|Year : 2014 | Volume
| Issue : 3 | Page : 294-300
Diagnosis of ventilator-associated pneumonia: Comparison between ante-mortem and post-mortem cultures in trauma patients
S Lalwani1, P Mathur2, V Tak2, S Janani1, SI Kumar1, R Bagla1, MC Misra3
1 Departments of Forensic Medicine, All India Institute of Medical Sciences, New Delhi, India
2 Departments of Laboratory Medicine, All India Institute of Medical Sciences, New Delhi, India
3 Department of Surgical Disciplines, AIIMS, New Delhi, India
|Date of Submission||04-Apr-2013|
|Date of Acceptance||10-Aug-2013|
|Date of Web Publication||10-Jul-2014|
Departments of Laboratory Medicine, All India Institute of Medical Sciences, New Delhi
Source of Support: None, Conflict of Interest: None
Purpose: To evaluate the diagnostic utility of ante-mortem tracheal aspirates for diagnosis of ventilator-associated pneumonia (VAP). Trauma victims represent an otherwise healthy population, who are on multiple invasive life-support devices, which predispose them to severe infections like VAP. The diagnosis of VAP is challenging, due to the difficulty in obtaining a representative sample from lungs. We studied the diagnostic utility of tracheal aspirates by comparing its results with the post-mortem lung cultures. Materials and Methods: A total of 106 fatal trauma patients were included in the study. Lung samples and cardiac blood were taken for culture at the time of autopsy. The results of ante-mortem and post-mortem cultures were compared. Results: Septicemia was the cause of death in 51 (48%) of the fatal cases and VAP was identified in 36 (34%) cases. A total of 96 (90.5%) cases had pathogens isolated from lung samples. In 62 (58%) cases, the same organism was isolated from ante-mortem and post-mortem respiratory samples. Conclusions: Culture results of a properly collected tracheal aspirate should be taken into consideration along with Centre for Disease Control and Prevention (CDC's) diagnostic criteria to maximise the diagnosis of VAP.
Keywords: Autopsy, Blood stream infections, intensive care unit, ventilator-associated pneumonia, trauma
|How to cite this article:|
Lalwani S, Mathur P, Tak V, Janani S, Kumar S I, Bagla R, Misra M C. Diagnosis of ventilator-associated pneumonia: Comparison between ante-mortem and post-mortem cultures in trauma patients. Indian J Med Microbiol 2014;32:294-300
|How to cite this URL:|
Lalwani S, Mathur P, Tak V, Janani S, Kumar S I, Bagla R, Misra M C. Diagnosis of ventilator-associated pneumonia: Comparison between ante-mortem and post-mortem cultures in trauma patients. Indian J Med Microbiol [serial online] 2014 [cited 2019 Sep 16];32:294-300. Available from: http://www.ijmm.org/text.asp?2014/32/3/294/136572
| ~ Introduction|| |
Trauma contributes to a significant proportion of mortality and morbidity in the age group of 15-45 years. Trauma victims are usually middle aged males in their economically productive age. Immediate mortality in trauma patients is usually due to post-trauma complications like severe organ injury and shock. Hospital-acquired infections are the predominant contributors of fatality in patients who survive the initial 5 days of trauma. ,, Majority of trauma patients on ventilators develop pneumonia if the ventilators are used for more than 5 days. Ventilator-associated pneumonia (VAP) is reported to contribute to a significant proportion of mortality in trauma patients. Despite advances in techniques for the management of ventilator-dependent patients and the availability of effective decontamination procedures for respiratory equipment, VAP complicates the course of 8-28% of the patients receiving mechanical ventilation. 
In ventilated patients, there is a strong suspicion of VAP. Diagnosis of VAP is challenging and therefore, the Centres for Disease Control and prevention (CDC), USA has suggested a multimodal approach to its diagnosis. This includes the radiological, clinical and microbiological features. The most commonly sent sample for ante-mortem diagnosis of VAP is tracheal aspirate for culture, which may signify tracheal colonisation rather than true lower respiratory infection. , Inappropriately collected tracheal aspirates are often contaminated with upper respiratory nosocomial colonisers, which often yield three types of organisms and are therefore not processed further. In the present era of multidrug resistance, selecting an appropriate antimicrobial to treat truly infected patients, at the same time restricting antimicrobial use in patients who only have respiratory cultures positive without any other evidence of VAP is a challenging task.
