Journal of Clinical Nursing 2003; 12: 484–489

A prospective, randomized study of ventilator-associated

pneumonia in patients using a closed vs. open suction system
SANDRA SALLOUM ZEITOUN MSN, RN
Doctoral Student, Clinical Nurse Specialist at the ICU, Hospital SEPACO, Nursing Department,
Federal University of Sa˜o Paulo, Sa˜o Paulo City, Brazil

ALBA LÚCIA
LUCIA BOTURA LEITE DE BARROS P h D , R N
Associate Professor, Nursing Department and Director of Nursing, Hospital Sa˜o Paulo, Federal
University of Sa˜o Paulo, Sa˜o Paulo City, Brazil

SOLANGE DICCINI P h D, RN
Adjoint Professor, Nursing Department, Federal University of Sa˜o Paulo, Sa˜o Paulo City, Brazil

Accepted for publication 17 October 2002

Summary
• The objective of this study was to verify the incidence of nosocomial
pneumonia in intubated and extended mechanically ventilated patients having
endotracheal suctioning by an open vs. closed suction method aiming to decrease
nosocomial pneumonia.
• Twenty-four (51.1%) patients received open-tracheal suction and 23 (48.9%)
received closed-tracheal suction. The inclusion criteria were: surgical and
medical patients older than 13 years, undergoing mechanical ventilation for more
than 48 hours. Additional data were gathered using the Acute Physiology and
Chronic Health Evaluation II, and details on smoking, alcoholism, diabetes
mellitus, renal failure, previous lung disease, and previous use of antibiotics,
steroids, H2 antagonists and antacids.
• Among the 24 patients having open-tracheal suction, 11 developed nosocomial
pneumonia while of the 23 patients undergoing closed-tracheal suction, seven
developed infection (P ¼ 0.278). Risk factors for nosocomial pneumonia were
not significantly different between the two groups. In the final logistical
regression model the following variables remained: groups (open and closed)
[odds ratio (OR) ¼ 0.014; confidence interval (CI) ¼ 0.001–0.416; P ¼ 0.014]
and use of prior antibiotics (OR ¼ 2.297; CI ¼ 1.244–4.242; P ¼ 0.008).
• Use of a closed suction system did not decrease the incidence of nosocomial
pneumonia when compared with the open system. The exogenous risk factors
were the most important for acquiring this infection.

Correspondence to: Sandra Salloum Zeitoun, Rua Madre de Deus,

795 ap. 22 Sa˜o Paulo, Brazil (tel.: +55 11 6605 6965; e-mail:
szeitoun@ig.com.br).

