One of the main challenges during intrathoracic procedures is to obtain adequate visibility within the thoracic cavity concomitantly with lung movement. In 1949, Carlens developed the OLV technique, which involves the insertion of a DLT into one of the main bronchi, thus separating each hemithorax for different ventilation protocols.1
Various techniques are currently used to successfully implement OLV. Two frequently used devices are bronchial blockers and the DLT. Bronchial blockers have been used for OLV in dogs.2–7 Bronchial blockers are thin balloon-tipped catheters that are inserted into a main bronchus to allow obstruction of a hemithorax.1 Compared with most types of bronchial blockers, a DLT has the advantage in that it can be used to allow an influx of gas to or ventilation of both lungs. However, there are only a few reports8–10 that describe successful intubation of dogs with a DLT.
For procedures that require OLV, a primary concern for the anesthetist is to develop an adequate anesthetic protocol that has minimal effects on pulmonary perfusion because most commonly used volatile anesthetic agents inhibit HPV, which is responsible for the deviation of blood flow to properly ventilated lung tissue.11–13 Some anesthetists continue to use inhalation anesthesia during OLV because volatile anesthetic agents cause a decrease in CO, which, although generally considered an adverse effect, counteracts the HPV inhibition induced by those agents in a dose-dependent manner. Thus, the net effect is a decrease in blood flow to the nonventilated lung.11,13,14 Other anesthetists advocate the use of propofol for OLV.8,15–17 In 1 study,17 pigs in which anesthesia was maintained with propofol during OLV had better oxygenation and a decreased Qs/Qt, compared with pigs in which anesthesia was maintained with desflurane.
Most studies11,15,16,18–20 conducted to evaluate anesthetic protocols for OLV have involved human patients. The aim of the study reported here was to assess pulmonary hemodynamics and alveolar oxygenation in dogs anesthetized with propofol or isoflurane during OLV in a closed-thoracic experimental model. Our hypothesis was that, when anesthetized with isoflurane, dogs would have impaired arterial oxygenation and an increased Qs/Qt relative to when they were anesthetized with propofol.
Materials and Methods
Animals
All study procedures were approved by the São Paulo State University (UNESP) Animal Usage Ethics Committee, (protocol No. 00118/2013). Six Beagles (2 males and 4 females) from a research colony with a mean ± SD age of 3.6 ± 0.2 years and body weight of 11.5 ± 2.4 kg were used for the study. All dogs were considered healthy on the basis of results of a physical examination, CBC, and measurement of serum creatinine concentration and alanine aminotransferase activity and tested negative for canine leishmaniasis (the study was conducted in an area where leishmaniasis was endemic).
Study design
The study had a crossover design. Each dog underwent both TLV and OLV by use of a DLT while anesthetized with each of 3 anesthetic protocols (a CRI of propofol [propofol], isoflurane administered at the MAC [IsoMAC], and isoflurane administered at 1.5 MAC [Iso1.5MAC]). There was a washout period of at least 7 days between anesthetic sessions.
Study protocol
Prior to initiation of the live-animal experiments, the anesthetist responsible for positioning the DLT practiced the technique on recently euthanized canine cadavers that weighed between 8 and 20 kg. All dogs were euthanized for reasons unrelated to the study. The training comprised placing the DLT in the left main bronchus followed by careful auscultation of each hemithorax to confirm correct ventilation of the lungs. The thoracic cavity was then incised for direct visualization of the lungs during ventilation. The intubation procedure was continued until the DLT was successfully placed in the left main bronchus of 10 consecutive cadavers. The training required a total of 30 cadavers.
For all dogs, food was withheld for 12 hours and water was withheld for 2 hours prior to each anesthetic session. At least 15 days prior to initiation of the study, the MAC of isoflurane was determined for each dog by use of electrical stimulationa as described.21 The order in which each dog received the 3 anesthetic protocols was determined by use of a random number generator.
Anesthesia was induced by the administration of an IV bolus of propofol (concentration, 10 mg/mL) titrated to effect for the propofol protocol and 5% isoflurane in oxygen (flow rate, 3 L/min) via a face mask and partial rebreathing circle anesthesia systemb for the IsoMAC and Iso1.5MAC protocols until intubation could be achieved. Female dogs (mean ± SD body weight, 10.2 ± 1.4 kg) were intubated with a size 26 left-bronchus DLT,c and male dogs (mean ± SD body weight, 14.2 ± 0.7 kg) were intubated with a size 28 left-bronchus DLT.c A carinal hook was not used for intubation. Correct positioning of the DLT was confirmed by careful auscultation of each hemithorax as described.9
Following confirmation that the DLT was in the correct position, dogs were positioned in right lateral recumbency to mimic the positioning used for a left thoracotomy, a commonly performed procedure in dogs.22 For all anesthetic sessions, dogs were administered 100% oxygen (flow rate, 2 L/min) delivered through a partial rebreathing circle anesthesia systemb and a CRI of lactated Ringer solution (3 mL/kg/h) delivered through a 20-gauge, 32-mm-long catheter that was aseptically placed in the left cephalic vein. Dogs were also instrumented and connected to an electronic multivariable patient monitord that provided continuous or semicontinuous measurement of ECG, HR, oxygen saturation as determined by pulse oximetry, indirect blood pressure, and expired gas concentrations. The capnograph portion of the monitor was calibrated before each anesthetic session by use of a standard gas sample.e A 22-gauge, 25-mm-long catheterf was aseptically placed in the right dorsal pedal artery and connected to the patient monitord for invasive blood pressure measurement. A 75-cm, 5F Swan-Ganz catheterg with an 8-mm balloon (capacity, 0.7 mL) was aseptically inserted into the pulmonary artery via the left jugular vein for measurement of core body temperature and CO by a thermodilution method. The catheter was connected to a CO module,h and correct positioning of the catheter in the pulmonary artery was confirmed by observation of the distinct waveforms of the right atrium, right ventricle, and pulmonary artery on the module screen. The module's transducers were positioned at the level of the heart and zeroed to room air. Following placement, all catheters were flushed with heparinized saline (0.9% NaCl) solution (5,000 U of heparin/mL diluted 1:1,000 with saline solution). Core body temperature (as measured by the Swan-Ganz catheter) was maintained at approximately 36.5°C with the aid of a warming blanketi throughout each anesthetic session.
