Anaesthesiology Intensive Therapy, 2009,XLI,1; 38-41

Atypical and opportunistic pulmonary infections after cardiac surgery

*Jacek Puzio1, Ewa Kucewicz1, Mirosława Sioła2, Anna Dworniczak3, Jacek Wojarski 4, Sławomir Żegleń4, Waldemar Pakosiewicz1, Tomasz Maciejewski1, Marian Zembala2

1Oddział Kliniczny Kardioanestezji i Intensywnej Terapii Śląskiego Centrum Chorób Serca w Zabrzu

2Pracownia Mikrobiologii Śląskiego Centrum Chorób Serca w Zabrzu

3Klinika Chorób Płuc i Gruźlicy SPSK Nr 3 w Zabrzu

4Oddział Kliniczny Kardiochirurgii i Transplantologii Śląskiego Centrum Chorób Serca w Zabrzu


  • Table 1. Demographic characteristics of the population
  • Table 2. Types of cardiac surgery
  • Table 3. Preoperative conditions
  • Table 4. Pathogens in bronchial tree aspirates
  • Table 5. Concomitant mixed infections
  • Table 6. Early postoperative complications

Background. New diseases and therapies that lead to marked immunocompromise, have brought to medicine so-called opportunistic infections, caused by pathogens that usually do not cause disease in the presence of a healthy immune system. First diagnosed in AIDS patients, opportunistic infections have appeared in many other situations, including transplant recipients, other immunosuppressed patients, and even in otherwise healthy patients undergoing major elective surgery. Among the most common pathogens causing atypical infections are: Pneumocistis jiroveci, Chlamydia, Mycoplasma and Legionella. The aim of this retrospective study was to assess the incidence of the above mentioned infections in cardiac surgical patients whose early postoperative period was complicated by respiratory distress.

Methods. The postoperative course of 5026 cardiac surgical patients was analysed retrospectively. In 196 cases (3.9%), important respiratory complications, defined as prolonged (>24h) requirement for mechanical ventilation, pulmonary oedema, atelectasis, pneumothorax, haemothorax, infiltrations present on chest x-ray, or ARDS, were detected.

Results. Positive atypical sputum cultures (Chlamydia pneumoniae, Mycoplasma pneumoniae or Pneumocistis jiroveci) were obtained from 27 patients (16%). Pneumocistis jiroveci was the most common pathogen identified. Legionella pneumoniae was never cultured. The average duration of mechanical ventilation in the study group was 253 h, and the mortality 18.5% (5 patients).

Conclusions. Opportunistic infection with atypical pathogens should be suspected in the postoperative period when complicated by prolonged respiratory distress and difficulties with weaning from mechanical ventilation. Since these infections do not present with obvious symptoms, the diagnostic process should be extended and appropriate treatment introduced.

Nowadays, thanks to the progress in medical science, the diseases that medicine was powerless against can be treated. Hope was given to all those patients, who quite recently had to accept dramatic consequences related to the diagnoses established. Modern therapies modify the immune system; intended immunosuppressive management is increasingly common in long-term therapy of patients after organ transplants. All this created favourable conditions for the development of specific infections, called opportunistic infections, i.e. benefiting from suitable circumstances. The neutrality of microorganisms towards the host is maintained only when the immune systems functions normally. Such a condition is named as obligate commensalism or opportunism. Once favourable conditions are created, opportunistic microorganisms change their nature from non-pathogenic to pathogenic [1]. Such infections are also characteristic of patients with acquired immunodeficiency syndrome (AIDS). Moreover, the immune incompetence may be a consequence of environmental effects and individual lifestyle. Opportunistic infections result from the invasion of all types of pathogens: bacteria, viruses, parasites and fungi [2].  

The first report about an opportunistic infection was published in 1981 and described pneumocistis pneumonia spreading in the homosexual community. The vulnerability to such infections of the respiratory system results from impaired local immunity due to dysfunctions of the mucous membrane (decreased count of lymphoid cells and damaged macrophages). The infections are also likely to result from the activation of pathogens present in the airway (carrier state).

