Journal of Clinical Epidemiology 66 (2013) 62e66

Longer duration of mechanical ventilation was found to be associatedwith ventilator-associated pneumonia in children aged 1 month

to 12 years in India

Shally Awasthia,*, Mohammad Tahazzula, Ankur Ambasta, Yogesh C. Govila, Amita JainbaDepartment of Pediatrics, King George’s Medical University, Shahmina Road, Lucknow 226003, Uttar Pradesh, India

bDepartment of Microbiology, King George’s Medical University, Shahmina Road, Lucknow 226003, Uttar Pradesh, India

Accepted 18 June 2012

Abstract

Objectives: To determine primarily (1) the incidence of ventilator-associated pneumonia (VAP) among ventilated patients aged 1month to 12 years and secondarily (2) the risk factors for VAP and (3) common organisms causing VAP.

Study Design and Setting: Prospective study in a tertiary care center in India. Consecutive ventilated patients aged �1 month and �12years and requiring mechanical ventilation (MV) for �48 hours were included after written informed parental consent. For the diagnosis ofVAP, National Nosocomial Infections Surveillance System criteria of 1996 were used.

Results: Incidence of VAP among patients aged 1 month to 12 years was 36.2% (38/105; 95% confidence interval [CI]: 27, 46). Inunconditional logistic regression analysis controlling for the presence of underlying illnesses, risk factor for VAP was O4 days of MV(adjusted odds ratio, 3.76; 95% CI: 1.41, 10.02; P5 0.008). Reintubation within 72 hours of extubation and more than two attendantsat the time of recruitment showed increased tendency for the development of VAP but did not reach statistical significance. Endotrachealand endobronchial aspirates were positive for organism in 19.05% (20/105) and 37.14% (39/105) of patients, respectively.

Conclusion: Almost one-third of ventilated patients develop VAP. Vigilance for the development of VAP has to be kept on thoserequiring O4 days of MV. Klebsiella and Staphylococcus aureus were common bacterial isolates in such patients. � 2013 ElsevierInc. All rights reserved.

Keywords: Ventilator-associated pneumonia; Intensive care unit; Risk factors; Attendants; Bacterial isolates

1. Introduction

Ventilator-associated pneumonia (VAP) continues tocomplicate the course of 9e27% of patients receiving me-chanical ventilation (MV) [1]. The attributable mortalityrate for VAP ranges from 24% to 50% [2] and can reach76% [3] in some specific settings or when lung infectionis caused by high-risk pathogens. Etiologic organismswidely differ according to the population of the patientsin an intensive care unit (ICU), duration of hospital stay,and prior antimicrobial therapy [4].

Although described as an important cause of mortalityand morbidity among the patients in ICU, there is paucityof data on the incidence, risk factors, and microbiologyof the causative organism of VAP in developing countriessuch as India. Also, there is a need to know the organisms

* Corresponding author. Tel.: þ91-522-2328463.

E-mail address: shallya@rediffmail.com (S. Awasthi).

0895-4356/$ - see front matter � 2013 Elsevier Inc. All rights reserved.

http://dx.doi.org/10.1016/j.jclinepi.2012.06.006

causing VAP in a particular ICU, which may be different indifferent ICUs. Knowledge of these factors will help in theprevention and management of VAP in specific settings.Therefore, the present study was undertaken to determineprimarily (1) the incidence of VAP among patients aged 1month to 12 years and secondarily (2) the risk factors forVAP among such patients and (3) common organisms caus-ing VAP.

2. Materials and methods

This was a prospective hospital-based study. The studywas conducted between September 2008 and August2009 in the pediatric ventilatory units of the Departmentsof Pediatrics and Microbiology, Chhatrapati Shahuji Maha-raj Medical University, Lucknow, after approval from theinstitutional ethics committee. This is a tertiary care teach-ing hospital with two ventilator units, one with six ventila-tors and the other with three ventilators.

Table 1. Primary diagnosis of patients put on mechanical ventilationand incidence of VAP

Primary diagnosis Total N VAP, n (%), P[ 0.55

Septicemia 21 9 (42.86)Neurologic 48 19 (39.58)Cardiac 12 3 (25.00)Cancer 7 1 (14.28)Othera 17 6 (35.30)Total 105 38 (36.20)

Abbreviation: VAP, ventilator-associated pneumonia.a Includes acute renal failure, tetanus, diarrhea with dehydration

with shock, and dengue shock syndrome.