Lungs represent the ideal specimen for diagnosis of lower respiratory tract infections, but obtaining lung samples is not feasible due to their highly invasive nature.  Autopsy-based studies are difficult due to the problem of obtaining consent and ethical considerations.  Since trauma cases are usually medico-legal, an autopsy is performed in all fatal trauma cases as a routine procedure. In this study, we report the profile of infections in trauma patients, who had a fatal outcome. The study is based on ante-mortem clinical findings and culture results and post-mortem culture results. We hypothesised that this study would enable us to evaluate the diagnostic utility of ante-mortem tracheal aspirates for predicting infection of lungs. The study would also help us to understand the aetiological profile of VAP in a situation where underlying risk factors like extremes of ages, diabetes and other morbid conditions are usually absent, since trauma victims are usually otherwise healthy.
| ~ Materials and Methods|| |
The study was conducted at the Jai Prakash Narain Apex (JPNA) Trauma Centre, All India Institute of Medical Sciences (AIIMS), New Delhi, India for a period of 6 months (October 2011 to March 2012). The study involved trauma victims who were admitted to the intensive care units (ICUs) and were on ventilators ante-mortem for at least 48 hours prior to death. All patients who died after 48 hours of their hospital admission with the above criteria were included in the study. An informed consent was taken from the immediate close relative (legally authorised representative) of the patients. Samples were taken during post-mortem only after taking informed consent from them. The study was approved by the Institute's ethical committee.
At the outset, we did not know which patient will eventually develop VAP/have a fatal outcome. Also, we did not know whether the organism/s isolated from ante-mortem tracheal aspirates would be isolated from lung samples, thus substantiating the proof that they actually caused VAP and possibly fatal outcome of the patient. Thus, many of the variables were uncertain. Since this study involved a continuum of follow-up of all ventilated patients till their final outcome (discharge/death), the methodology is divided into ante-mortem and post-mortem periods.
Ante-mortem follow-up of ventilated patients
Definition and surveillance methodology: VAP was defined in our surveillance system according to the CDC's diagnostic algorithm [Figure 1].  For detecting VAP, a daily follow-up was done on a performa, where all the variables [Figure 1] were noted for all ventilated patients. The radiological report was provided by the radiologist and all clinical indices were filled by clinicians. The surveillance cultures of tracheal aspirates were as a routine done in all ventilated patients on Mondays and Thursday, apart from peripheral blood sample culture when the patient was febrile.
The processing and reporting of samples was done as per standard microbiological methods. , The tracheal aspirates were cultured onto blood and MacConkey agar by the quantitative method. A report was generated only if the counts of pathogens exceed >10 5 colony forming unit (CFU)/ml from tracheal aspirate.  However, the diagnosis of VAP was done on the basis of CDC's defined criteria. 
In order to ascertain which pathogen might be responsible for VAP and may be isolated from lung samples if the patient had a fatal outcome (and only for the purpose of this study), for 6 months, all the microbial isolates (including three types or insignificant counts) isolated from tracheal aspirates of all the patients were identified by the VITEK-2 system. The identity of these pathogens from a sample was kept confidential in an electronic work sheet for future reference. The report of ≥ three types of organism or a count <10 5 CFU/ml from tracheal aspirates was not dispatched. Reports were sent only when one or two pathogen was isolated in a significant count. Thus, a complete record of the flora of tracheal aspirate of each patient over the 6-month period was maintained. All the strains were also stocked at -70°C for future reference. The strains of patients who were discharged were discarded. The strains from fatal cases were processed by the Vitek 2 system for identification and susceptibility testing and were further preserved. Apart from the total characterisation of respiratory isolates, the culture reports from other samples of all patients were also recorded. This enabled us to ascertain the source of bacteraemia in case the patient had a fatal outcome and the cardiac blood culture came positive post-mortem.