For the propofol protocol, a 20-gauge, 32-mm-long catheter was aseptically placed in the right cephalic vein, and anesthesia was maintained with a CRI of propofol (beginning at 0.4 mg/kg/min) administered with a syringe pumpj that was connected to the catheter. For each dog, the propofol infusion rate was adjusted as necessary throughout the instrumentation period on the basis of the presence of protective responses, reflexes, or movement during placement and confirmation of correct placement of the DLT tube and peripheral artery and Swan-Ganz catheters. When such responses were detected, the propofol infusion rate was increased by 0.1 mg/kg/min and maintained for 10 minutes. This procedure was repeated until no protective responses were detected, and that infusion rate was maintained for the duration of the propofol protocol.
For the IsoMAC and Iso1.5MAC protocols, anesthesia was maintained by the administration of isoflurane in oxygen through the DLT and partial rebreathing anesthesia system.b The amount of isoflurane delivered to each dog was adjusted as necessary until the expired end-tidal isoflurane concentration was equal to the MAC or 1.5 MAC for the IsoMAC and Iso1.5MAC protocols, respectively.
TLV and OLV
For all 3 protocols, the observation period did not begin until 15 minutes after instrumentation was completed to allow for physiologic variables to stabilize. The observation period lasted 2 hours and was comprised of 3 ventilation modes (30 minutes of TLV [T30], 60 minutes of OLV of the right lung divided into 20-minute intervals [T50, T70, and T90], and another 30 minutes of TLV [T120]). It was developed on the basis of a combination of methods used in other studies2,7,23 that initiated mechanical ventilation with TLV before OLV and another study16 in which TLV was repeated after OLV.
Pressure-controlled mechanical ventilationk was used during both TLV and OLV. Each dog was administered 0.2 mg of atracurium/kg, IV, immediately before initiation of the first 30 minutes of TLV and 0.1 mg of atracurium/kg, IV, every 30 minutes thereafter or at any time signs of muscular activity, such as attempts to breathe spontaneously against the mechanical ventilator, were observed throughout the observation period. During mechanical ventilation, peak inspiratory pressure was maintained at 15 cm H2O and the inspiratory-expiratory ratio was maintained at 1:2. The PEEP was maintained at 5 cm H2O via a PEEP valve on the ventilator in accordance with the protocol used in other studies6,8 involving dogs.
During TLV, a Y connectorl was attached to both extremities of the DLT, thereby joining those 2 extremities into 1 end. During OLV, a flowmeter, manometer, and 3-way stopcock were attached to the anesthesia machine to regulate oxygen flow to the nonventilated (left) lung, in which a CPAP of 4 to 6 cm H2O was maintained on the basis of reports8,15,24 of OLV in human patients. The 3-way stopcock was kept in a semi-open position to allow excess pressure to escape the system until the manometer indicated that the pressure was stable.
Mechanical ventilation was maintained after the observation period had ended so that the thorax could be carefully auscultated to check for any areas of silence. Auscultation during DLT positioning (baseline) was used for comparison with auscultation after the observation period. When loci without any lung sounds were identified, alveolar recruitment maneuvers were performed. Briefly, a peak inspiratory pressure of 30 cm H2O was maintained for 3 seconds and then repeated every 5 to 10 minutes until lung sounds were audible over all areas of the thorax and comparable to those auscultated at baseline. Dogs were monitored and auscultated by 2 anesthetists (BPF and TAT) until they were able to breathe spontaneously and maintain a Petco2 between 35 and 45 mm Hg, after which mechanical ventilation was discontinued, and the dogs were observed for at least 1 more hour to ensure that they had completely recovered from anesthesia.
Data collection
At each of the 5 specified times during the observation period (T30, T50, T70, T90, and T120), a 1-mL sample of arterial and 1-mL sample of mixed venous blood were collected for blood gas analysis.m The following variables were obtained at each time: arterial pH, HR, MAP, MPAP, PAOP, CO, Pao2, and Paco2. The following calculations were also performed and recorded: Pao2 – Pao2; Paco2 – Petco2; SV = CO/HR; PVR = ([MPAP – PAOP]/CO) × 79.9; Qs/Qt = (Cco2 × arterial oxygen concentration)/(Cco2 × oxygen concentration in mixed venous blood), where Cco2 = (1.34 × hemoglobin concentration) + (Pao2 × 0.0031)18; arterial oxygen concentration = (1.34 × hemoglobin concentration × [arterial oxygen saturation/100]) + (Pao2 × 0.0031)25; oxygen concentration in mixed venous blood = (1.34 × hemoglobin concentration × [oxygen saturation in mixed venous blood/100]) + (partial pressure of oxygen in mixed venous blood × 0.0031)25; and Pao2 = ([barometric pressure – 47] × fraction of inspired O2) – (Paco2 × 1.1), where barometric pressure is equivalent to 760 mm Hg, 47 is the water vapor pressure in mm Hg, and 1.1 is the inverse respiratory quotient (1/0.9). For each dog, the hemoglobin concentration obtained during the CBC that was performed prior to study enrollment was used for all calculations.
Statistical analysis
The distributions for all variables were assessed for normality with the Shapiro-Wilk test. Variables were compared among the 3 anesthetic protocols and over time within each protocol by use of repeated-measures ANOVA with a Tukey adjustment for multiple pairwise comparisons when necessary. Values of P < 0.05 were considered significant for all comparisons.