63S. Awasthi et al. / Journal of Clinical Epidemiology 66 (2013) 62e66

Consecutive ventilated patients aged �1 month and �12years and requiring MV for �48 hours were included afterwritten informed consent and followed for the duration ofMV. No one refused to give consent for enrollment. Forthe diagnosis of VAP, National Nosocomial Infections Sur-veillance (NNIS) System criteria of 1996 were used [5].According to these criteria, VAP is diagnosed whena new and persistent pulmonary infiltrate (unexplained) ap-pears on the chest radiograph after 48 hours of MVand anythree of the following criteria are positive: (1) temperatureinstability (fever with temperature O38�C or hypothermiawith temperature !35�C); (2) leukopenia (total leukocytecount !4,000/mm3) or leukocytosis (total leukocyte countO10,000/mm3); (3) positive endobronchial aspirate and/orpositive endotracheal aspirate culture for bacteria; (4)wheezing, rales, or rhonchi; (5) bradycardia (heart rate!60 beats per minute) or tachycardia (heart rateO120 beatsperminute); and (6) increased ventilator demand or increasedsupplemental oxygen requirement (PaO2/FIO2! 240, inwhich FIO2 stands for fraction of oxygen delivered byventilator).

All patients were ventilated by an orotracheal tube,which was changed only if blocked or displaced. Patientswere ventilated in synchronized intermittent mandatoryventilation mode using either a reusable or disposable ven-tilator circuit with a heated humidification system. Ventila-tor circuits were changed (if disposable) or sterilized (ifreusable) after every 1-week use for each patient. An openmethod was used for suctioning of secretions. Frequency ofendotracheal tube suction depended on the amount of secre-tions, and it varied from patient to patient. Patients wereventilated in supine position with frequent changing to rightlateral and left lateral decubitus positions. An orogastrictube was inserted in all patients. No prophylactic topicaloropharyngeal antibiotic and selective digestive tract de-contamination was done in any of the patients. Temperaturewas recorded 4 hourly by clinical thermometer from the ax-illa. Patients were clinically assessed for heart rate, respira-tory rate, temperature, and blood pressure (both systolicand diastolic) 4 hourly. At the time of recruitment, endotra-cheal and endobronchial aspirates were taken. The endotra-cheal aspirate was obtained for microbiological cultureafter hand washing, and while wearing sterile gloves. Endo-tracheal secretions were collected by instilling 5e10 mL ofsterile normal saline through an infant feeding tube inserted30 cm into the endotracheal tubes and then aspirating witha sterile 10-mL syringe. The endobronchial aspirate wascollected by instilling 5e10 mL of normal saline throughan infant feeding tube and aspirating back by using a mucusextractor. One end of the mucus extractor was connected tothe infant feeding tube and the other end to an open suctionpump. Both specimens were immediately transported to thelaboratory and processed according to the standard protocol[6]. All the bacterial isolates were identified to species levelby standard biochemical tests, and their antibiotic suscepti-bility testing was performed by the KirbyeBauer disk

diffusion method on MuellereHinton agar as per the guide-lines of the Clinical and Laboratory Standards Institute [7].

Data were collected on demographic variables such asage, sex, primary diagnosis, and indication for MV. Datawere also collected on risk factors such as duration ofMV, number of reintubations within 72 hours of extubation,number of endotracheal suctions done in the last 6 hours,number of attendants at the time of recruitment, type ofventilator circuit used (disposable or reusable), history ofprior hospitalization, and breach in continuity of skin.

Sample size was calculated to assess VAP incidence of30% with a precision of 9% and an alpha level of 95%.Hundred patients were to be recruited for this purpose.

Data were collected in a preformed questionnaire andentered into an Excel file (Microsoft, Redmond, WA). Uni-variate analysis was done to assess the distribution of base-line variables. Thereafter, Student t-test and chi-square testwere used to compare continuous and categorical variables,respectively, between patients with and without VAP. Un-conditional logistic regression (LR) analysis was done con-trolling for the presence of underlying illnesses to find therisk factors associated with VAP among those that had uni-variate association with VAPwith P-value!0.1. We also re-ported the adjusted odds ratio (OR) with 95% confidenceinterval (CI) and P-value.

