Jens Lohser

Acute kidney injury after lung resection surgery: incidence and perioperative risk factors.

BACKGROUND: Postoperative acute kidney injury (AKI) is associated with increased perioperative morbidity and mortality in a variety of surgical settings, but has not been well studied after lung resection surgery. In the present study, we defined the incidence of postoperative AKI, identified risk factors, and clarified the relationship between postoperative AKI and outcome in patients undergoing lung resection surgery. METHODS: A retrospective, observational study of patients who underwent lung resection surgery between January 2006 and March 2010 in a tertiary care academic center was conducted. Postoperative AKI was diagnosed within 72 hours after surgery based on the Acute Kidney Injury Network creatinine criteria. Logistic regression was used to model the association between perioperative factors and the risk of AKI within 72 hours after surgery. The relationship between postoperative AKI and patient outcome including mortality, days in hospital, and the requirement of reintubation was investigated. RESULTS: A total of 1129 patients (pneumonectomy n = 71, bilobectomy n = 30, lobectomy n = 580, segmentectomy n = 35, wedge resection/bullectomy n = 413) were included in the final analysis. Patients were an average of 61 years (SD 15) and 50% were female. AKI was diagnosed in 67 patients (5.9%) based on Acute Kidney Injury Network criteria (stage 1, n = 59; stage 2, n = 8; and stage 3, n = 0) within 72 hours after surgery, and only 1 patient required renal replacement therapy. Multivariate analysis demonstrated an independent association between postoperative AKI and hypertension (adjusted odds ratio [OR] 2.0, 95% confidence interval [CI]: 1.1–3.8), peripheral vascular disease (OR 4.4, 95% CI: 1.8–10), estimated glomerular filtration rate (OR 0.8, 95% CI: 0.69–0.93), preoperative use of angiotensin II receptor blockers (OR 2.2, 95% CI: 1.1–4.4), intraoperative hydroxyethyl starch administration (OR 1.5, 95% CI: 1.1–2.1), and thoracoscopic (versus open) procedures (OR 0.37, 95% CI: 0.15–0.90). Development of AKI was associated with increased rates of tracheal reintubation (12% vs 2%, P < 0.001), postoperative mechanical ventilation (15% vs 3%, P < 0.001), and prolonged hospital length of stay (10 vs 8 days, P < 0.001). There was no difference in mortality between the 2 groups (3% vs 1%, P = 0.12). CONCLUSIONS: Preoperative risk factors for AKI after lung resection surgery overlap with those established for other surgical procedures. Perioperative management seems to influence the risk of AKI after lung resection; in particular, the use of synthetic colloids may increase the risk, whereas thoracoscopic procedures may decrease the risk of AKI. Early postoperative AKI is associated with respiratory complications and prolonged hospitalization.

Lung Injury After One-Lung Ventilation: A Review of the Pathophysiologic Mechanisms Affecting the Ventilated and the Collapsed Lung.

Lung injury is the leading cause of death after thoracic surgery. Initially recognized after pneumonectomy, it has since been described after any period of 1-lung ventilation (OLV), even in the absence of lung resection. Overhydration and high tidal volumes were thought to be responsible at various points; however, it is now recognized that the pathophysiology is more complex and multifactorial. All causative mechanisms known to trigger ventilator-induced lung injury have been described in the OLV setting. The ventilated lung is exposed to high strain secondary to large, nonphysiologic tidal volumes and loss of the normal functional residual capacity. In addition, the ventilated lung experiences oxidative stress, as well as capillary shear stress because of hyperperfusion. Surgical manipulation and/or resection of the collapsed lung may induce lung injury. Re-expansion of the collapsed lung at the conclusion of OLV invariably induces duration-dependent, ischemia-reperfusion injury. Inflammatory cytokines are released in response to localized injury and may promote local and contralateral lung injury. Protective ventilation and volatile anesthesia lessen the degree of injury; however, increases in biochemical and histologic markers of lung injury appear unavoidable. The endothelial glycocalyx may represent a common pathway for lung injury creation during OLV, because it is damaged by most of the recognized lung injurious mechanisms. Experimental therapies to stabilize the endothelial glycocalyx may afford the ability to reduce lung injury in the future. In the interim, protective ventilation with tidal volumes of 4 to 5 mL/kg predicted body weight, positive end-expiratory pressure of 5 to 10 cm H2O, and routine lung recruitment should be used during OLV in an attempt to minimize harmful lung stress and strain. Additional strategies to reduce lung injury include routine volatile anesthesia and efforts to minimize OLV duration and hyperoxia.

