Key Words: Ventilator-associated pneumonia, Endotracheal tube, Colonisation, Mechanical ventilation, Antibiotic susceptibility.

Nosocomial pneumonia (NP) is a major concern in the management of patients who require ventilatory support. NP contributes to the morbidity and mortality in mechanically ventilated patients ^1-5 . A knowledge of the risk factors and common causative organisms can help in decreasing the incidence of NP.

This study aimed to determine the incidence, risk factors, mortality, and causative organisms of NP in patients with endotracheal (ET) tube (with or without mechanical ventilation).

Materials and Methods

This prospective study was conducted over a period of six months in the Paediatric Intensive Care Unit (PICU) of a tertiary care teaching hospital. Patients less than one month, trauma cases and surgical cases were not included as they are not admitted to our PICU.

All patients with ET intubation were included in the study. Those with prepneumonia were excluded from the study. The patients had a portex ET tube inserted and removed using strict aseptic technique. The common indications for ET tube insertion were for - maintenance of airway patency, respiratory failure and mechanical ventilation (MV), elective ventilation, etc. The frequency of suctioning of the ET tube was every two hourly or more frequently if required and all aseptic precautions were followed while suctioning. Patients were extubated as per the clinical need.

After extubation, the tip of the ET tube was cut with a sterile blade and sent in a sterile tube for aerobic bacterial culture. We did not study the role of viruses, fungi, and anaerobic organisms in the development of NP. The qualitative method used for culture included incubation in glucose broth for 4 hours followed by smear examination and culture on sheep blood agar and Mac Conkey's agar. The culture plates were incubated overnight and examined for growth. Organisms were identified on the basis of colony characteristics and biochemical reactions. The antibiotic susceptibility pattern of organisms was determined using Kirby-Bauer method by disk diffusion technique and the results were interpreted as per the National Committee for Clinical Laboratory Standards guidelines ⁶ .

'Colonization' of ET tube was defined as isolation of organisms from the tip of the ET tube. 'Nosocomial Pneumonia' was diagnosed when all of the following 4 criteria were met ⁷  -

  Following risk factors were studied to evaluate their significance in developing NP - age, sex, immunocompromised status (protein energy malnutrition-grade III/IV by IAP classification and HIV positive cases), altered sensorium, duration of MV and duration of stay in the PICU. All patients kept in the PICU for 3 days or lesser were followed up subsequently for 1 week (after discharge from the PICU) to look for development of NP. The chest radiograph and other investigations were repeated as indicated by the clinical profile of each patient. Chi Square test ⁸  was applied for studying the significance of these risk factors in causing NP. The probability value ('p' value) was obtained using standard charts and considered significant if below 0.05.

Results

In the study, 69 patients (49 males : 20 females) had an ET tube inserted. The mean age was 2.95 years (range - 1 month to 11 years). The average duration of stay in the PICU was 7.23 days per patient (range - 1 to 39 days; median - 5 days). Fiftyout of 69 patients underwent MV via the ET tube. The average duration of MV was 3.675 days per patient (range - 8 hours to 15 days; median - 2 days). The total number of ET tubes inserted was 88 (in 69 patients). The ET tube was changed once in 11 patients and twice in 4 patients and lasted for an average of 3.46 days (range - 12 hours to 15 days; median - 3 days). The average duration of change of each ET tube was 2.713 days.

ET tube colonisation with bacteria was seen in 70 out of 88 ET tube tips cultured. Eighteen ET tube tip cultures did not show any growth. The total number of organisms isolated from 70 ET tube tips was 96. Two organisms were isolated from 26 ET tube tips.

NP developed in 19 out of 59 patients with MV (32.20%; 8.92/100 days of MV; 3.65/100 patient-days). NP developed in 19 out of the 69 patients with ET intubation (27.54%; 7.96/100 days of ET intubation). All these 19 patients had undergone MV. Patients with ET tube without MV did not develop NP. The incidence of NP was significantly higher in intubated patients with MV as compared to those who were not ventilated.

Of the 10 patients who were intubated but not ventilated, 2 had ET tube colonisation (E. coli & Klebsiella grown from 1 ET tube each) and 8 were not colonized. None of these developed NP.

Age, sex, altered sensorium and immunocompromised status (including 3 HIV positive cases) did not increase the incidence of NP. MV for more than 48 hours and a PICU stay of more than 3 days significantly increased the incidence of NP (Table 1).

Nine out of 19 mechanically ventilated patients with NP died (mortality - 47.37%) and 18 out of 40 mechanically ventilated patients without NP died (mortality - 45%). The difference was not statistically significant and mortality was not found to be higher in patients with NP as compared to those without it.