The term atypical pneumonia was introduced by Reiman in 1938 to describe the infection of non-characteristic course resistant to sulphonamide therapy. Some authors recommend the term of primary atypical pneumonia to differentiate it from infections of a different aetiology and atypical course [3]. The third term used in literature is “infections with atypical pathogens”: [4]; the majority of them are caused by Chlamydia, Mycoplasma and Legionella.

Moreover, mixed infections are observed. The severe course, symmetric lesions with pleural involvement and enlarged hilar lymph nodes are characteristic of such infections.

The aim of the present study was to assess the incidence of pulmonary infections with atypical bacteria (Legionalla pneumonie, Chlamydia pneumonie, Mycoplasma pneumonie)  and Pneumocistis jiroveci fungi and their effect on the course of treatment of patients after cardiac surgery.


The retrospective analysis covered the period between 1.01.2004 and 15.03.2007. The indication for collection of bronchial tree secretion to detect the presence of atypical and opportunistic pathogens was prolonged respiratory failure resistant to typical therapy. Moreover, the duration of mechanical lung ventilation, necessary reintubation and tracheostomy, other complications, length of stay in the postoperative therapy unit and of hospitalization as well as mortality and further outcomes of patients immediately after discharge were assessed.

Microbiological diagnostics was carried out using immunofluorescence assays. In direct immunofluorescence, the specimens are exposed to conjugates containing specific antibodies. The fluorochromium-labelled antibody binds with the antigen and the resultant complex is identified under a fluorescent microscope; the presence of elementary bodies is verified at 1000x magnification. Indirect immunofluorescence is used to examine the humoral response in the individual classes of immunoglobulins: IgA, IgM, IgG. During the first stage, the unlabelled antibody binds with the antigen. The second stage involves binding of the complex with the labelled antibody. Both methods have their merits and demerits. The first one is more specific and quicker. The indirect method visualizes bigger molecules, easier to detect under the microscope. The result is interpreted as positive if 10 or more round elementary bodies are visible in the specimen. In practice, 5-9 elementary bodies in the specimen are considered as the positive result. The prerequisite of reliable results is short storage of the material, which should be delivered to the laboratory immediately after collection.

The quantitative data were presented as a mean and standard deviation whereas qualitative ones as numbers and percentages. The data were collected and processed using Office software with Microsoft Excel spreadsheet.


In the period analysed, 5026 cardiac surgical procedures were performed in adults in the Silesian Centre for Heart Diseases in Zabrze. The mean EuroSCORE (the European System for Cardiac Operative Risk Evaluation) was 5.5 pts (Table 1). Patients undergoing coronary artery bypass grafting (CABG) constituted 51% of the population (Table 2); 37% of patients had advanced circulatory pathologies (NYHA III or IV); 10 patients were diabetic and another 10 had concomitant respiratory diseases (Table 3).

Postoperative respiratory complications developed in 196 (3.9%) patients - required  mechanical lung ventilation for more than 24 h and respiratory dysfunctions due to atelectasis, oedema, pleural fluid or haematoma, pulmonary infiltrations, pulmonary oedema, persistent haemostasis or ARDS. Tracheostomy was performed in 12 patients. Re-intubation, sometimes repeated, was required in 59.3% of patients [6].  

The presence of atypical and opportunistic pathogens in the bronchial tree secretion was detected in 27 patients (16%), 9 females and 18 males.  

Pneumocistis jiroveci was most commonly isolated (Table 4). No Legionella pneumonie was cultured.  On average, the pathogens were isolated on the 9th postoperative day. In 8 patients, the bronchial tree secretion showed also some other bacteria (mixed infection) (Table 5).