Results
The mean ± SD hemoglobin concentration for all dogs prior to study initiation was 15.9 ± 1.0 g/dL. For the propofol protocol, the mean ± SD induction dose and infusion rate of the propofol were 5.0 ± 0.6 mg/kg and 0.6 ± 0.2 mg/kg/min, respectively. For the IsoMAC and Iso1.5MAC protocols combined, the mean ± SD time required for anesthesia induction (ie, administration of isoflurane via face mask) was 7.8 ± 1.7 minutes. One dog required alveolar recruitment maneuvers for 1 hour after the observation period for the Iso1.5MAC protocol had ended. Adequate ventilation was successfully reestablished for that dog, and it went on to recover from anesthesia for that protocol without any other problems. It also recovered from anesthesia without any problems when administered the other 2 (IsoMAC and propofol) protocols. Recovery from all anesthetic sessions was uneventful for the other dogs.
Variables assessed were tabulated for each dog (Supplementary Table S1, available at avmajournals.avma.org/doi/suppl/10.2460/ajvr.78.10.117) and summarized for each protocol over time (Table 1). Mean HR, PAOP, PVR, and Paco2 – Petco2 did not differ among the 3 protocols at any time (T30, T50, T70, T90, and T120) assessed. The mean MAP for the Iso1.5MAC protocol was significantly less than that for the propofol protocol at all 5 times. The mean CO for the Iso1.5MAC protocol was significantly less than that for the IsoMAC protocol at T50, T90, and T120, whereas the mean SV for Iso1.5MAC protocol was significantly less than that for the IsoMAC protocol only at T50. Although the mean MPAP did not differ significantly among the 3 protocols at any time, it increased within each protocol over time. Within the Iso1.5MAC protocol, the mean Qs/Qt at T50 was significantly greater than that at T120; however, the mean Qs/Qt did not differ significantly over time for the other 2 protocols or among the 3 protocols at any time. For all 3 protocols, the mean arterial pH was significantly lower, whereas the mean Paco2 was significantly greater during OLV (T50, T70, and T90), compared with that during initial TLV (T30). The mean Pao2 did not differ significantly among the 3 protocols at any time but decreased significantly within each protocol when TLV was switched to OLV, and that significant decrease persisted for the duration of OLV. For the Iso1.5MAC protocol, the mean Pao2 – Pao2 at T70 was significantly greater than that at T120 (Figure 1).
Mean ± SD Pao2 – Pao2 for 6 healthy adult Beagles that underwent TLV and OLV in a closed-thoracic experimental model while anesthetized with each of 3 protocols (CRI of propofol [mean ± SD, 0.6 ± 0.23 mg/kg/min; black bars] and isoflurane at the MAC [striped bars] and 1.5 MAC [white bars]). For each dog, there was at least a 7-day interval between anesthetic sessions, and the MAC of isoflurane was determined by use of electrical stimulation as described.21 For each protocol, the observation period began 15 minutes after instrumentation of the dog was completed to allow physiologic variables to stabilize. The observation period lasted 2 hours and was comprised of 3 ventilation modes (30 minutes of TLV [T30], 60 minutes of OLV of the right lung divided into 20-minute intervals [T50, T70, and T90], and another 30 minutes of TLV [T120]).
Citation: American Journal of Veterinary Research 78, 10; 10.2460/ajvr.78.10.1117
Mean ± SD values for pulmonary hemodynamic and blood gas variables for 6 healthy adult Beagles that underwent TLV and OLV in a closed-thoracic experimental model while anesthetized with each of 3 protocols (CRI of propofol [mean ± SD, 0.6 ± 0.23 mg/kg/min; propofol] and isoflurane at the MAC [IsoMAC] and 1.5 MAC [Iso1.5MAC]).
| Time | ||||||
|---|---|---|---|---|---|---|
| Variable | Anesthetic protocol | T30 | T50 | T70 | T90 | T120 |
| HR (beats/min) | Propofol | 95 ± 32 | 96 ± 31 | 100 ± 33 | 108 ± 33 | 99 ± 38 |
| IsoMAC | 104 ± 27 | 120 ± 17 | 116 ± 17 | 115 ± 16 | 121 ± 12 | |
| Iso1.5MAC | 113 ± 14 | 116 ± 6 | 115 ± 4 | 116 ± 10 | 114 ± 13 | |