The various organisms isolated from ET tube tips and their antibiotic susceptibility pattern is given in Table 2. The organisms commonly isolated were E. coli (34.4%), Klebsiella (30.2%), Pseudomonas (11.5%), Proteus (11.5%), and Acinetobacter (5.2%). Other organisms isolated included 2 isolates each of Enterobacter and Citrobacter, and 1 isolate each of coagulase negative Staphylococci, Non-lactose fermentors and Salmonella. E. coli and Klebsiella had maximum susceptibility to amikacin and cefotaxime.

Discussion

Endotracheal (ET) intubation is frequently done in critically ill paediatric patients as an emergency or elective procedure. Pneumonia developing after ET intubation with MV is known despite meticulous care of the ET tube. An infective, highly viscous and adhesive layer is formed inside the ET tube as determined by factors like surface properties of the ET tube, humidity of ventilatory gases, mode of ventilation and nursing care ⁹ . ET tube can contribute to the pathogenesis of NP when the infective particles of this adhesive layer are dislodged into the lower airways by shear forces imparted by the respiratory gas flow during MV ⁹ . Colonisation of the ET tube occurs frequently in the PICU but all patients do not develop NP. Colonisation was seen in 70 out of 88 ET tube tips (79.54%) in our study. This was higher than that (51.8%) reported by Albert S et al ¹⁰ .

The incidence of NP in patients with ET intubation was 27.54% (7.96/100 days of ET intubation, 8.92/100 days of MV) in our study. This was lower than the study by Albert S et al (45%) ¹⁰  and Ruiz-Santana S et al (38%) ¹¹ . Other studies have reported an incidence varying from 9.6% to 28 .9% ^12-15 .

The incidence of NP was significantly higher in intubated patients who were mechanically ventilated as compared to those who were not ventilated. Patients without MV did not develop NP. Similar findings have been reported previously ^4,12,16 .

Factors like age, sex, immunocompromised status and altered sensorium did not affect the incidence of NP in our study. Mosconi P et al ¹⁷ , Joshi N et al ¹⁸  and George DL ¹⁹  have suggested that an altered mental status or depressed level of consciousness with impaired airway reflexes predispose the patients to aspiration of gastric or oropharyngeal secretions, increasing the risk of NP in mechanically ventilated patients.

Our study showed a significantly higher incidence of NP in patients with MV for more than 48 hours. Similar results were shown by Torres A et al ⁶  and Mosconi P et al ¹⁷ . Fagon JY et a1 ²⁰  showed an acturial risk of NP to be 6.5% at 10 days, 19% at 20 days and 28% at 30 days of MV. The risk of development of ventilator-associated pneumonia is said to be 1-3% per day of MV ^19,20 . Thus, longer the duration of MV, higher is the risk of NP. Prolonged stay in PICU (more than 3 days) significantly increased the risk of acquiring NP in our study and patients with a PICU stay of 3 days or lesser did not develop NP. This finding was similar to that previously reported ^12,14,18 .

The    mortality in cases with NP was 47.37% in our study and has ranged from 33% to 71% in various studies ^{5,6,11,18,20,21} . It was not significantly higher than the patients without NP as previously reported ^{6,17,18,20,21} .

The microbial flora associated with NP reflects the common organisms present in gut, oropharynx and the environment i.e. the gram negative organisms ^3,4,6,22-24 . E. coli was the commonest organism isolated by us. Most of the previous studies had isolated Pseudomonas aeruginosa as the commonest organism ^{3,5,8,10,20,25} . Pseudomonas was the third common organism in the present study. Others have reported Staphylococcus aureus as an important aetiological agent of NP ^{3,6,10,13,20,25} . The knowledge of the common organisms and their antibiotic susceptibility pattern is important for institution of appropriate antimicrobial therapy.

We could not study the effect of antibiotics and H2 -blockers on development of NP. We were unable to employ sophisticated techniques like bronchoscopic aspiration, bronchoalveolar lavage and protected specimen brushings ^{3,10,19,20,25,26}  for the diagnosis of NP. These studies have better accuracy in diagnosing the aetiological agent of NP but have the disadvantages of being costly and requiring special equipment and expertise.

Conclusions

Mechanically ventilated patients have a higher risk of developing NP. The duration of MV of more than 48 hours and a PICU stay of more than 3 days significantly increased the risk of developing NP in the present study. Mortality was not significantly higher in patients with NP as compared to those without NP. E. coli and Klebsiella were the common organisms isolated from the ET tube tip cultures with maximum susceptibility to amikacin and cefotaxime.

Acknowledgement

The authors thank the Dean, Seth G. S. Medical College and K. E. M. Hospital for granting permission to publish this article.

References