The detailed records of all patients were entered in an electronic format. The information included the detailed post-trauma admission history, number of days on mechanical ventilation, number of vascular catheter days, number of urinary catheter days, presence of acute respiratory distress syndrome, clinical symptoms, microbiological and radiological findings, surgical procedures, history of blood transfusion and cause of death (if the patient had a fatal outcome).
If any patient had a fatal outcome, the hospital infection control (HIC) nurse was informed by the floor nurse, who also contacted the forensic specialist for arranging the autopsy. All the ante-mortem records of this patient were then used for comparison with the post-mortem records.
The autopsy was done as per standard protocols. The post-mortem examinations were performed within 1-12 hours of death in all except three cases, where the autopsy was performed between 24-48 hours after death. Prior to the performance of autopsy, the bodies were kept in cold rooms (4°C) to avoid the possibility of contamination.
Collection of lung sample
For the purpose of this study, a small piece of lung (2-3 g) was collected from the base and apex of both lungs using a sterile scalpel and scissors.  The sample was immediately put in sterile, wide-mouth, screw capped vial and transported to the laboratory.
Collection of cardiac blood
As soon as the chest cavity was opened, approximately 8-10 ml of cardiac blood was aseptically aspirated with a sterile needle and syringe.  A part of this was inoculated immediately in BacT Alert blood culture bottles (Biomerieux Ltd, France) and the remaining was directly cultured onto the blood, Mac Conkey and chocolate agar plates with sterile precautions.
Contamination of the samples was prevented by observing the following precautions: early refrigeration and restriction of movements of the body was done after death (to prevent passive recirculation of blood from contaminated areas and thus decrease the potential for false-positive blood cultures); post-mortem was performed at the earliest in order to minimise any bacterial overgrowth; blood was taken from heart using sterile syringe and needle. Before sampling, the skin of the chest was thoroughly cleaned with an alcoholic-iodine preparation. The surface of the heart was sterilised using a heated broad bladed spatula and a sterile needle and syringe was used to collect the sample. For taking sample of the lung tissue, the surface of the organ in an area of 2 × 2 cm 2 was sterilised using a heated broad bladed spatula and then sterile scissors were used to obtain samples for culture. The samples were sent to the laboratory without any delay.
All the samples obtained at autopsy were cultured onto blood, MacConkey and chocolate agar and processed according to standard microbiological methods. , All microorganism isolated from the sample were identified by the Vitek 2 system (Biomerieux, France) using identification (ID) Gram positive (GP)/Gram negative (GN)/yeast (YST) cards. Antimicrobial susceptibility was also done by the Vitek 2 system.
Correlation of ante- and post-mortem data
The post-mortem growth from lung samples and cardiac blood was correlated with the ante-mortem culture reports (of 7 days prior to the date of death). This was done by retrieving the data from the electronic worksheet, wherein the detailed identification (genus, species, antibiogram) of all pathogens isolated from ante-mortem cultures were recorded. These reports were compared with the culture and susceptibility results of post-mortem specimens. Pathogens isolated from lung/cardiac blood were considered significant, except when post-mortem samples yielded common skin contaminants (e.g. diphtheroids, Bacillus sp., Propionibacterium sp., coagulase-negative Staphylococci or Micrococci) or >/= three types of bacteria, which were considered as contaminated samples. Organisms were defined as identical if the genus and species identification and antibiogram of the organisms were similar.