3. Results

From September 2008 to August 2009, 105 patientswere enrolled of which 72.4% (n5 76) were males.Table 1 shows the primary diagnosis of patients enrolledin the study, which was similar in those who developedor did not develop VAP. Using NNIS System criteria(1996) [5], we found the incidence of VAP among patientsaged 1 month to 12 years who are on MV to be 36.2%(n5 38; 95% CI: 27, 46). Table 2 shows the results of uni-variate analysis for association of various risk factors withVAP. In univariate analysis, patients who required O4 daysof MV had increased chance to develop VAP (OR, 2.81;95% CI: 1.15, 6.97; P5 0.012).

Table 3 reports the findings of unconditional LR analysisto find the association of various risk factors with VAP, con-trolled for the underlying diagnosis, using cancer as thecomparator because it had the lowest frequency count. In

Table 2. Univariate analysis to assess the association of various categorical risk factors with VAP

Categorical risk factors Coding TotalNumber of patients who

developed VAP (n) Crude OR (95% CI) P-value

Use of reusable ventilator tubes Yes 95 35 1.36 (0.29, 7.16) 0.7No 10 3

History of prior hospitalization Yes 41 13 0.72 (0.65, 1.20) 0.4No 64 25

Breach in continuity of skin Yes 5 3 2.79 (0.35, 25.24) 0.25No 100 35

Age !36 mo Yes 51 18 0.93 (0.39, 2.22) 0.85No 54 20

Reintubation within 72 hr of extubation Yes 38 18 2.12 (0.86, 5.24) 0.07No 67 20

Number of endotracheal suctions done in last 6 hr O1 Yes 16 9 2.66 (0.80, 8.96) 0.06No 89 29

Number of attendants at the time of recruitment O2 Yes 54 24 2.11 (0.87, 5.20) 0.07No 51 14

Duration of mechanical ventilation O4 d Yes 44 22 2.81 (1.15, 6.97) 0.012No 61 16

Abbreviations: VAP, ventilator-associated pneumonia; OR, odds ratio; CI, confidence interval.

64 S. Awasthi et al. / Journal of Clinical Epidemiology 66 (2013) 62e66

the LR model, the only factor significantly associated withVAP was MV for more than 4 days (adjusted OR, 3.76;95% CI: 1.41, 10.02; P5 0.008). An increased tendencyfor the development of VAP was seen in patients with morethan two attendants at the time of recruitment, and in pa-tients reintubated within 72 hours of extubation, but thesedid not achieve statistical significance.

Serious adverse outcome (death or refusal for furthertreatment) was found in 30 of 38 patients with VAP(78.9%), of which 21 expired and 9 refused for furthertreatment, and in 49 of 67 patients without VAP (73.1%),of which 28 expired and 21 refused for further treatment.The patients who refused further treatment were moribund.Mean duration of hospital stay in patients who developedand did not develop VAP was 10.07 days (standard devia-tion [SD], 6.44) and 10.20 days (SD, 6.67), respectively(P5 0.922).

Table 3. Unconditional logistic regression to assess risk factors forVAP

VariablesAdjusted oddsratios (95% CI) P-value

Underlying diagnosis (cancer asthe reference)Cardiac 0.62 (0.14, 2.73) 0.53Others 0.76 (0.21, 2.70) 0.67Septicemia 1.90 (0.58, 6.23) 0.29Neurologic 0.95 (0.37, 2.43) 0.92Reintubation within 72 hr ofextubation

2.65 (0.97, 7.24) 0.05

Number of endotrachealsuctions done in last 6 hr O1

3.39 (0.89, 12.87) 0.07

Number of attendants at thetime of recruitment O2

2.50 (0.94, 6.65) 0.06

Duration of mechanicalventilation O4 d

3.76 (1.41, 10.02) 0.008

Abbreviations: VAP, ventilator-associated pneumonia; CI, confi-dence interval.

We report the distribution of organisms isolated from theendotracheal and endobronchial aspirate culture in Table 4.