484  2003 Blackwell Publishing Ltd

Issues in clinical nursing
Keywords: adverse effects, artificial ventilation, endotracheal intubation, endo-
tracheal suction, pneumonia.
Introduction
Pneumonia is the inflammatory response of the host to the
uncontrolled multiplication of organisms invading the
distal airway (Meduri, 1993; George, 1996). The definition
‘ventilator associated pneumonia’ (VAP) is used when
pneumonia is diagnosed in an intubated, mechanically
ventilated patient after more then 48 hours of ventilation.
Early onset VAP occurs within the first 4 days of
mechanical ventilation, and late-onset VAP occurs there-
after (Bonten & Bergmans, 1999). According to data from
the National Nosocomial Infection Surveillance System of
the Centers for Disease Control (George, 1996), pneumo-
nia is the second most common nosocomial infection
overall and the most common infection in Intensive Care
Units (ICU), occurring in between six and 10 cases for
every 1000 admissions (Hayner & Baughman, 1995). In
intubated patients, the infection incidence is seven to 21
times higher compared with those not needing mechanical
ventilation (Martino, 1998). Despite the advent of new
effective broad-spectrum antimicrobials, its morbidity and
mortality remain high (Rello et al., 1992).
Risk factors for VAP include patient-related conditions,
conditions that favour colonization and bronchoaspiration,
cross-infection, respiratory therapy equipment, invasive
devices, use of antibiotics, H2 blockers, steroids and
endotracheal suction.
Several factors can contribute to the appearance of
secretions in intubated and ventilated patients. Endotra-
cheal tubes impair coughing and mucociliary clearance and
the contaminated secretions can pool above the infected
endotracheal tube cuff and are not easily removed by
suctioning (American Thoracic Society, 1995; Kollef
et al., 1999). Endotracheal suctioning is an essential part
of care for intubated patients to guarantee good ventilation
and oxygenation. However, it must be done using aseptic
technique because patients must be disconnected from the
ventilator. There is consensus that staff hands are the
main vehicles for transmission of the infection (Craven
et al., 1998).
Recently, we acquired a closed tracheal suction system.
It consists of a disposable suction catheter available in
several diameters. A plastic sheet envelops the catheter to
prevent bacterial contamination. The distal end has a
rubber stopper to allow removal of secretions when by
vacuum. The proximal tip has an irrigation port and a
T-tube with a standard connection that allows it to be
 2003 Blackwell Publishing Ltd, Journal of Clinical Nursing, 12, 484–489
A study of ventilator-associated pneumonia 485
placed between the ventilator and endotracheal tube. The
confirmed advantages of this system include maintenance
of cardiovascular parameters, because the patient is not
disconnected from the mechanical ventilator to perform
endotracheal suction, readiness for suction, and lower
costs, because only one catheter is used in each 24 hours.
It also dispenses with the use of gloves, mask and glasses
during the procedure, and decreases environmental cross-
contamination and hypoxaemia (Grossi & Santos, 1994;
Johnson et al., 1994; Kollef et al., 1999).
However, no information was available about whether
the new system would reduce infection rates. Therefore we
decided to carry out a study with the following objective:
To evaluate the incidence of nosocomial pneumonia in
intubated and ventilated patients undergoing closed
suction and open suction methods.
Methods
DESIGN AND SAMPLE
The study design was a randomized assay, with parallel
groups, conducted in an ICU in a general hospital in São
Paulo, Brazil. Approval for the study was given by the
appropriate ethics committees. Surgical and medical
patients older than 13 years, undergoing mechanical
ventilation for more than 48 hours were included in the
study. The exclusion criteria were patients who had been
intubated or had a tracheostomy performed at another
hospital; presence of pulmonary infection at the time of
admission (Deppe et al., 1990), patients diagnosed as
having AIDS or severe neutropenia (<500 polymorpho-
nuclear cells/mm3) (Brun-Buisson et al., 1998) and early
reintubation.
The VAP was defined by clinical criteria as follows:
fever (axillary temperature ‡37.8 C); radiographic
appearance of new or progressive pulmonary infiltrate,
leucocytosis (‡10.000/mm3) and purulent tracheobron-
chial secretions or change in their characteristics. All cases
were reviewed by a Clinical Evaluation Committee,
composed of two infection specialists and one intensive
care specialist, to establish whether VAP had occurred or
not. The study endpoint was 48 hours postextubation,
because in this period the risk of developing VAP still
exists (Gaynes & Horan, 1996).
Patients included in the study were divided into two
groups: those intubated on even dates received open
486 S. S. Zeitoun et al.

endotracheal suction, and those intubated on odd dates
had closed suction.
Additional data gathered were Acute Physiology and
Chronic Health Evaluation (APACHE) II scores to verify
differences in illness severity; risk factors for the devel-
opment of VAP such as smoking, alcoholism, previous
lung disease, diabetes mellitus, renal failure, and time of
use of antibiotics, H2 antagonist, antacids and steroids
prior to the development of VAP. The APACHE II is a
prognostic scoring system that can be used to predict the
risk of death for multidiagnostic ICU patient groups. It
consists of a numeric score (ranging from 0 to 71) derived
from three components: (1) points for extent of abnor-
mality of 12 physiological parameters, (2) points for
increasing age, and (3) points for chronic health abnor-
malities. The APACHE II scoring uses the worst values
for the initial ICU day. An increasing score reflects an
increased severity of disease and a higher risk of death in
hospital (Seneff et al., 1995).
Among the 24 patients having open suction, 11 (45.8%)
developed VAP, while of the 23 having closed suction,
seven (30.4%) developed VAP (P ¼ 0.278).
No statistically significant differences were found
between the two groups for smoking (P ¼ 0.450), alco-
holism (P ¼ 0.489), previous lung disease (P ¼ 1.000),
diabetes mellitus (P ¼ 0.724) and renal failure
(P ¼ 1.000) (Table 1).
When these risk factors were compared with the
development of VAP in both groups, there were no
statistically significance different (Tables 2 and 3).
Interesting results were obtained when the data on use
of steroids, antibiotics and H2 antagonist were analysed.
Although the ‘open’ group had used antibiotics and H2
antagonists for less time, this use was significant for the
development of VAP. However, in the ‘closed’ group, the
Table 1 Prevalence of risk factors for the development of ventilator
associated pneumonia (VAP) with the use of open and closed
endotracheal suction systems

