Dexmedetomidine is a highly selective α2-adrenergic receptor agonist associated with sedative and analgesic sparing effects, reduced delirium and agitation, perioperative sympatholysis, cardiovascular stabilizing effects, and preservation of respiratory function. Therefore, it has been successfully employed as a total anesthetic agent in day care minimally invasive procedures and other minor procedures facilitating early discharge [14, 15]. In 2006, Lee et al. reported that intraoperative dexmedetomidine administration could improve the quality of recovery, postoperative pulmonary function, and other outcomes, and reduce hospital stay in patients undergoing VATS [16]. These findings were related to reduced pain, postoperative nausea and vomiting, emergence agitation, and opioid consumption following dexmedetomidine administration, making it a suitable choice for patients undergoing VATS.
These characteristics of dexmedetomidine are just what non-intubated VATS requires. Therefore, our study retrospectively determined the perioperative outcomes in 100 patients who underwent non-intubated VATS with thoracic epidural analgesia, intercostal nerve block, propofol, and either dexmedetomidine or alfentanil.
Doo et al. reported a prevalence of dexmedetomidine-induced hemodynamic instability of 14.7% in patients administered at a loading dose of 1 µg/kg, followed by 0.2–0.6 µg/kg/hr for sedation. Female sex and obesity were found to be associated with a higher probability of developing dexmedetomidine-induced hemodynamic instability [17]. Hypotension occurred more frequently when dexmedetomidine was rapidly titrated [18]. Taittonen et al suggested that premedication dose with 0.33–0.67 mg/kg IV given 15 min before surgery could minimize side effects of hypotension and bradycardia [19]. Therefore, in our study, dexmedetomidine infusion was administered at a very slow rate, with loading dose at 0.5 µg/kg for 30 min and then maintenance dose at 0.5 µg/kg/hr until the surgery finished. Hence, the duration of anesthetic induction was longer in group D than in group O. Compared to group O, group D showed a significant decrease in heart rate, systolic blood pressure, diastolic blood pressure, and mean arterial pressure, from baseline values, and maintained a relatively stable hemodynamic status. In addition to the side effects of dexmedetomidine, the effects of vasodilation by propofol and thoracic epidural analgesia might not be excluded due to hypotension and bradycardia in both groups [12, 20]. Adequate pain control can prevent elevated blood pressure and accelerated heart rate [21]. Fluid challenge and ephedrine were administered for treatment.
Intraoperative one-lung arterial blood gas revealed lower pH and significant CO2 retention in group O, which indicates the preservation effects of CO2 elimination and the respiratory function of dexmedetomidine in group D. Although dexmedetomidine infusions did not result in clinically significant respiratory depression, sedation with dexmedetomidine reduced both hypoxic and hypercapnic regulation of breathing to a similar extent as sedation with propofol [22]. In addition, dexmedetomidine-induced sedation causes upper airway obstruction and episodes of apnea to the same degree as propofol-induced sedation [23]. Therefore, the respiratory condition and expiratory ETCO2 were closely monitored in our study. Pre-one lung arterial blood gas data obtained after prescribing midazolam and loading dose of dexmedetomidine revealed no significant difference compared to the alfentanil group premedicated with midazolam.
We routinely applied THRIVE for the entire duration of non-intubated VATS. One-lung ventilation during non-intubated VATS carries the risk of hypoxemia and hypercapnia, whereas THRIVE increases the oxygenation reserve status and efficiency of CO2 elimination [24]. Increased oxygenation and CO2 elimination reduce the respiratory dead space owing to the higher flow of oxygen [25–27]. In our study, pre-one-lung and one-lung oxygenation was similar between the two groups.
In our study, eight patients suffered from PONV, which included one patient in group D and seven patients in group O. PONV high-risk patients should be predicted early using the Apfel score, and prokinetic premedication should be considered [28]. Serotonin receptor antagonists are effective in preventing and treating PONV with minor side effects. It binds to the serotonin receptor in both the vagal afferents of the gastrointestinal tract and the chemoreceptor trigger zone. Besides PONV, other common opioid-related side effects such as dyspnea, constipation, dizziness, and skin itching were also noted. In a meta-analysis by Ohishi et al., median occurrence rates of somnolence and dizziness in opioid-treated patients were 21% (range: 10–39%) and 22% (range: 10–37%), respectively [29]. In our study, dizziness was significantly more common in group O than in group D (group D vs. group O: 0 vs. 4, p < 0.05). Prochlorperazine was prescribed in ordinary wards postoperatively for these patients. The occurrence of these side effects might be the reason why patients in group O had a significantly prolonged postoperative hospital stay than those in group D.
Our protocol was designed to control pain by intraoperative loading, continuous doses, and postoperative single-dose fentanyl (50 µg) through thoracic epidural analgesia, intraoperative intercostal nerve block as well as postoperative opioid and adjuvant analgesics, such as NSAIDs. Thoracic epidural placed before surgery has traditionally been the gold standard for achieving good analgesia acutely. Thereby, it reduces the likelihood of post-thoracotomy pain syndrome development and improves pulmonary mechanics and function by reducing splinting in thoracic surgery [30]. Another option is the intercostal nerve blockade [31], which has been considered a superior alternative to thoracic epidural if combined with a multimodal pain control regimen [32]. Morphine was administered in the postoperative room in our regimen when the numerical rating scale was rated higher than three points by patients. The consumption of morphine, diclofenac tablets, and Traceton® tablets was significantly decreased in group D, and the patients tolerated the treatment well without complications.
In 2015, Hung et al. reported that 13 patients converted to intubation and 1 patient converted to thoracotomy in non-intubated thoracoscopic lobectomy for lung cancer. Air leakage, subcutaneous emphysema, bleeding, and pulmonary and cardiovascular complications were also recorded [12]. In a retrospective study by Liu et al. in 2021, two patients were converted to multiportal incisions from uniportal thoracoscopic segmentectomy because of severe adhesion in the pleural cavity during the operation. In addition, one patient had prolonged air leakage [33]. In our study, there was no conversion to thoracotomy or multiportal incision. Surgical complications, including subcutaneous emphysema, air leakage, lung atelectasis, pneumothorax, and fever, were recorded, which were usually resolved or treated without sequelae. Patients were discharged earlier in group D than in group because of lesser opioid related discomforts. Oral NSAIDs tablets were prescribed for continuous OPD pain control.
This study has several limitations. First, this retrospective study evaluated anesthesia records of 100 patients between December 2016 and May 2022, which was conducted at a single-center, where statistical power might be limited by sample size. Second, patients in this study were sedated with either one of the two anesthesia regimens, selected by the anesthesiologist on duty. However, we applied a propensity bipartite matching score to approximate a random experiment that eliminates problems and selection bias associated with observational data analysis. Third, both techniques might not be equally distributed during the study period. Non-intubated VATS was a relatively new technique; thus, it would be an improvement in surgical and anesthesia techniques with time. Due to these limitations, prospective randomized controlled trials are required for further validation.