| MAP (mm Hg) | Propofol | 81 ± 11a | 85 ± 14a | 85 ± 10a | 90 ± 10a | 91 ± 12a |
| IsoMAC | 74 ± 6a,b | 79 ± 12a,b | 78 ± 9a,b | 75 ± 13b | 83 ± 8a,b | |
| Iso1.5MAC | 69 ± 7b | 68 ± 8b | 64 ± 8b | 66 ± 9b | 63 ± 5b | |
| SV (mL/beat/kg) | Propofol | 1.9 ± 0.1 | 2.0 ± 0.2a,b | 2.0 ± 0.3 | 1.9 ± 0.3 | 1.9 ± 0.2 |
| IsoMAC | 2.1 ± 0.5 | 2.1 ± 0.3a | 2.1 ± 0.3 | 2.3 ± 0.3 | 2.0 ± 0.6 | |
| Iso1.5MAC | 1.6 ± 0.4 | 1.7 ± 0.4b | 1.8 ± 0.5 | 1.8 ± 0.5 | 1.6 ± 0.3 | |
| CO (mL/min/kg) | Propofol | 182 ± 72 | 193 ± 64a,b | 201 ± 80 | 208 ± 74a,b | 182 ± 74a,b |
| IsoMAC | 213 ± 74 | 261 ± 68a | 244 ± 55 | 261 ± 62a | 243 ± 61a | |
| Iso1.5MAC | 183 ± 66 | 194 ± 52b | 205 ± 68 | 214 ± 87b | 185 ± 57b | |
| MPAP (mm Hg) | Propofol | 16 ± 3 | 17 ± 3 | 17 ± 3 | 18 ± 3* | 17 ± 4 |
| IsoMAC | 14 ± 4 | 16 ± 1 | 18 ± 2* | 18 ± 2* | 19 ± 3* | |
| Iso1.5MAC | 16 ± 2 | 17 ± 2 | 17 ± 2 | 18 ± 2* | 17 ± 3 | |
| PAOP (mm Hg) | Propofol | 6 ± 4 | 6 ± 4 | 7 ± 4 | 6 ± 4 | 7 ± 3 |
| IsoMAC | 7 ± 2 | 6 ± 2 | 7 ± 1 | 6 ± 2 | 6 ± 3 | |
| Iso1.5MAC | 7 ± 1 | 7 ± 2 | 7 ± 2 | 8 ± 2 | 6 ± 2 | |
| PVR (dynes•cm−5) | Propofol | 396 ± 160 | 406 ± 153 | 391 ± 175 | 420 ± 171 | 448 ± 156 |
| IsoMAC | 256 ± 108 | 307 ± 93 | 339 ± 108 | 341 ± 108 | 382 ± 114 | |
| Iso1.5MAC | 394 ± 168 | 384 ± 163 | 365 ± 163 | 348 ± 116 | 432 ± 164 | |
| Qs/Qt (%) | Propofol | 1.1 ± 0.5 | 1.6 ± 0.9 | 1.3 ± 0.9 | 1.5 ± 1.1 | 1.2 ± 1.0 |
| IsoMAC | 1.2 ± 0.7 | 1.5 ± 1.0 | 1.5 ± 1.0 | 0.9 ± 0.4 | 1.3 ± 0.7 | |
| Iso1.5MAC | 1.8 ± 0.8 | 2.3 ± 1.3 | 2.2 ± 1.0 | 2.3 ± 1.2 | 1.2 ± 0.6† | |
| Arterial pH | Propofol | 7.32 ± 0.08 | 7.25 ± 0.07* | 7.24 ± 0.06* | 7.23 ± 0.07* | 7.28 ± 0.07 |
| IsoMAC | 7.29 ± 0.08 | 7.20 ± 0.09* | 7.18 ± 0.08* | 7.25 ± 0.06* | 7.31 ± 0.05 | |
| Iso1.5MAC | 7.31 ± 0.05 | 7.24 ± 0.04* | 7.23 ± 0.08* | 7.22 ± 0.07* | 7.27 ± 0.09 | |
| Pao2 (mm Hg) | Propofol | 519 ± 72 | 472 ± 85 | 489 ± 82 | 477 ± 86 | 525 ± 74 |
| IsoMAC | 489 ± 54 | 445 ± 70 | 443 ± 76 | 477 ± 34 | 456 ± 82 | |
| Iso1.5MAC | 449 ± 103 | 392 ± 66 | 381 ± 82 | 392 ± 82 | 469 ± 86 | |
| Pao2 (mm Hg) | Propofol | 630 ± 12 | 617 ± 10* | 616 ± 9* | 615 ± 12* | 622 ± 13 |
| IsoMAC | 621 ± 15 | 601 ± 22* | 597 ± 18* | 596 ± 10* | 610 ± 14 | |
| Iso1.5MAC | 620 ± 7 | 607 ± 6 | 603 ± 13* | 597 ± 13* | 608 ± 17 | |
| Paco2 (mm Hg) | Propofol | 45 ± 7 | 53 ± 9* | 55 ± 7* | 56 ± 8* | 49 ± 7 |
| IsoMAC | 49 ± 10 | 61 ± 9* | 62 ± 7* | 64 ± 8* | 54 ± 11 | |
| Iso1.5MAC | 46 ± 3 | 55 ± 5 | 57 ± 11* | 61 ± 10* | 53 ± 13 | |
| Paco2 – Petco2 (mm Hg) | Propofol | 10 ± 3 | 13 ± 7 | 14 ± 6 | 14 ± 4 | 12 ± 2 |
| IsoMAC | 12 ± 5 | 16 ± 3 | 15 ± 3 | 15 ± 5 | 12 ± 6 | |
| Iso1.5MAC | 12 ± 2 | 13 ± 4 | 12 ± 2 | 14 ± 3 | 11 ± 8 | |
For each dog, there was at least a 7-day interval between anesthetic sessions, and the MAC of isoflurane was determined by use of electrical stimulation as described.21 For each protocol, the observation period began 15 minutes after instrumentation of the dog was completed to allow physiologic variables to stabilize. The observation period lasted 2 hours and was comprised of 3 ventilation modes (30 minutes of TLV [T30], 60 minutes of OLV of the right lung divided into 20-minute intervals [T50, T70, and T90], and another 30 minutes of TLV [T120]).
Within a variable and protocol, value differs significantly (P < 0.05) from that at T30.
Within a variable and protocol, value differs significantly (P < 0.05) from that at T50.
Within a variable and time, values with different superscript letters differ significantly (P < 0.05); the absence of superscript letters indicates that the mean did not differ significantly among the 3 protocols.
Discussion
Results of the present study indicated that propofol and isoflurane affected the pulmonary hemodynamics and alveolar oxygenation of dogs in a similar manner and are viable anesthetic agents for use in dogs undergoing procedures that require OLV. Although a few significant differences were identified over time for some pulmonary variables, those differences were likely a reflection of the transition from TLV to OLV and then back to TLV.