4. Discussion

In the present study, we used the NNIS System criteria(1996) for the diagnosis of VAP [5], which include clinicalobservation, ventilator settings, and noninvasive investiga-tions such as chest x-rays and endotracheal and endobron-chial cultures for bacteria. We found the incidence of VAPin patients aged 1 month to 12 years who were on MV to be36.2% (38/105; 95% CI: 27, 46). Our findings agree withtwo studies conducted in India that reported the incidenceof VAP in children to be 33.3% [8] and 45.4% [9]. In West-ern studies, the incidence of VAP ranged from 28% to 67%[10e13].

We assessed the association between VAP and its knownrisk factors namely duration of MV, type of ventilatory tube(disposable or reusable), history of prior hospitalization,breach in continuity of skin, younger age, and frequencyof endotracheal suction and reintubation. We found no as-sociation between the primary diagnosis and the incidenceof VAP, as reported earlier by Apostolopoulou et al. [14].

Table 4. Distribution of organisms isolated from the endotracheal andendobronchial aspirate culture

OrganismEndotracheal culture,

n (%), N[ 20Endobronchial culture,

n (%), N[ 39

Klebsiella 6 (30.00) 17 (43.59)Staphylococcus aureus 5 (25.00) 8 (20.51)Candida 3 (15.00) 2 (5.13)Bacillus cereus 2 (10.00) 1 (2.56)Pseudomonas

aeruginosa1 (5.00) 5 (12.82)

Escherichia coli 1 (5.00) 4 (10.26)Enterococcus 2 (10.00) 0 (0.00)Acinetobacter 0 (0.00) 2 (5.13)

65S. Awasthi et al. / Journal of Clinical Epidemiology 66 (2013) 62e66

On univariate as well as LR model analysis, we foundthatO4 days of MV was a significant risk factor for the de-velopment of VAP. Similar findings were reported by others[18e23]. Hence, efforts must be made to keep the durationof MV as short as possible. Those requiring more than oneendotracheal suction in 6 hours may need to be monitoredwith greater vigilance for the development of VAP. Furtherstudies are needed in this area.

We also found that reintubation after extubation had in-creased tendency for the development of VAP in the LRmodel (adjusted OR, 2.65; 95% CI: 0.97, 7.24; P5 0.05).A similar finding was reported by Bauer et al. [15] who rec-ommended that odds of failurebe considered before extuba-tion. Routine periodic change of endotracheal tube shouldbe discouraged.

Because a greater proportion of patients who developedVAP had serious adverse outcomes (death or refusal for fur-ther treatment) as compared with those who did not developVAP, the former had shorter total length of hospital stay ascompared with the latter, though statistically insignificant.Similar findings of increased mortality in VAP patientswere reported by others [9,24,25].

Common bacterial isolates from endotracheal and endo-bronchial aspirates were Klebsiella and Staphylococcus au-reus as reported by others [20,26e28]. However, we alsoisolated Pseudomonas and Candida from these aspirates,and both these organisms may be of nosocomial origin[17,29,30].

This was a hospital-based study to assess primarily theincidence and secondarily the risk factors and etiologicagents of VAP. The basis of sample size calculation wasto get precision around the incidence of VAP; therefore,the sample size was inadequate to study the associationof various risk factors with VAP. We had included a wideage range of patients, which is a weakness of this study.However, we had assessed the association of the numberof attendants in ICU with VAP, which has possibly not beendone in any other study so far.

With increasing number of intensive care setups comingup in developing as well as developed countries, there is anurgent need to prevent known complications of MV such asVAP as well as identify cost-effective strategies for thetreatment. A number of studies have been conducted to as-sess the cost-effectiveness of various protocols for treatingVAP [16,28e30], but only few are randomized controlledtrials [31]. We recommend conducting randomized con-trolled trials to look for measures to minimize VAP andto assess their cost-effectiveness.

5. Conclusion

Almost one-third of ventilated patients develop VAP.Because patients who require O4 days of MV are at in-creased risk for VAP, they have to be monitored with vigi-lance for its development. Bacteria were cultured from

endobronchial aspirates of more than one-third of patientswith VAP. Klebsiella and S aureus were the most commonbacterial isolates in such patients.