| ~ Results|| |
A total of 114 cases fulfilling the inclusion criteria as defined above were autopsied. Informed consent was obtained from a total of 111 fatal cases, which were included in the study. Of these, five cases were excluded, since their specimens (lung/cardiac blood) were contaminated on culture (as defined above). Thus, a total of 106 cases were included in the final analysis. The age of the patients ranged from 9 months to 82 years (average 37.4 years). A total of 92 (87%) of them were males. The length of stay of the 106 cases included in this study ranged from 3 to 174 days (average 22 days). The clinical cause of deaths in these cases was septicaemia in 51 (48%), severe head injury in 38 (36%), haemorrhagic shock in 7 (6.6%), spinal injury in 5 (5%), acute renal failure in 2 (2%) and polytrauma, abdominal injuries and chest injuries in 1 each. [Table 1] describes the ante-mortem and post-mortem culture reports of the cases as compared with their ante-mortem clinical picture. In 62 (58%) cases (S.No 2 + 3 + 4 in [Table 1]), the same organism was isolated from ante-mortem and post-mortem respiratory samples. Of these, 10 cases had the same organism from ante-mortem blood and cardiac blood also (S.No. 2); 39 had the same organisms from ante-mortem and post-mortem respiratory samples and from cardiac blood (S.No. 3), signifying that there was significant respiratory tract infection in these patients, which may have contributed to their fatal outcome.
|Table 1: Ante - mortem and post - mortem culture results of respiratory and blood specimens correlated with clinical presentation (n=106) |
Click here to view
Of these 62 cases, in 31 (50%), the ante-mortem respiratory sample immediate prior to death had grown three types of organisms, one or two of which were found in the post-mortem samples. Of these 31 cases, in 7 (22.5%), the pathogen was found only in post-mortem lung tissue specimens, whereas in the remaining 24 cases, the same pathogen/s were found in lung tissue specimens as well as cardiac blood. Of these 31 cases, in 17 (55%) cases, two of the three pathogens were recovered from post-mortem lung tissue samples and in the remaining 14 (45%) cases, only one of the three pathogens were recovered. Of the 24 cases whose cardiac blood also grew the same pathogen, in 19 (79%), only one of the three organisms was isolated, whereas in 5 (21%), two of the three organisms were recovered from cardiac blood. In one patient, the ante-mortem and post-mortem respiratory specimens grew the same pathogen (Acinetobacter baumannii), whereas the ante-mortem and post-mortem blood grew Klebsiella. pneumoniae (represented in S.No. 4 of [Table 1]). On review of the records, the patient's CVP tip sample had also grown K. pneumoniae, which could have been the source of blood stream infection. In two patients, the source of pathogens in blood (ante-mortem and post-mortem) was traced to wound infection and in one patient to urinary tract infection. The respiratory tract specimens of these patients were sterile (S.No 8 in [Table 1]). [Table 2] describes the microbial aetiology in the 10 cases that had the same growth in all the ante-mortem and post-mortem specimens.
|Table 2: Aetiology in cases that had the same organisms in ante- and post - mortem respiratory and blood specimens (10 cases) |
Click here to view
Of the 96 patients who had pathogens isolated from lung tissues, only 36 (37.5%) had been clinically diagnosed as VAP, based on CDC's diagnostic algorithm [Table 1]. Thus, of the 106 fatal cases, 36 (34%) had clinically defined VAP, all of whom had pathogens isolated from lungs.
| ~ Discussion|| |
VAP is a diagnostic dilemma for clinicians and microbiologists. The best sample representative of VAP for microbiological diagnosis is still being debated in large-scale studies. It is for this reason that the CDC has recommended an algorithm for diagnosis of VAP, which begins with a suggestive chest X-ray finding. 
In our study, we found a high proportion of patients who had the same organisms in the post-mortem lung tissue specimens, as were recovered from ante-mortem tracheal aspirates. However, a large number of these patients (50%) had tracheal cultures, which were reported as ≥ three types, that is considered as contaminated samples. In other words, a report of the infecting organism/s was dispatched along with its antimicrobial susceptibility result in only the remaining half of the 62 cases, which had growth in the lungs. This is a common problem with tracheal aspirates, which are often contaminated with upper respiratory flora and therefore reported as >three types or as respiratory colonisers.