The propofol infusion rate used during the propofol protocol was adjusted as necessary to eliminate protective responses (eg, laryngeal and palpebral reflexes) or movement during instrumentation. It is possible that that depth of anesthesia induced by the propofol protocol was not comparable to that induced by the isoflurane protocols (IsoMAC and Iso1.5MAC). Perhaps it would have been more appropriate to determine the minimum infusion rate of propofol prior to the study to ensure that anesthetic depth was comparable among propofol and isoflurane groups. Regardless, results of the comparisons among the 3 anesthetic protocols evaluated in this study should be interpreted cautiously considering the wide use of both propofol and isoflurane in clinical practice.
In healthy dogs, the mean ± SD HR is approximately 90 ± 21 beats/min.25 The mean HR for the dogs of the present study remained within acceptable limits throughout the duration of the observation period regardless of anesthetic protocol or ventilation (TLV or OLV) method used. This indicated that the HR of dogs was not substantially altered during OLV and was consistent with results of other studies of OLV in dogs during open-3 and closed-thoracic5 experimental models and human patients undergoing thoracic procedures.19
Isoflurane causes vasodilation by decreasing intracellular calcium concentration.26 For dogs, hypotension is defined as an MAP < 60 mm Hg.25 Among the 3 anesthetic protocols evaluated in the present study, the Iso1.5MAC protocol resulted in the lowest mean MAP throughout the entire observation period; however, the mean MAP for that protocol was never < 60 mm Hg. Among individual dogs, the lowest MAP recorded during the present study was 54 mm Hg. Two dogs developed hypotension at various times during OLV when anesthetized with the Iso1.5MAC protocol, but the hypotension resolved without treatment. It is unlikely that mild and transient hypotension similar to that observed in those 2 dogs would cause substantial complications during OLV; however, it is important to remember that the dogs of this study were healthy and fairly young, and the same may not be true for dogs undergoing a surgical procedure that requires OLV because those dogs likely have cardiorespiratory abnormalities.
For the dogs of the present study, mean HR and MAP did not change significantly when TLV was transitioned to OLV and back again. The finding that OLV does not affect HR and MAP was consistent with results of studies involving propofol- and isoflurane-anesthetized human patients23 and isoflurane-anesthetized dogs during open-3 and closed-thoracic5 experimental models. Although the mean MAP for the Iso1.5MAC protocol was significantly lower than that for the other 2 protocols throughout the observation period, that finding was likely a reflection of the anesthetic protocol used rather than the type of ventilation because the same ventilation protocol was used for all 3 anesthetic protocols. If the differences in MAP were associated with type of ventilation, those significant differences should have been observed only during TLV (T30 and T120) or OLV (T50, T70, and T90).
Isoflurane also causes a dose-dependent decrease in CO.27,28 The mean CO for the IsoMAC protocol was consistently greater than the mean CO for the Iso1.5MAC protocol, which was likely a reflection of the lower isoflurane concentration that was administered during the IsoMAC protocol relative to the Iso1.5MAC protocol. Positive responses to a stimulus such as sympathetic changes, muscle twitches, and protective reflexes are eliminated when the MAC for an anesthetic is achieved in a patient.21 Nevertheless, dogs are more prone to respond to stimulation when anesthesia is maintained at the MAC than when anesthesia is maintained at 1.5 MAC because they are at a lighter plane of anesthesia. However, the neuromuscular blockade administered during the observation period may have masked or mitigated the positive responses of dogs to stimulation.
One-lung ventilation does not alter CO in human patients.20 Similarly, OLV was not associated with substantial changes in CO for the dogs of the present study regardless of anesthetic protocol administered or for halothane-anesthetized dogs of another study,2 although the mean CO for the dogs of that study2 was lower than that for the dogs of this study. The mean CO did not differ significantly between OLV and TLV for the dogs of other studies3,5 in which anesthesia was maintained with isoflurane at 1.5 MAC. Collectively, these results indicate that OLV does not affect the CO of dogs.
In the present study, the mean MPAP increased significantly during the observation period for all 3 anesthetic protocols, although it remained within the reference range (11 to 19 mm Hg) for dogs.25 Those findings were consistent with those of other studies3,5 that compared OLV with TLV in dogs anesthetized with isoflurane at 1.5 MAC. In another study,6 the increase in the MPAP of dogs during OLV was associated with the PEEP but not the zero end-expiratory pressure, and a gradual increase in PEEP during the observation period of the present study may have caused the observed increase in MPAP. However, when the MPAP of healthy dogs undergoing experimental thoracoscopy was compared at the zero end-expiratory pressure and PEEPs of 2.5 and 5.0 cm H2O, it did not change significantly when dogs were transitioned from TLV to OLV.4 The reason for the discrepant results regarding MPAP between that study4 and the present study may be associated with the difference in the experimental models used for the 2 studies. The present study used a closed-thoracic experimental model, whereas that other study4 used an open-thoracic (thoracoscopy) experimental model. During thoracoscopy the thoracic cavity goes from a negative-pressure environment to a positive-pressure environment, which increases the pressure exerted on the large vessels such as the pulmonary artery.
The mean PAOP did not change significantly during the observation period for the dogs of the present study regardless of the anesthetic protocol used or for dogs of another study5 than underwent OLV during a closed-thoracic experimental model. Conversely, the PAOP increased significantly when TLV was transitioned to OLV in dogs undergoing an open-thoracic experimental model3; however, the maximum increase in PAOP detected during that study3 was approximately 10 mm Hg, which is not clinically relevant. Moreover, an open-thoracic experimental model was used in that study,3 and the introduction of positive pressure into the thorax affects blood flow in the pulmonary vessels, particularly the capillaries.