The mean concentration of procalcytonin on collection day was 2.83 ng mL-1. The radiographic picture of the lungs was normal in 7 patients; in the remaining 20, various, non-characteristic pathologic features (inflammatory infiltrations, atelectasis, parenchymal densities, pleural fluid, haemostasis) were observed. The infection was accompanied by abundant secretion in 6 (22.2%) cases; body temperature above 38oC was found in 4 (14.8%) patients. In 25 patients, FIO2 during lung ventilation on the infection diagnosis day was 38%, SpO2 – 86 mmHg. Prolonged sedation was administered to15 patients (55.6%). The mean lung ventilation time was 253 h; 16 patients required re-intubation, tracheostomy was performed in 12 patients. Postoperative complications unrelated to infections are presented in Table 6.

The mean length of stay in the postoperative therapy unit was 17.5 days, of hospitalization – 24 days. Four patients were discharged home, 15 were sent to other centres for further hospitalization, 3 – to intensive therapy units. The mortality rates in the population examined was 18.5% (5 patients).


Epidemiological studies demonstrated that pneumonia was the sixth cause of death in the population of North America and the fourth one in Japan [5]. Ventilator-associated pneumonia (VAP) is defined as an infection developing during mechanical ventilation due to respiratory insufficiency. This type of infection should be differentiated from severe environmental and hospital pneumonia. The majority of VAP cases are caused by typical pathogens, although some atypical and commensal bacteria are also involved [6]. Legionella pneumonie is an extremely rare cause of VAP. Patients undergoing mechanical lung ventilation are not exposed to tap water or aerosols formed during showering. Intubation is protective. The causative factor may be ice cubes for sucking to reduce the thirst.

Mycoplasmatic pneumonia can result from bacterial carrier state in the nasopharynx. The incidence of such pneumonias seems to be understated. Low detectability is likely to be related to problems with identification of the microorganism. A similar situation is observed in the case of Chlamydia pneumonie infections. In the present study, pathogens were identified using the immunofluorescence assay, which detects the presence of inclusion bodies in the infected cells of tissue culture. The Canadian researches, who analysed the causes of hospital-acquired infections in 135 consecutive patients, demonstrated Mycoplazma in one patient and Chlamydia in another one [7]. Opportunistic and atypical pneumonia cases during mechanical ventilation affect mainly patients with impaired immunity. Therefore, special attention should be paid to elderly, wasted patients and those with the incompetent immune system. The complicated postoperative course and increased catabolism with the breakdown of proteins promote destabilization of the immune system.

Atypical pathogens can cause mild, medium or severe pneumonia [4].  In our study, the general status of some patients was satisfactory, no radiographic pulmonary lesions were observed. The symptoms suggestive of widening the diagnostics included prolonged weaning from mechanical lung ventilation, airway secretion, and persistent weakness. In extremely severe forms, atypical pneumonia symptoms (Chlamydia pneumonie, Pneumocistis carinii) were confirmed on autopsy.

Bacteria colonizing the upper airways are involved in the aetiology of pneumonia in patients after tracheostomy. Another risk factor of VAP is long-term sedation and resuscitation [6]. In our study, one female patient who had sudden cardiac arrest immediately after admission to the postoperative therapy unit, developed acute respiratory failure complicated by the mixed infection. The atypical aetiology was confirmed on the 14th day of therapy. The patient died due to multi-organ failure, lungs included. ARDS is a recognized risk factor of VAP [8].

The explicit relation between serological indices of infection with atypical bacteria and acute CNS vascular incidents was not demonstrated; however, there are preliminary reports implicating such a relation [9]. Long-term atypical pneumonia may be associated with the risk of stroke or transient ischemic attack (TIA). The presence of IgG antibodies against atypical pathogens are likely to reflect recurrent infections and be related to the cerebral vascular disease. The hypothesis mentioned requires further large-scale prospective studies. In our analysis, the diagnosis of nervous system-related complications preceded the diagnosis of atypical pneumonia. If the hypothesis of bacterial origin of CNS vascular lesions is assumed, it may be supposed that patients with neurological complications were carriers of atypical bacteria and had such infections in the past, which could have increased the susceptibility of the brain to injuring stimuli.

