/kg/24 hr). Ten hours after the end of surgery, the CVP was 6 cm H₂O. At 18 hr after surgery, urine production was adequate, and the dog was discharged from the intensive care unit. On re-evaluation through telephone communication 3 mo postsurgically, the owner reported that the dog did not exhibit any signs of cardiorespiratory compromise.

Discussion

The physiologic consequences and complications of pneumonectomy in dogs have been described.^3,4,6 However, there are no reports in the literature concerning the potential complications and problems that veterinarians might face when attempting to anesthetize an animal that has previously undergone a total left- or right-sided lung resection and requires anesthesia for a different nonrespiratory surgical procedure. The most frequently reported complications of pneumonectomy pertain to the respiratory and cardiovascular system. Gastrointestinal complications, such as esophageal dysfunction due to mediastinal shift, can also occur.^3,4,8

As part of the preanesthetic evaluation of the case reported here, arterial blood gases were measured. This examination revealed normal values for both PaCO₂ and PaO₂. The calculated P_AO₂-PaO₂ was also within normal limits, indicating efficient gas exchange. Nevertheless, preoxygenation with 100% O₂ via a face-mask was performed before induction of anesthesia to prevent potential hypoxemia after induction. Crash-induction of anesthesia with quick acquisition of control of the airway was deemed appropriate and was carried out using a combination of a benzodiazepine, a potent opioid, and a general anesthetic. This reduced the required doses and consequently the potential side effects of each of the drugs. Anesthesia was maintained with isoflurane in oxygen. Because volatile anesthetics have proven to be potent bronchodilators, the use of isoflurane probably contributed to improved oxygenation due to bronchodilation. Transient apnea after induction of anesthesia was treated with manual ventilation for a few minutes using tidal volumes that would be considered small for a dog with two lungs. When blood gas analysis of a second arterial sample revealed respiratory acidosis, it was decided that hypercapnia should not be allowed, and IPPV using a ventilator was initiated. Oxygen exchange at this point was not problematic (P_AO₂-PaO₂ was within normal limits). Respiratory acidosis is not unusual in dogs following pneumonectomy, and the CRI of fentanyl in this case may have contributed to respiratory depression, exacerbating the elevation of PaCO₂.⁵

In humans with two lungs that are undergoing one lung ventilation, it has been suggested that tidal volumes similar to those during two lung ventilation should be used. A tidal volume of 10–12 mL/kg at a rate to maintain a PaCO₂ of 35±3 mm Hg has been recommended. Tidal volumes <10 mL/kg are generally discouraged because of concerns about dependent lung atelectasis, but there are no clear guidelines established for tidal volume during one-lung ventilation.⁹ In contrast, the case described here differs from situations of one lung ventilation in normal subjects because in this case the animal had already had a pneumonectomy resulting in all blood flow being directed to the existing lung. Moreover, there was a concern that “high” tidal volumes (i.e., 10–20 mL/kg as would be used in an animal with two lungs) might cause alveolar trauma in an animal with decreased lung compliance because less compliant lungs are at greater risk for rupturing when high volumes and pressures are used during IPPV.¹⁰ The tidal volume and respiratory rate of the spontaneously breathing animal was taken as a guide. A tidal volume of approximately 6.5 mL/kg, which was mildly elevated compared with the tidal volumes that were observed during spontaneous breathing (approximately 5 mL/kg) and a rate increased by approximately 40% compared with the rate of spontaneous breathing, were chosen. This ventilation resulted in a maximum airway pressure of 12–20 cm H₂O and a P_ECO₂ of 40–45 mm Hg, which are in the upper normal range. No attempts to further increase ventilation were made.

The remaining lung after pneumonectomy is more sensitive to PEEP, as manifested by a greater increase in the pulmonary vascular resistance and a lower cardiac output at a given level of PEEP.⁵ It has been established that PEEP can provoke cardiovascular effects and, specifically, a reduction in cardiac output. The mechanisms for this alteration may be a reduction in venous return because of the increase in mean intrathoracic pressure, an increase in right ventricular afterload, a decrease in right and left ventricular preload, and a change in ventricular function.¹¹ However, in patients with normal lungs, the effects of PEEP on pulmonary function are mostly beneficial. This includes improved V/Q matching and hence improved oxygenation either via opening or “recruitment” of collapsed alveoli or prevention of collapse of unstable alveoli.¹² Pulmonary vascular resistance is increased after the application of PEEP. Values in excess of 5 cm H₂O can increase pulmonary vascular resistance in the ventilated lung. However, it has been suggested that PEEP during one lung ventilation in a dog with a closed thoracic cavity probably provides sufficient space for the ventilated lung to expand, which prevents most of the negative effects of PEEP on hemodynamic variables.¹¹ This may be the case for a pneumonectomized animal also. Thus, the use of PEEP during one lung ventilation in dogs with a closed thoracic cavity has been recommended because cardiac output is not affected and any gain in arterial O₂ will be beneficial for O₂ delivery to tissues in critically ill patients.¹¹ Taking these into account, in addition to the fact that application of PEEP at <5 cm H₂O is unlikely to increase pulmonary vascular resistance, it was decided to apply a PEEP of 4 cm H₂O. Intensive hemodynamic monitoring was also applied to detect detrimental effects of PEEP on cardiovascular performance.

During resection of the protruding vaginal wall, hypotension was noted, which probably reflected the combined effects of: positive pressure ventilation, which increased intrathoracic pressure thus hindering venous return to the heart; and epidural anesthesia, which may have resulted in sympathetic blockade and vasodilation in the hind end of the animal. Treatment included increasing the rate of isotonic crystalloid administration from 10 mL/kg/hr to 20 mL/kg/hr for a short period of time. This resulted in the desired increase of mean arterial pressure to levels >60 mm Hg. More aggressive fluid administration or the use of α₁-agonists could have been more effective in increasing systemic arterial pressure, but could also have resulted in increased pulmonary arterial pressure. The potential for pulmonary hypertension due to pneumonectomy dictated a less aggressive treatment of systemic hypotension using moderate infusion rates of fluids. Moreover, the possibility that arterial pressure might increase when the celiotomy involving the cranial abdomen (an area usually not affected by epidural anesthesia at the usual doses) would be performed, right after the vaginal wall resection, was taken into consideration. The fact that arterial blood pressure indeed increased further justified the conservative use of fluids and withholding of inotropes and vasoactive drugs. The risk of pulmonary hypertension aggravating cardiac function was also the reason why a conservative rate of fluid administration was chosen for the management of decreased CVP postoperatively in the intensive care unit.⁴

Conclusion

Successful perioperative management of a left-sided pneumonectomized animal is reported. Intensive hemodynamic and respiratory monitoring and appropriate response and treatment of any detected abnormalities, taking into consideration the pathophysiologic alterations occurring in such an animal, are required for a successful outcome.