One-lung ventilation did not significantly affect the PVR of the dogs of the present study and other studies3,5 regardless of whether they were anesthetized with propofol or isoflurane. Likewise, OLV does not affect the PVR in human patients anesthetized with sevoflurane,15,20 isoflurane,11,15 or propofol.11
The effect of OLV on the Qs/Qt is a major concern. In human patients anesthetized with isoflurane or sevoflurane, the Qs/Qt during OLV is significantly greater than that during TLV.15 Volatile anesthetic agents inhibit HPV and increase Qs/Qt.12,13 However, the mean Qs/Qt during OLV for human patients anesthetized with propofol did not differ significantly from the mean Qs/Qt during OLV for human patients anesthetized with sevoflurane.18 In the present study, the mean Qs/Qt during OLV for the Iso1.5MAC protocol was consistently greater, albeit not significantly so, than that for the IsoMAC protocol, which was likely the result of the dose-dependent inhibitory effect of isoflurane on HPV. The dose-dependent depressive effects of inhalant anesthetics on CO might mitigate HPV inhibition.12,14 Mitigating HPV inhibition would be beneficial during OLV. However, that benefit was not observed in the present study because the mean Qs/Qt for the Iso1.5MAC protocol did not differ significantly from that for the IsoMAC or propofol protocol at any time during the observation period. It is important to consider that the propofol infusion rate used for the propofol protocol might not have been equipotent to the IsoMAC and Iso1.5MAC protocols, even though clinical assessment of the anesthetic depth of dogs was similar among all 3 protocols.
Results of multiple studies8,15,24 indicate that the Qs/Qt during OLV can be decreased by maintaining CPAP in the nonventilated lung, something that may not be possible when some types of bronchial blockers are used.1 In dogs, use of some types of bronchial blockers can result in an increase of up to 30% in the Qs/Qt during OLV.5 In the present study, use of a DLT and maintaining CPAP in the nonventilated lung likely minimized the effect of OLV on the Qs/Qt. However, that is simply an assumption because there was no control group to assess the effect of CPAP in the nonventilated lung on variable results. Also, blood flow to the lungs is affected by gravity.19 It is possible the thoracic anatomy of dogs may facilitate pulmonary circulation during OLV when they are positioned in lateral recumbency.
Another option for improving oxygenation and blood circulation during OLV is maintaining a consistent PEEP in the dependent (ventilated) lung. A PEEP of approximately 5 cm H2O is usually recommended for OLV and has been associated with a decrease in the Qs/Qt in various studies.4,26,29 In the present study, the PEEP of the ventilated lung was maintained at 5 cm H2O during OLV, which might have also minimized the effects of OLV on the Qs/Qt. Although the Qs/Qt was low in this study, it is important to emphasize that all study dogs were healthy without evidence of thoracic disease. Systemic changes associated with thoracic diseases that require surgical intervention generally increase anesthetic risk and may also increase the Qs/Qt. Given that the mean Qs/Qt for the propofol protocol remained fairly stable and the mean Qs/Qt was greatest for the Iso1.5MAC protocol throughout the observation period of this study, we suggest that propofol may be preferable to isoflurane for anesthetizing dogs that require OLV, and isoflurane should be administered only at low concentrations when it must be used.
Although the transition from TLV to OLV caused a decrease in Pao2 for the dogs of the present study, it remained at approximately 600 mg Hg, which indicated that alveolar oxygenation was not impaired during OLV. A decrease in Pao2 should be expected during OLV because less lung tissue is available for oxygen exchange.
In the present study, changes in Pao2 – Pao2 were similar to those for Qs/Qt, with significant changes over time detected only for the Iso1.5MAC protocol. The Pao2 – Pao2 increased from 194 to 550 mm Hg 60 minutes after initiation of OLV by use of a wire-guided bronchial blocker in a dog undergoing thoracoscopic partial pericardiectomy,7 and that dramatic increase was likely associated with lung collapse that usually occurs with the use of such devices. Changes in Pao2 – Pao2 associated with OLV for the dogs of the present study were not so dramatic and were similar to those of clinically normal dogs of another study5 that also underwent a closed-thoracic scenario. In that study,5 the mean Pao2 – Pao2 increased from 102 to 375 mm Hg during the first 15 minutes of OLV. Collectively, these findings indicated that OLV by use of a DLT can be safely performed in dogs without any clinically relevant changes in arterial oxygenation.
In the present study, only 1 dog developed a Pao2 – Pao2 > 300 mm Hg (during the Iso1.5MAC protocol), but alveolar oxygenation was not impaired. This was the dog that required alveolar recruitment maneuvers during recovery from the Iso1.5MAC protocol. It is possible that the DLT was incorrectly positioned in that dog during that protocol. Intubation was performed prior to instrumentation, and although the DLT was confirmed to be in the correct position immediately after placement, it may have been inadvertently moved during instrumentation. Thus, correct positioning of the DLT should be verified following instrumentation in future studies. The silent lung loci detected prior to and during alveolar recruitment maneuvers for this dog corresponded anatomically to the left caudal lung lobe.
Results of serial monitoring of blood gas variables indicated that the dogs of the present study had respiratory acidosis throughout the observation period, which was exacerbated during OLV. Mechanical ventilation settings used in this study were expected to maintain the Petco2 within reference limits (35 to 45 mm Hg), but that was not achieved during OLV, even though the ventilator respiratory rate was set at 30 breaths/min. A similar phenomenon was observed for dogs of another study5 during OLV in a closed-thoracic experimental model, although the increases in Petco2 and Paco2 in that study5 were not as extreme as those observed in the present study. For the dogs of this study, the mean Petco2 and Paco2 both decreased between T90 and T120 when dogs were transitioned from OLV back to TLV. Because Petco2 was measured in real time by capnography and Paco2 was measured only when blood samples were collected for blood gas analysis, the Paco2 was frequently > 45 mm Hg while the Petco2 was within the intended range. This was reflected in the fairly large values recorded for Paco2 – Petco2. Although the mean Paco2 did not differ significantly among the 3 anesthetic protocols at any time during the observation period, it was consistently greatest for the IsoMAC protocol.