Acknowledgments

Prof G.G. Agarwal assisted in statistical analysis anddata interpretation.

Contributors: Shally Awasthi was responsible for thedesign and supervision of data collection, analysis, inter-pretation, and manuscript writing and is the guarantor.Mohammad Tahazzul was responsible for data collection,data entry, analysis, and manuscript writing. Late YogeshC. Govil was responsible for data interpretation and manu-script writing. Amita Jain and Ankur Ambast were respon-sible for microbiology laboratory work, data interpretation,and manuscript writing.

Role of funding sources: This was the MD thesis of Mo-hammad Tahazzul. It was partly supported by UPCST grantto Prof Shally Awasthi.

References

[1] American Thoracic Society. Guidelines for the management of adults

with hospital-acquired, ventilator associated and healthcare-associated

pneumonia. Am J Respir Crit Care Med 2005;171:388e416.

[2] Erbay RH, Yalcin AN, Zencir M, Serin S, Atalay H. Costs and risk

factors for ventilator-associated pneumonia in a Turkish university

hospital’s intensive care unit: a case-control study. BMC Pulm Med

2004;4:3.

[3] Fagon JY, Chastre J, Domart Y, Trouillet JL, Pierre J, Darne C, et al.

Nosocomial pneumonia in patients receiving continuous mechanical

ventilation. Prospective analysis of 52 episodes with use of a pro-

tected specimen brush and quantitative culture techniques. Am Rev

Respir Dis 1989;139:877e84.

[4] Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Respir

Crit Care Med 2002;165:867e903.

[5] Rea-Neto A, Youssef NC, Tuche F, Brunkhorst F, Ranieri VM,

Reinhart K, et al. Diagnosis of ventilator-associated pneumonia:

a systematic review of the literature. Crit Care 2008;12:56e8.

[6] Wu CL, Yang DI, Wang NY, Kuo HT, Chen PZ. Quantitative culture

of endotracheal aspirates in the diagnosis of ventilator-associated

pneumonia in patients with treatment failure. Chest 2002;122:662e8.[7] Clinical and Laboratory Standards Institute. Performance standards

for antimicrobial disk tests; Approved standards. CLSI Document

M2-A9, Vol. 26. 9th ed. No 1. Wayne, PA: 2006.

[8] Deep A, Ghildiyal R, Kandian S, Shinker N. Clinical and microbio-

logical profile of nosocomial infections in the pediatric intensive care

unit (PICU). Indian Pediatr 2004;41(12):1238e46.

[9] Dey A, Bairy I. Incidence of multidrug resistant organisms causing

ventilator associated pneumonia in a tertiary care hospital: 9 months

prospective study. Ann Thorac Med 2007;2:52e7.

[10] Salata RA, Lederman MM, Shlaes DM, Jacobs MR, Eckstein E,

Tweardy D, et al. Diagnosis of nosocomial pneumonia in intubated,

intensive care unit patients. Am Rev Respir Dis 1987;135:426e32.

[11] Gonzalez C, Rubio M, Romero-Vivas J, Gonzalez M, Picazo JJ. Bac-

teremic pneumonia due to Staphylococcus aureus: a comparison of

disease caused by methicillin-resistant and methicillin-susceptible or-

ganisms. Clin Infect Dis 1999;29:1171e7.

[12] Chastre J, Fagon JY. Invasive diagnostic testing should be routinely

used to manage ventilated patients with suspected pneumonia. Am

J Respir Crit Care Med 1994;150:570e4.

66 S. Awasthi et al. / Journal of Clinical Epidemiology 66 (2013) 62e66

[13] Kerver AJ, Rommes JH, Mevissen-Verhage EA, Hulstaert PF, Vos A,

Verhoef J, et al. Colonization and infection in surgical intensive care

patients: a prospective study. Intensive Care Med 1987;13:347e51.

[14] Apostolopoulou E, Bakakos P, Katostaras T, Gregerakos L. Incidence

and risk factors for ventilator associated pneumonia in 4multidisciplin-

ary intensive care units in Athens, Greece. Respir Care 2003;48:

681e8.