There is a strict relation between severe forms of pneumocystis pneumonia and long-term steroid therapy [10]. Patients, who developed acute respiratory failure during lung infections, were characterized by high early mortality rates (66%). It was demonstrated that poor prognosis was correlated with high APACHE scores on admission, delayed intubation, length of mechanical ventilation and pleural oedema. Therefore, mandatory prophylactic examinations for Pneumocistis pneumonia were introduced in patients receiving: >16 mg of prednisone a day for more than 2 months, cytostatics and immunosuppressants. In the present study, only one patient was on long-term steroid therapy yet the dose was much lower than the one mentioned above. Moreover, in the population studied, delayed intubation was not a risk factor. It seems that the population of patients undergoing cardiac surgery differs from the population admitted with symptoms of severe pneumonia, thus, risk factors cannot be compared between such diverse groups.

Furthermore, the theory suggesting severe clinical course in patients with mixed infections does not work. The likely reason is that in the postoperative therapy unit, the secretion for bacteriological examinations is collected from patients undergoing mechanical lung ventilation before all characteristic features of pneumonia have developed. Thanks to that, once the infection develops the targeted antibiotic therapy is immediately instituted. In such cases, atypical pneumonia overlaps with the effectively treated infection of different aetiology. In the present study, only one patient with mixed infection died. It seems, however, that the bad outcome resulted from sudden cardiac arrest and ARDS in the early postoperative period.


1. Atypical and opportunistic pathogens are likely to cause pulmonary infections in the early postoperative period.

2. Prolonged respiratory failure and difficulties in weaning from mechanical ventilation should suggest possible atypical infections and lead to widened diagnostic procedures.

3. The generally accepted risk factors of such infections do not have to be present in cardiac patients in the early postoperative period.



1.   Prokopowicz D: Zagrozenia oportunistyczne. Wyd.: Ekonomia iSrodowisko. Bialystok 2005.

2.   Nitsch-Osuch A, Wardyn KA, Choroszy-Król I: Zakazenia wywolane patogenami atypowymi wlekarskiej. Wyd.: Górnicki. Wroclaw 2007.

3.   Park DR: The microbiology of ventilator-associated pneumonia. Resp Care 2005; 50: 742-765.

4.  Fields BS, Benson RF, Besser RE: Legionella and Legionnaires disease: 25 years of investigation. Clin Micribiol Rev 2002; 15: 506-526.

5.  Jani K, Metha NJ: Pneumocistis carinii pneumonia presenting as aof unknown origin in patient without AIDS. Heart and Lung 2002; 31: 50-52.

6.   Blasi F: Atypical pathogenes and respiratory tract infections. Eur Respir J 2004; 24: 171-181.

7.   Ngeh J, Goodbourn C: Chlamydia pneumonie, Mycoplasma pneumonie and Legionella pneumophila in elderly patients with stroke (C-PEPS, M-PEPS, L-PEPS): acase-control study on the infectious burden of atypical respiratory pathogens in elderly patients with acute cerebrovascular disease. Stroke 2005; 36: 259-265

8.   Mikaelsson L, Jacobsson G, Andersson R: Pneumocistis pneumonia – aretrospective study 1991-2001 in Gothenburg, Sweden. J Inf 2006; 53: 260-265

9.   Festic E, Gajic O, Limper AH, Aksamit TR: Acute respiratory failure due to Pneumocystis pneumonia in patients without Human Immunodeficiency Virus infection: outcome and associated features. Chest 2005; 128: 573-579.

10. Ho JC, Ooi GC, Mok TY, Chan JW, Hung I, Lam B, Wong PC, Li PC, Ho PL, Lam WK, Ng ChK, Ip MS, Lai KN, Chan-Yeung M, Tsang KW: High-dose pulse versus nonpulse corticosteroid regimens in severe acute respiratory syndrome. Am J Respir Crit Care Med 2003; 168: 1449-1456.



*Jacek Puzio

Oddzial Kliniczny Kardioanestezji iTerapii
Slaskiego Centrum Chorób Serca
ul. Szpitalna 2, 41-800 Zabrze
tel.: 0-32237 24

Received: 25.10.2008.
Accepted: 12.01.2009.