+++
Sternal Complications
++
The bilateral transsternal thoracotomy provides excellent exposure to the hila and pleural spaces, but problems have been reported with poor sternal healing. Brown and colleagues9 report an institutional prevalence of 36% for sternal disruption in transverse bilateral thoracosternotomy for lung transplantation, and they cited disruption rates of 20% to 60% at centers worldwide. Lung transplant recipients are particularly prone to poor sternal healing (Fig. 112-3) owing to their debilitated state and the routine use of postoperative corticosteroids.4 Sternal override is a common complication that results from a tendency of the distal sternum to angulate and displace anteriorly, a translational movement that is not prevented by the sternal wires. The addition of coaxial stabilization, either with long, thin Kirschner wires or with short, stout Steinmann pins placed within the cancellous bone of the sternum, reduces the incidence of sternal override and translational movement at the bony closure. However, these wires have a tendency to migrate, causing other problems (Fig. 112-4). We have removed numerous wires that have migrated from the sternum to various locations in the body. Such retrievals require interventions of various complexity, ranging from the administration of local anesthetic to liberate a wire that has eroded through the anterior chest wall to a laparoscopic procedure under general anesthesia to remove a Kirschner wire from the pouch of Douglas. Case reports suggest that the use of longitudinal titanium plates for sternal fixation may provide improved results for high risk patients or those requiring reclosure after sternal dehiscence. The Synthes Titanium Sternal Fixation System (Synthes, Westchester, PA) involves titanium plates that can be molded to the contour of the sternum, and includes a drill/titanium screw gauging system to ensure that the screws pierce both the anterior and posterior tables of the sternum, without protruding beyond the posterior border.9 As a prophylactic maneuver, the procedure does add time and cost to the operation, and further studies are needed to define which patients would benefit from such precautions.10,11
++
++
++
Deep sternal wound infection is an additional serious complication of transverse sternotomy. We have encountered this problem in several patients, and it has required operative and bedside wound debridement with additional antibiotics and a prolonged hospital stay. The estimated prevalence for all sternal closure complications in our historical control group is 34%. With this in mind, we routinely avoid sternal division and have found that adequate exposure in many circumstances can be provided by bilateral anterior thoracotomies alone. Additionally, in rare selected patients, we also advocate modified approaches such as a combined left posterolateral and right anterior thoracotomy to optimize the left hilar exposure without the need for sternal division or separate positioning, preparation, or draping. We have found that the use of a cardiac positioning device designed for off-pump cardiac stabilization (apical suction cup) can be useful to expose the left hilum and avoid sternal division and cardiopulmonary bypass.
+++
Primary Graft Dysfunction
++
Primary graft dysfunction is one of the most important complications of lung transplantation and represents a common cause of early mortality and prolonged ICU stay. The frequency of primary graft dysfunction at our institution is 23%.12 The impact of primary graft dysfunction is enormous. Patients experiencing initial graft dysfunction at our institution had a mortality of 28.8% compared with 4.2% in patients without the condition.
++
Primary graft dysfunction is commonly referred to as ischemia–reperfusion injury. A number of factors such as poor preservation techniques, prolonged ischemic time, and unsuspected donor lung pathology such as contusion, pulmonary thromboembolism, or aspiration all play a role in the development of primary graft dysfunction. Additionally, we have reported a statistically significant difference in the distribution of primary graft dysfunction according to underlying diagnosis leading to transplantation. There appears to be more primary graft dysfunction in patients transplanted for pulmonary hypertension and less in patients transplanted for emphysema. Hyperacute rejection is exceedingly rare, but it must be a consideration in cases of early severe lung dysfunction. Ischemia–reperfusion injury is characterized by noncardiogenic pulmonary edema and progressive lung injury over the first few hours after implantation (Fig. 112-5). Pathologically severe ischemia–reperfusion injury has the appearance of diffuse alveolar damage. Regardless of cause, it is important to establish a diagnosis of early graft dysfunction and to rule out other treatable conditions. We perform open lung biopsy at the time of implantation if graft dysfunction is immediately apparent in the OR. Serologic evaluation for antihuman leukocyte antigen antibodies also may reveal evidence of hyperacute rejection in some patients.
++
++
Efforts to prevent primary graft dysfunction have included paying close attention to the inflation and ventilation of the donor lungs as well as the optimization of preservation methods.13 Considering that lung hyperinflation can produce a striking model of postreperfusion pulmonary edema, we are particularly careful to avoid lung hyperinflation during harvest and storage of the donor lungs.