One-lung ventilation and arterial oxygenation.

Purpose of review Hypoxemia during one-lung ventilation (OLV) has become less common; however, it may still occur in about 10% of cases. We review recent developments which may affect the incidence and treatment of hypoxemia during OLV. Recent findings Changes in surgical techniques are affecting oxygenation during OLV. The increased use of the supine position may adversely affect the prevalence of hypoxemia, whereas the increased application of thoracoscopic techniques is limiting the treatment options. Treatment options such as global or selective recruitment maneuvers and drug effects of dexmedetomidine and epoprostenol on arterial oxygenation during OLV are discussed. Capnometry prior to, or early during OLV, may in fact be able to predict the degree of hypoxemia during OLV. Persistent controversies surrounding the effect of epidural anesthesia, ventilatory modalities and gravity are reviewed. Summary Interesting concepts have emerged from case reports and small studies on the treatment and prediction of hypoxemia during OLV. Definitive studies on the most effective ventilatory mode remain elusive. End-organ effects of OLV are an exciting new concept that may shape clinical practice and research going forward.

Anesthesia for thoracic surgery in morbidly obese patients.

Purpose of review This review considers the anesthetic management of obese patients undergoing thoracic surgery. Extremely or morbidly obese patients differ from patients of normal weight in several ways. Obese patients have altered anatomy and physiology, and usually have associated comorbid medical conditions that may complicate their operative course and increase their risks for postoperative complications. Recent findings During anesthetic induction and laryngoscopy for tracheal intubation the morbidly obese patient should be in the reverse Trendelenburg position with the head and neck elevated above the table. Placement of a double-lumen tube should be no more difficult in an obese patient than in a normal-weight patient. There are no clear advantages for any of the commonly available inhalational anesthetic agents and each can be used for general anesthesia. Summary With proper attention to their special needs, the morbidly obese patient can safely undergo thoracic surgery and one-lung ventilation.

Current trends in anesthesia for esophagectomy.

Purpose of review Despite marked improvements in perioperative outcomes, esophagectomy continues to be a high-risk operation associated with significant morbidity and mortality. Progress has been achieved through evidence-based changes in preoperative optimization, intraoperative ventilation strategies, fluid therapy, and analgesia, as well as expedited postoperative recovery pathways. This review will summarize the recent literature on the anesthetic management of patients undergoing esophageal resection. Recent findings The current focus in publications on the perioperative management of esophagectomy patients can be summarized under the umbrella term of enhanced recovery pathways, focusing on ventilation, fluid therapy, analgesia and minimally invasive surgical approaches. Lung protective ventilation reduces pulmonary complications in cases requiring one-lung ventilation. Excess fluid administration contributes to morbidity while restrictive approaches have not resulted in an increased risk of acute kidney injury. Goal-directed fluid therapy remains intuitive yet unproven. Thoracic epidural analgesia reduces the systemic inflammatory response, pulmonary complications, and enhances postoperative pain control, yet if causing perioperative hypotension may be associated with anastomotic leaks. Enhanced recovery pathways have facilitated low morbidity and mortality rates in a high-risk population but are heterogeneous and limited by a weak evidence base. Minimally invasive surgical approaches are increasingly popular and appear to have at least equivalent outcomes to open procedures. Summary The morbidity and mortality after esophagectomy remains high despite significant improvements over the last decades. Enhanced recovery pathways appear promising in achieving further marginal gains but at present are lacking large scale, prospective, multicenter evidence.

Tension Pneumothorax During One-lung Ventilation

In Response: We agree with Bo et al. that meta-analyses of articles describing the results of well done randomized trials may represent valid evidence supporting treatment and are important contributions to the literature. The key here of course is the term “well done randomized trials.” In addition, a limitation of meta-analysis is that the results of some trials may not be included in the meta-analysis, thus skewing the results. Our intention was not to minimize the importance of meta-analyses by relegating them to a position below validation and evaluation studies, and acknowledge that with the hierarchy we described, the importance of meta-analyses might be misconstrued by our readers. However, although we agree that it may have been appropriate to rank meta-analyses more on a par with randomized trials, even if our analyses had been performed using the suggested hierarchy, our results would have been unchanged.