The present study had several limitations. A primary concern associated with OLV is atelectasis. Auscultation alone cannot definitively determine whether atelectasis is or is not present. Continued monitoring of the Qs/Qt after the observation period during anesthesia recovery could have been used to assess atelectasis but was not performed because the Qs/Qt remained within the reference range throughout the observation period for all dogs. The DLTs used for the dogs of this study were developed for human patients, and there are substantial anatomic differences between the 2 species that limit the use of DLTs developed for humans in dogs. During the intubation training phase, it was noted that the tracheal diameter of canine cadavers that weighed < 10 kg was quite small relative to the diameter of the smallest DLT, and canine cadavers that weighed > 25 kg generally had long tracheas, which precluded bronchial intubation even with the largest DLT available.
Another limitation of the present study was the fact that neuromuscular stimulation was not used to assess efficacy of atracurium administration. It is possible the dog that required alveolar recruitment maneuvers during anesthesia recovery had inadequate ventilation because of persistence of the atracurium-induced neuromuscular blockade. Although atracurium was administered at constant intervals to all dogs, the efficacy of the drug might have varied among individual dogs, and it was not possible to determine the extent of the neuromuscular blockade without neuromuscular stimulation. Additionally, atracurium can cause histamine release, which in turn can cause hypotension.30 However, atracurium-induced histamine release is rarely observed following administration of clinical doses of the drug, and cardiovascular changes caused by atracurium alone are minimal.30
To our knowledge, the present study was the first in which pulmonary hemodynamics and alveolar oxygenation of healthy dogs during OLV in a closed-thoracic experimental model were compared between IV and inhalant anesthetic protocols. Results indicated that pulmonary hemodynamics and alveolar oxygenation during OLV generally did not differ significantly between anesthetic protocols involving the use of a CRI of propofol and those involving isoflurane at the MAC or 1.5 MAC. Thus, OLV can be successfully performed by use of a DLT in dogs anesthetized with either propofol or low concentrations of isoflurane. Frequent serial blood gas analyses are also recommended so that the Paco2 – Petco2 can be calculated and the mechanical ventilator settings adjusted to minimize the magnitude of respiratory acidosis induced during OLV.
Acknowledgments
This manuscript represents a portion of a thesis submitted by Dr. Floriano to the Faculty of Veterinary Medicine of Araçatuba as partial fulfillments of the requirements for a Doctor of Philosophy degree.
Supported by the São Paulo Research Foundation (FAPESP No. 2013/05062-4).
The authors thank Professor Silvia Helena Venturoli Perri for assistance with statistical analyses.
ABBREVIATIONS
| Cco2 | Capillary oxygen concentration |
| CO | Cardiac output |
| CPAP | Constant positive airway pressure |
| CRI | Constant rate infusion |
| DLT | Double-lumen endotracheal tube |
| HPV | Hypoxic pulmonary vasoconstriction |
| HR | Heart rate |
| MAC | Minimum alveolar concentration |
| MAP | Mean arterial blood pressure |
| MPAP | Mean pulmonary artery pressure |
| OLV | One-lung ventilation |
| Pao2 | Alveolar partial pressure of oxygen |
| Pao2 – Pao2 | Alveolar-arterial difference in partial pressure of oxygen |
| PAOP | Pulmonary artery occlusion pressure |
| PEEP | Positive end-expiratory pressure |
| Petco2 | End-tidal partial pressure of carbon dioxide |
| PVR | Pulmonary vascular resistance |
| Qs/Qt | Arteriovenous shunt fraction |
| SV | Stroke volume |
| TLV | Two-lung ventilation |
Footnotes
Stimulator Grass-S48, Astromed Inc, Los Angeles, Calif.
Anesthetic Machine Nikkei K, Takaoka, São Paulo, Brazil.
Ruschelit Bronchopart, Willy Rüsch AG, Kernen, Germany.
Cardiocap 5, Datex Ohmeda, Helsinki, Finland.
Quick Cal Calibration Gas, Datex-Engstrom Division Instrumentarium Corp, Helsinki, Finland.
Nipro 22G, Nipro Medical, Sorocaba, Brazil.
Pediatric Swan-Ganz catheter 132F5/75 cm, Edwards Life-sciences, São Paulo, Brazil.
DX 2020 Cardiac output module, Dixtal, Manaus, Brazil.
Warm Air Blower Unit TC3000, Gaymar Industries Inc, Orchard Park, NY.
Syringe pump BSS 200-Biosensor, São Paulo, Brazil.
Mechanical Ventilator Nikkei K, Takaoka, São Paulo, Brazil.
Y-Connector Ruschelit Bronchopart, Willy Rüsch AG, Kernen, Germany.
i-Stat, Abbott Laboratorie, São Paulo, Brazil.
References
1. Bjork VO, Carlens E, Friberg O. Endobronchial anesthesia. Anesthesiology 1953; 14: 60–72.
2. Cantwell SL, Duke T, Walsh PJ, et al. One-lung versus two-lung ventilation in the closed-thoracic anesthetized dog: a comparison of cardiopulmonary parameters. Vet Surg 2000; 29: 365–373.
3. Kudnig ST, Monnet E, Riquelme M, et al. Effect of one-lung ventilation on oxygen delivery in anesthetized dogs with an open thoracic cavity. Am J Vet Res 2003; 64: 443–448.
4. Kudnig ST, Monnet E, Riquelme M, et al. Effect of positive end-expiratory pressure on oxygen delivery during 1-lung ventilation for thoracoscopy in normal dogs. Vet Surg 2006; 35: 534–542.
5. Riquelme M, Monnet E, Kudnig ST, et al. Cardiopulmonary changes induced during one-lung ventilation in anesthetized dogs with a closed thoracic cavity. Am J Vet Res 2005; 66: 973–977.