++
Extensive research has been devoted to refining the preservation solution to maximally reduce cellular injury during ischemic time. It is clear from experimental14,15 and clinical work16 that the low potassium dextran solution provides superior protection compared with the high potassium solution used previously. In addition, experimental work suggests that adding nitroprusside (a potent nitric oxide donor) to the flush solution at the time of harvest provides a preservation advantage.17
++
Other strategies to reduce primary graft dysfunction include controlled reperfusion, a concept that was used originally to reduce cardiac dysfunction after reperfusion of acutely ischemic myocardium at the time of coronary artery revascularization. The clinical use of controlled reperfusion, where recipient blood depleted of leukocytes is administered into the pulmonary arteries of the newly transplanted lungs for the initial reperfusion phase,18,19 has shown promise as a preventive strategy for ischemia–reperfusion injury.
++
Ex vivo lung perfusion (EVLP), where harvested lungs are perfused and ventilated ex vivo at body temperature to allow for an evaluation of physiologic integrity, is another novel strategy under intense study. This technique aims to increase lung utilization and improve outcomes by providing time under physiologically protective conditions to adequately assess and optimize suboptimal organs. A prospective clinical trial compared outcomes after high-risk donor lungs (N = 23) were subjected to 4 hours of EVLP, after which N = 20 were deemed acceptable for transplantation. They found no difference in the incidence of primary graft dysfunction in the high-risk organs that went on to implantation when compared to the control group (N = 116) of conventionally selected donor lungs. No conclusion regarding the relationship between EVLP and primary graft dysfunction can be made based on this small and nonrandomized study, but further studies are warranted.20
++
In cases of established ischemia–reperfusion injury, treatment is supportive. Management strategies are similar to those utilized for ARDS, and include diuresis and balancing the appropriate ventilatory support without ventilator-induced injury. In most cases, the reperfusion injury peaks and begins to resolve over 24 to 48 hours. We have shown previously that inhaled nitric oxide is beneficial in severe reperfusion injury because it decreases pulmonary artery pressure and improves the PaO2/FiO2 ratio.21 More recently, inhaled prostacyclin has been investigated and shows promise as an economic alternative to nitric oxide.22,23
++
While standard intensive ventilatory and pharmacologic interventions generally suffice, severe graft dysfunction or coexisting cardiac failure may require extracorporeal membrane oxygenation (ECMO) support. We reported the selective use of ECMO after lung transplantation24 and found satisfactory results when lung failure occurs immediately after transplantation (<24 hours). More recently, Bermudez et al. described the University of Pittsburgh experience with ECMO (7.6% of 763 lung transplantation patients) where support was initiated between 0 and 7 days after transplant. Thirty day survival was 56%. Long-term survival rates in the ECMO patients was significantly worse at 1 year and 5 years than the overall survival rates for the entire cohort (40% and 25% as compared to 82% and 54%, respectively).25
++
An alternative approach to severe, reversible allograft dysfunction has been reported by Eriksson and Steen,26 who have used core cooling successfully to reduce oxygen requirements and avoid ECMO, giving a chance for the lung to heal. As a last resort, acute retransplantation may be lifesaving. In our series, we have performed a small number of retransplants emergently for primary graft failure (Fig. 112-6).7 The indications for such expensive and dramatic therapy include the presence of unrelenting lung dysfunction, respiratory failure, and the absence of systemic complications (renal failure, infection) which would greatly diminish the chance of success.
++
++
Airway complications formerly were a major cause of morbidity and mortality after pulmonary transplantation. In the current era of transplantation, airway complications remain a source of morbidity but do not appear to be associated with poorer survival.7 Using standard methods of implantation, the donor bronchus is rendered ischemic, without reconstitution of its systemic bronchial artery circulation. The donor bronchus relies on collateral pulmonary artery blood flow during the first few days after transplantation. It has been demonstrated that pulmonary collateral flow contributes substantially to bronchial viability at the level of the distal bronchus and lobar origin. A shortened donor bronchial length reduces the length of donor bronchus dependent on collateral flow. Work from our group has shown that a determined effort to shorten the donor bronchus prior to implantation is associated with a decreased rate of anastomotic complications.27 Superior preservation, improved sepsis prophylaxis, and better immunosuppression have reduced the incidence of airway complication. Across the authors' entire series, the rate of airway anastomotic complications of all degrees of severity is 9.3%.12
++
Airway anastomotic complications include bronchial stenosis, dehiscence, excessive granulation tissue formation, tracheobronchial malacia, fistulas, and infections. Bronchial stenosis represents the most commonly reported problem. Airway complications can be identified in a number of ways. Routine postoperative bronchoscopic surveillance generally provides early evidence of an anastomotic complication. On occasion, CT scan, performed for some other indication, demonstrates an unexpected airway stenosis or dehiscence. In fact, we learned that CT scanning is a useful diagnostic tool for evaluating documented or suspected donor airway complications (Fig. 112-7). Late airway stenoses generally manifest with symptoms of dyspnea, wheeze, or decreased forced expiratory volume in 1 second (FEV1). Bronchoscopic assessment confirms the diagnosis (Fig. 112-8).
++
++