[15] Bauer TT, Ferrer R, Angrill J, Schultze-Werninghaus G, Torres A.

Ventilator-associated pneumonia: incidence, risk factors, and micro-

biology. Semin Respir Infect 2000;15:272e9.

[16] Eagye KJ, David PN, Joseph LK. Impact of superinfection on hospi-

tal length of stay and costs in patient with ventilator associated pneu-

monia. Semin Respir Crit Care Med 2009;30(1):116e23.

[17] Mark ER, et al. Nosocomial infection in pediatric intensive care unit.

In: Derek SW, Thomas PS, Hector RW, editors. Pediatric critical care

medicine: Basic science and clinical evidence. Ed 1. London:

Springer; 2007:1473e4.

[18] Papazian L, Bregeon F, Thirion X, Gregoire R, Saux P, Denis JP, et al.

Effect of ventilator associated pneumonia on mortality and morbidity.

Am J Respir Crit Care Med 1996;154:91e7.[19] Bekele A, Donald EC, Antonio A, Marin HK, Regina S. Morbidity of

ventilator associated pneumonia in a randomized study of a silver

coated endotracheal tube. Chest 2007;132:4.

[20] Jordi R, Daniel AO, Gerry O, Montserrat VL, Lisa B, Rebecca R,

et al. Epidemiology and outcome of ventilator associated pneumonia

in a large US database. Chest 2002;122:2115e21.

[21] Heyland DK, Cook DJ, Griffith L, Keenan SP, Brun-Buisson C. The at-

tributablemorbidity andmortality of ventilator associated pneumonia in

the critically ill patient. Am JRespir Crit CareMed 1999;159:1249e56.

[22] Fagon JY, Chastre J, Domart Y, Trouillet JL, Gibert C. Mortality due

to ventilator associated pneumonia or colonization with Pseudomo-

nas or Acinetobacter species: assessment by quantitative culture of

samples obtained by a protected specimen brush. Clin Infect Dis

1996;23:538e42.

[23] Justin LR, Steven MG, Pam C, Cynthica Fatica RN, Jeffrey H, et al.

Predictors of long-term mortality in patients with ventilator associ-

ated pneumonia. Am J Med 2006;119:897.

[24] Rakshit P, Nagar US, Deshpande AK. Incidence, clinical outcome,

and risk stratification of ventilator associated pneumoniada prospec-

tive cohort study. Indian J Crit Care Med 2005;9:211e6.

[25] Ahmad SM, Choudhary J, Ahmad M, Arora V, et al. Treatment of

ventilator associated pneumonia with piperacillin-tazobactum and

amikacin vs cefepime and levofloxacin: a randomized prospective

study. Indian J Crit Care Med 2007;11:117e21.

[26] Masterton RG, Galloway A, French G, et al. Guidelines for the man-

agement of hospital-acquired pneumonia in the UK: report of the

working party on the British Society for Antimicrobial Chemother-

apy. J Antimicrob Chemother 2008;62:5e34.

[27] Masterton R, Craven D, Rello J, et al. Hospital acquired pneumonia

guidelines in Europe: a review of their status and future development.

J Antimicrob Chemother 2007;60:206e13.

[28] Niederman MS. The impact of antibiotic resistance on clinical

outcomes and the cost of care. Crit Care Med 2001;29(suppl):

N114e7.

[29] Singh N, Rogers P, Atwood CW, Wagener MM, Yu VL. Short-course

empiric antibiotic therapy for patients with pulmonary infiltrates in

the intensive care unit: a proposed solution for indiscriminate antibi-

otic prescription. Am J Respir Crit Care Med 2000;162(2 pt 1):

505e11.

[30] Hamel MB, Phillips RS, David RB, Teno J, Connors AF, Desbiens N,

et al. Outcomes and cost-effectiveness of ventilator support and ag-

gressive care of the patients with acute respiratory failure due to

pneumonia or acute distress syndrome. Am J Med 2000;109:614e20.

[31] Memish ZA, Oni GA, Djazamati W, Cunningham G, Mah MW. A

randomized clinical trial to compare effects of heat and moisture ex-

changer with a heated humidifying system on the occurrence rate of

ventilator associated pneumonia. Am J Infect Control 2001;

29(5):301e5.