6. Riquelme M, Monnet E, Kudnig ST, et al. Cardiopulmonary effects of positive end-expiratory pressure during one-lung ventilation in anesthetized dogs with a closed thoracic cavity. Am J Vet Res 2005; 66: 978–983.
7. Bauquier SH, Culp WT, Lin RC, et al. One-lung ventilation using a wire-guided endobronchial blocker for thoracoscopic pericardial fenestration in a dog. Can Vet J 2010; 51: 1135–1138.
8. Adami C, Axiak S, Rytz U, et al. Alternating one lung ventilation using a double lumen endobronchial tube and providing CPAP to the non-ventilated lung in a dog. Vet Anaesth Analg 2011; 38: 70–76.
9. Mayhew PD, Culp WT, Pascoe PJ, et al. Evaluation of blind thoracoscopic-assisted placement of three double-lumen endobronchial tube designs for one-lung ventilation in dogs. Vet Surg 2012; 41: 664–670.
10. Mayhew PD, Friedberg JS. Video-assisted thoracoscopic resection of noninvasive thymomas using one-lung ventilation in two dogs. Vet Surg 2008; 37: 756–762.
11. Kellow NH, Scott AD, White SA, et al. Comparison of the effects of propofol and isoflurane anaesthesia on right ventricular function and shunt fraction during thoracic surgery. Br J Anaesth 1995; 75: 578–582.
12. Loer SA, Scheeren TW, Tarnow J. Desflurane inhibits hypoxic pulmonary vasoconstriction in isolated rabbit lungs. Anesthesiology 1995; 83: 552–556.
13. Uno H, Ishibe Y, Umeda T, et al. Evaluation of inhibitory effect of isoflurane on the hypoxic pulmonary vasoconstriction response in dogs. Masui 1994; 43: 1288–1296.
14. Schwarzkopf K, Schreiber T, Bauer R, et al. The effects of increasing concentrations of isoflurane and desflurane on pulmonary perfusion and systemic oxygenation during one-lung ventilation in pigs. Anesth Analg 2001; 93: 1434–1438.
15. Abe K, Mashimo T, Yoshiya I. Arterial oxygenation and shunt fraction during one-lung ventilation: a comparison of isoflurane and sevoflurane. Anesth Analg 1998; 86: 1266–1270.
16. Huang CH, Wang YP, Wu PY, et al. Propofol infusion shortens and attenuates oxidative stress during one lung ventilation. Acta Anaesthesiol Taiwan 2008; 46: 160–165.
17. Karzai W, Haberstroh J, Priebe HJ. Effects of desflurane and propofol on arterial oxygenation during one-lung ventilation in the pig. Acta Anaesthesiol Scand 1998; 42: 648–652.
18. Beck DH, Doepfmer UR, Sinemus C, et al. Effects of sevoflurane and propofol on pulmonary shunt fraction during one-lung ventilation for thoracic surgery. Br J Anaesth 2001; 86: 38–43.
19. Pruszkowski O, Dalibon N, Moutafis M, et al. Effects of propofol vs sevoflurane on arterial oxygenation during one-lung ventilation. Br J Anaesth 2007; 98: 539–544.
20. Saito M, Cho S, Morooka H, et al. Effects of sevoflurane compared with those of isoflurane on arterial oxygenation and hemodynamics during one-lung ventilation. J Anesth 2000; 14: 1–5.
21. Valverde A, Morey TE, Hernández J, et al. Validation of several types of noxious stimuli for use in determining the minimum alveolar concentration for inhalation anesthetics in dogs and rabbits. Am J Vet Res 2003; 64: 957–962.
22. Macphail CM. Cirurgia do sistema cardiovascular. In: Fossum TW, ed. Cirurgia de pequenos animais. 4th ed. Rio de Janeiro: Elsevier, 2014; 856–905.
23. Dabir S, Mohammad-Taheri Z, Parsa T, et al. Effects of propofol versus isoflurane on liver function after open thoracotomy. Asian Cardiovasc Thorac Ann 2015; 23: 292–298.
24. Kim YD, Ko S, Kim D, et al. The effects of incremental continuous positive airway pressure on arterial oxygenation and pulmonary shunt during one-lung ventilation. Korean J Anesthesiol 2012; 62: 256–259.
25. Haskins SC. Monitoring anesthetized patients. In: Grimm KA, Lamont LA, Tranquilli WJ, et al, eds. Veterinary anesthesia and analgesia: the fifth edition of Lumb and Jones. 5th ed. Ames, Iowa: Wiley Blackwell, 2015; 86–113.
26. Muir WW. Cardiovascular physiology. In: Grimm KA, Lamont LA, Tranquilli WJ, et al, eds. Veterinary anesthesia and analgesia: the fifth edition of Lumb and Jones. 5th ed. Ames, Iowa: Wiley Blackwell, 2015; 417–472.
27. Shih AC, Vigani A, Loring N, et al. Cardiopulmonary effects of a new inspiratory impedance threshold device in anesthetized hypotensive dogs. Vet Anaesth Analg 2010; 37: 215–221.
28. Floriano BP, Wagatsuma JT, Ferreira JZ, et al. Effects on indicators of tissue perfusion in dogs anesthetized with isoflurane at two multiples of the minimum alveolar concentration. Am J Vet Res 2016; 77: 24–31.
29. Inomata S, Nishikawa T, Saito S, et al. “Best” PEEP during one-lung ventilation. Br J Anaesth 1997; 78: 754–756.
30. Basta SJ, Savarese JJ, Ali HH, et al. Histamine-releasing potencies of atracurium, dimethyl tubocurarine and tubocurarine. Br J Anaesth 1983; 55(suppl 1):105S–106S.
