Respiratory infections after lung cancer resection

Bronchogenic carcinoma is the most frequent neoplasia (900,000 and 330,000 cases per year in men and women, respectively) and the most common cause of death from cancer in the world in both men and women [1]. The overall probability of being alive 5 years after diagnosis is 15%, a percentage that has barely changed in the past 35 years [2]. This is due, in part, to diagnosis in the advanced stages of the disease (less than 30% of cases are diagnosed in stages I and II when the disease is limited and surgical resection offers a probability of survival at 5 years of 70 and 40%, respectively). Early diagnosis is therefore vital in order to take advantage of surgical treatment, which is the option that currently offers the greatest probability of cure to the greatest number of patients.

The advances made in anesthesia and minimally invasive surgery have allowed surgical indications to be extended to those patients who are at greater risk of suffering from complications or dying after a lung resection. Moreover, multidisciplinary treatments of bronchogenic carcinoma have led to more pneumonectomies, with a subsequent increase in the morbidity and mortality associated with this resection. Reducing the complications and mortality associated with the surgical treatment of bronchogenic carcinoma is another important aspect to consider if the potential benefits to a greater number of patients treated with a lung resection are not to be significantly burdened by an increase in postoperative morbidity and mortality.

Respiratory complications are currently the most frequent and serious, as well as being the primary cause of death, after a lung resection owing to a bronchogenic carcinoma [3–7]. These are an ill-defined, heterogeneous group of diseases that are very inter-related and are accompanied in a syndromal manner by a greater or lesser degree of respiratory insufficiency, with various causes and pathogenic mechanisms. Some of these complications are related to surgery (bronchopleural fistula, postoperative hemorrhage and lobar torsion) or hemodynamic alterations, pulmonary capillary permeability or the balance of liquids (noncardiogenic lung edema, acute damage to the lungs or acute respiratory distress syndrome [ARDS]). There are also infectious complications, such as purulent tracheobronchitis and pneumonia, which involve factors, such as postoperative pain, nutritional state, pulmonary function or bronchial colonization. Finally, there are cardiovascular complications, including arrhythmias and pulmonary thromboembolism, and a group of respiratory complications of multifaceted etiology such as atelectasia and acute respiratory failure.

Pneumonia and ARDS are the most serious complications and the cause of 22–75% of deaths after lung resection owing to a carcinoma [3–6,8]. Pneumonia appears to be one of the main causes (if not the primary cause) of death after a lung resection in most published series. It is estimated that pneumonia has an incidence of 8% in surgical patients and this represents half of all nosocomial pneumonias. However, the risk to patients subjected to abdominal or thoracic surgery is 38-times greater than in other patients subjected to different operations. The incidence of pneumonia after a lung resection is 2–20% and the associated mortality is 15–75%; significantly higher than in operated patients who do not develop this disease [9–11].

The growing importance of respiratory infections as a cause of death after a lung cancer resection, as well as the resulting increase in the hospitalization period and expenses, has spurred a new interest in their prevention, diagnosis and treatment [12,13].

Respiratory infection after lung cancer resection

Nosocomial pneumonia has a prevalence of 0.5–1.0% in acute patients admitted to hospital and it comprises 15–18% of all infections acquired in hospital (the third most frequent after urinary infection [42%] and infection of the surgical wound [24%]). However, it is the infection that gives rise to the greatest morbidity and mortality in hospitals [14,15].

Patients subjected to a lung cancer resection present a series of factors that can predispose them to a respiratory infection:

•	Most suffer from a chronic obstructive lung disease; in 85% of cases, they are smokers whose distal airway has been colonized;
•	Patients are subjected to selective tracheobronchial intubation, mechanical ventilation and, in many cases, several fibrobronchoscopies to check or reposition the tube used for the selective intubation or aspiration of secretions;
•	Decrease in the capacity to take deep breaths and in the effectiveness of coughing can lead to the retention of secretions and the appearance of obstructive phenomena, such as lung segments, and partial collapses of the lung;
•	Neoplasia can give rise to alterations in the nutritional state, structural alterations of the airway (stenosis or obstruction) or tissue-necrosis phenomena;
•	Sometimes, patients have received treatments prior to surgery that are capable of reducing the body’s defenses, such as chemotherapy, or of provoking structural changes in the lung parenchyma, such as radiotherapy.

Pneumonias after lung resection are mostly bacterial and lobar, with a typical clinical presentation and a particularly fast onset and evolution. They are usually caused by Haemophilus influenzae, Streptococcus pneumoniae, Pseudomona aeruginosa, Klebsiella pneumoniae, Staphylococcus aureus and other Gram-negative aerobic and anaerobic bacteria [16].

Risk factors

From the pathogenic viewpoint, pneumonia is the result of an alteration in the balance between the defenses of the host and the presence in the respiratory system of a microorganism in sufficient quantities or of sufficient virulence to give rise to the disease. There are numerous descriptions of risk factors for pneumonia that can coincide in the perioperative period and that would act by altering this balance between the host’s defense system and the causal microorganism in some way. These risk factors include alteration of the nutritional state [17]; old age [18,19]; deteriorated lung function [20]; thoracotomy and lung resection themselves [4,21]; type and side of the lung resection [8,22]; alteration to the cleaning mechanisms of the bronchial tree [23]; reintubation [24]; smoking [13,25]; and a prolonged preoperative hospital stay, which has been linked to an increase in oropharyngeal colonization, particularly by Gram-negative bacilli [25,26].