Jourdain et al., in their study, observed that tracheal aspirates have a high sensitivity but low specificity for diagnosis of VAP.  In a study by Delclaux et al., a count of <10 3 CFU/ml in a sample obtained blindly by passing a plugged telescopic catheter through the endotracheal tube evolved to pneumonia within 2-6 days in the majority of patients. The colonising microorganisms were the same as those that caused subsequent pneumonia.  Since lungs represent the actual site of infection, lung biopsy remains the gold standard, although it is highly invasive and not recommended for routine diagnosis. , A properly collected lung sample at autopsy therefore represents the ideal sample for establishing the microbial aetiology of pneumonia. 
In contrast to the study of Kurtin et al., who conducted cultures of cardiac blood in 50 autopsies,  the positivity rates in our study was much higher. This may be because a large number of patients actually had a clinical suspicion of VAP/septicaemia in our study. Also, most of the autopsy-based studies in the literature are more than four decades old, and do not reflect the current diagnostic dilemmas of hospital-acquired infections. ,,,,, As discussed in a recent excellent review on the role of post-mortem studies in aetiological diagnosis of pneumonia, the ante-mortem diagnosis of pneumonia compromises several factors, the primary one being the inability to acquire a suitable representative sample.  Therefore, we feel that the pathogens isolated from lung samples of fatal trauma cases in our study represent actual infections of lungs, which were missed ante-mortem.
In our study, only 37.5% of patients who had pathogens in lungs were diagnosed as VAP on the basis of CDC's diagnostic algorithm. Based on our findings, we feel that a properly collected tracheal aspirate may actually represent pneumonia, if the clinical and radiological findings are also suggestive.
Studies based on autopsies are often challenged on the grounds of possibility of contamination of samples. However, Morris et al. in a review of published studies on autopsies observed that proper storage of bodies prevents translocation of pathogens and a properly conducted autopsy reduces the risk of contamination of samples.  Since samples from the apex and base from both the lungs were taken in our study, we feel that an adequate sampling of the actual site of infection was done by us.
A major lacuna of our study was that we could not perform molecular typing of the ante-mortem and post-mortem strains to ascertain their similarity and therefore the true causal association as cause of death.
We also did not assess the quality of respiratory tract specimens for rejection of inappropriate sample. Although a semi-quantitative/quantitative culture of bronchoscopic alveolar lavage (BAL)/tracheal aspirate is recommended by the CDC for aiding the diagnosis of VAP, a few studies have found that for BAL, a total cell count be performed to assess adequacy and a differential count to assess cellularity. For quality assessment, the percentages of squamous and bronchial epithelial cells may be used to predict heavy upper respiratory contamination, with more than 1% of the total cells being proposed as a rejection criterion, even if only a few studies have directly assessed this point.  Modified Giemsa staining may also be done, as it offers a number of advantages over Gram staining, including better visualisation of host cell morphology, improved detection of bacteria, particularly intracellular bacteria and detection of some protozoan and fungal pathogens. ,
We could also not ascertain the confounding effect of antimicrobial treatment on post-mortem culture results. Since we found a large number of patients who had culture positivity with the same species, we postulate that VAP in our patients had a rapidly deteriorating course, despite initiation of appropriate culture-result-based treatment.
Since trauma victims are predominantly middle aged, otherwise healthy males, infections in these patients may truly represent a hospital acquisition.
To conclude, culture results of a properly conducted tracheal aspirate should be taken into consideration along with the CDC's diagnostic criteria to maximise the diagnosis of VAP and institute appropriate and timely antimicrobial treatment for this potentially life threatening infection.
| ~ Acknowledgement|| |
The electronic surveillance of hospital-acquired infections at our centre is being funded by a research grant from the Indian Council of Medical Research.
| ~ References|| |
|1.||Pories SE, Gamelli RL, Mead PB, Goodwin G, Harris F, Vacek P. The epidemiologic features of nosocomial infections in patients with trauma. Arch Surg 1991;126:97-9. |
|2.||Caplan ES, Hoyt NJ. Identification and treatment of infections in multiply traumatized patients. Am J Med 1985;79 (suppl 1A):68-76. |
|3.||Mathur P. Infection in Traumatized Patients: A growing medico-surgical concern: Indian J Med Microbiol 2008;26:212-6. |
|4.||Chastre J, Fagon JY. Ventilator-associated Pneumonia. Am J Respir Crit Care Med 2002;165:867-903. |
|5.||Mayhall CG. Ventilator-Associated Pneumonia or Not? Contemporary Diagnosis. Emerg Infects Dis 2001;7:200-4. |
|6.||Turner GD, Bunthi C, Wonodi C. The role of post-mortem studies in pneumonia etiology research. Clin Infect Dis 2012;54 (Suppl 2):S159-64. |
|7.||7. CDC. National Healthcare Safety Network. Device Associated Infections Module. Protocols and Instructions. 2010. Available from: www.cdc.gov/nhsn/psc-da.html. and 7. http://www.cdc.gov/nhsn/library.html [Last accessed date 14 Aug 2013 |
|8.||Collee JG, Diguid JP, Fraser AG. Mackie and Mc Cartney practical Medical Microbiology: 14 th ed. Edinburgh, UK: Churchill Livingstone; 1996. |
|9.||Forbes BA, Sahun DF, Weissfeld AS. Bailey and Scott's diagnostic microbiology: 10 th ed. St. Louis, Missouri, USA: Mosby; 1998. |
|10.||Porzecanski I, Bowton DL. Diagnosis and treatment of ventilator-associated pneumonia. Chest 2006;130:597-604. |
|11.||Vij K. Textbook of Forensic Medicine and Toxicology: 4 th ed. New Delhi, India: Elsevier; 2008. |
|12.||Jourdain B, Novara A, Joly Guillou ML, Dombret MC, Calvat S, Trouillet JL, et al. Role of quantitative cultures of endotracheal aspirates in the diagnosis of nosocomial pneumonia. Am J Respir Crit Care Med 1995;152:241-6. |
|13.||Delclaux C, Roupie E, Blot F, Brochard L, Lemaire F, Brun-Buisson C. Lower respiratory tract colonization and infection during severe acute respiratory distress syndrome. Incidence and diagnosis. Am J Respir Crit Care Med 1997;156:1092-8. |
|14.||Chintu C, Mudenda V, Lucas S. Lung diseases at necropsy in African children dying from respiratory illnesses: A descriptive necropsy study. Lancet 2002;360:985-90. |
|15.||Kurtin JJ. Studies in autopsy bacteriology. Am J Clin Pathol 1958;30:239-43. |
|16.||Carpenter HM, Wilkins RM. Autopsy bacteriology: Review of 2033 cases. Arch Pathol 1964;77:73-81. |
|17.||Wood WH, Oldstone M, Schultz RB. A re-evaluation of blood culture as an autopsy procedure. Am J Clin Pathol 1965;43:241-7. |
|18.||O'Toole WF, Saxena HM, Golden A, Ritts RE. Studies of post-mortem microbiology using sterile autopsy technique. Arch Pathol 1965;80:540-7. |
|19.||Koneman EW, Minckler TM, Shires DB, De Jongh DS. Postmortem microbiology II. Selection of cases for culture. Am J Clin Pathol 1971;55:17-23. |
|20.||Dolan CT, Brown AL, Ritts RE. Microbiological examination of post-mortem tissues. Arch Pathol 1971;92:206-11. |
|21.||Morris JA, Harrison LM, Partridge SM. Postmortem bacteriology: A re-evaluation. J Clin Pathol 2006;59:1-9. |
|22.||Fagon JY, Chastre J, Domart Y, Trouillet JL, Pierre J, Darne C, et al. Nosocomial pneumonia in patients receiving continued mechanical ventilation: Prospective analysis of 52 episodes with use of protected specimen brush and quantitative culture techniques. Am Rev Respir Dis 1989;139:877-884". |
|23.||Meduri GU, Mauldin GL, Wunderink RG, Leeper KV Jr, Jones CB, Tolley E, et al. Causes of fever and pulmonary densities in patients with clinical manifestations of ventilator-associated pneumonia. Chest 1994;106:221-35. |
[Table 1], [Table 2]