Bronchial colonization

The role of the bronchial colonization of the distal airway in the development of a postoperative respiratory infection is unclear. Bronchial colonization would be the result of an equilibrium in which the host’s (deteriorated) defenses succeed in limiting, but not eradicating, the microorganisms adhering to the bronchial epithelium. Lung infection would be the result of a lowering of the host’s cellular and humoral defenses at a specific time or of the presence of an inoculum in a sufficient quantity or of a sufficient virulence to produce it. In our experience, after a prospective study of 78 patients operated consecutively for a lung cancer resection and subjected to bilateral protected bronchial brushings immediately before thoracotomy, the colonization by potentially pathogenic microorganisms in the distal airway was a prognostic parameter, regardless of the risk of postoperative respiratory infection (odds ratio: 1.5; 95% confidence interval: 1, 21–1) [27].

Several studies have linked preoperative bronchial colonization with the appearance of postoperative respiratory infections. Wansbrough-Jones and colleagues performed cultures from samples obtained by means of preoperative bronchoalveolar washing in 75 patients subjected to a pulmonary or pleural resection [28]. They found pathogenic germs in 12 out of 52 samples, with H. influenzae found most frequently. Of the patients, 11% developed an infectious respiratory complication in the postoperative stage, five out of 12 patients had pathogens in the airway (42%) and only two out of 42 patients had no pathogens (4.8%) – a statistically significant difference. The authors suggested that treating this colonization before surgery could reduce the number of postoperative lung infections. Sok and colleagues designed a prospective study in order to verify the origin of the pathogens that cause pleuropulmonary infections after lung cancer resection [29]. They studied samples of sputum 3 days before, during and 3 days after the operation. The infections appeared at 4.3 ± 2.9 days. In 75% of cases, the microorganisms that caused the lung infections were Gram-negative bacilli and Candida albicans. These were isolated in 18, 13 and 63% of sputum samples, pre-, intra- and postoperatively, respectively. They also found a strong association between the pathogens found in the sputum obtained 3 days after the operation and those found in the sputum of patients who developed a lung infection. The authors suggested that the colonization of the airway generally occurs during the postoperative period, from the patient’s oral cavity, pharynx and hypopharynx.

Patients affected by lung cancer present with a series of factors that predispose them to bronchial colonization; the most important of these is the association with chronic obstructive lung disease. The distal airways of these patients are colonized in up to 85% of cases. Other factors are immunosuppression originated by the neoplasia itself or by chemotherapy treatments; structural alterations to the bronchial tree due to cancer; deterioration of the general con-dition; performance of one or more fibrobronchoscopies; and presence of a bronchial obstruction that produces a combination of phenomena, such as atelectasia, bronchiectasias, mucus retention and bronchial inflammation. Many authors consider that obstructive pneumonitis is a noninfectious lung consolidation. By contrast, other authors have demonstrated that infection is a frequently associated finding.

Antibiotic prophylaxis in lung-resection surgery

Bernard and colleagues, in a randomized prospective study on patients subjected to lung surgery, compared two prophylaxis regimes with cephuroxime [30]. They observed that administration every 6 h for 48 h, as opposed to administration in a single dose, significantly reduced the number of respiratory infections and empiema. In 1994, Train and colleagues compared the efficacy of amoxicillin with cephamandol in preventing pulmonary infections in lung-resection surgery and they found no significant differences between the two antibiotics [31].

Boldt and colleagues, in 1999, suggested that the best schedule for antibiotic prophylaxis will depend, among other factors, on the bacteria that are present at the time when these patients undergo surgery [32]. They carried out a randomized prospective study in which they compared ampicillin sulbactam (group 1) and cephazoline (group 2). Furthermore, every patient was subjected to a preoperative bronchial aspirate through a double-light tube in order to carry out a microbiological study and determine whether isolated germs were sensitive to the antibiotics used. The isolated microorganisms were mainly H. influenzae, S. pneumoniae, Streptococcus viridans, S. aureus and K. pneumoniae. The authors showed that group 1 had significantly fewer pulmonary infections than group 2 and, in group 1, all the isolated bacteria were sensitive to the antibiotic used, while in group 2, eight out of the 25 isolated bacteria were not sensitive. When these findings were analyzed, they suggested that a microbiological examination can be useful for evaluating the efficacy of a prophylaxis regime for lung infection.

Studies by Sok and colleagues suggest that the colonization of the airway by pathogens in patients operated for lung cancer resection is basically produced during the postoperative period via the oral cavity, pharynx and hypopharynx [29]. This colonization is significantly associated with an increase in postoperative respiratory infections. Prevention of colonization in this period would be decisive for preventing these infections. The authors suggest that the prophylaxis schedule administered in the perioperative period would not strictly be applicable to respiratory infections after a lung resection. In our experience [33,34] 40% of patients with a bronchogenic carcinoma already have their distal airway colonized by pathogenic microorganisms at the time of the operation; postoperative lung infections are significantly more frequent in those patients who have their airway colonized in this way and concordance between the pathogens that colonized the airway and the causes of the postoperative respiratory infections is 51%.

The ideal regime for antibiotic prophylaxis for the prevention of respiratory infections was not discovered and this is still the case 20 years later.