DATA ANALYSIS Open system Closed system

Variable n (%) n (%) P-value
Quantitative variables and their distribution between the
groups were analysed using the Mann–Whitney U-test. Smoking 5 (20.8) 7 (30.4) 0.450
Differences between the groups in relation to dichotomous Alcoholism 0 (0) 1 (4.3) 0.489
Previous pneumopathy 5 (20.8) 4 (17.4) 1.000
variables were compared using the chi-square test, with
Diabetes mellitus 4 (16.7) 5 (21.7) 0.724
Fisher’s correction when indicated. Odds ratios (OR) with Renal failure 2 (8.3) 1 (4.3) 1.000
confidence intervals (CI) were used to measure associ-
ations between all variables, use of closed and open suction
systems and pneumonia. OR with CI were used to
Table 2 Prevalence of risk factors for the development of ventilator
measure associations between all variables, including the associated pneumonia (VAP) in the open suction group
use of closed and open suction systems and pneumonia.
The technique of multiple logistic regression was applied VAP
in order to form an assemblage of independent variables Yes (%) No (%)
Variable P-value
(exposure variable, cofactors, control variable and confu-
sion) that lead to illness (pneumonia), which was the Smoking 3 (27.3) 2 (15.4) 0.630
dependent dichotomous variable (Norman & Streine, Previous lung disease 2 (18.2) 3 (23.1) 1.000
Diabetes mellitus 2 (18.2) 2 (15.4) 1.000
1994).
Renal failure – 2 (15.4) 0.482

Results
Table 3 Prevalence of risk factors for the development of ventilator
Forty-seven patients were randomized, with 24 (51.1%) associated pneumonia (VAP) in the closed suction group
receiving open suction and 23 (48.9%) closed suction.
There was no statistically significant difference VAP
(P ¼ 0.296) when subjects were grouped by cause (car-
Variable Yes (%) No (%) P-value
diac, neurological, pulmonary, septic shock and other
causes). The APACHE II scores were not significantly Smoking 1 (14.3) 6 (37.5) 0.366
different between the two groups, and had a median of 24 Alcoholism 1 (14.3) – 0.304
for the ‘open’ group and 22 for the ‘closed’ group, showing Previous lung disease – 4 (25.0) 0.273
Diabetes mellitus 1 (14.3) 4 (25.0) 1.000
homogeneity between the groups in relation to the severity Renal failure – 1 (6.3) 1.000
of illness.

 2003 Blackwell Publishing Ltd, Journal of Clinical Nursing, 12, 484–489

Issues in clinical nursing A study of ventilator-associated pneumonia 487

Table 4 Distribution of risk factors for the development of Table 6 Final logistical regression model of independent variables
ventilator associated pneumonia (VAP) in the open suction group chosen as risk factors for the development of ventilator associated
pneumonia
VAP
IC (95%)
Variable Yes No P-value Variable Odds ratio (inferior–superior) P-value

Prior steroids 8 (5–16) 6 (3–14) 0.043 Groups* 0.014 <0.001–0.416 0.014

Prior antibiotics 9 (6–11) 5 (3–7) 0.002 Prior antibiotics 2.297 1.244–4.242 0.008
H2 antagonist 8 (3–16) 6 (3–8) 0.046
*Open and closed endotracheal suction systems.

Table 5 Distribution of risk factors for the development of

ventilator associated pneumonia (VAP) in the closed suction group
VAP
Variable Yes No
Prior steroids 7.5 (3–14) 8 (4–12)
Prior antibiotics 14 (7–16) 8 (4–11)
H2 antagonist 10 (5–16) 8 (5–12)
greater length of use of these drugs did not show a
statistically significant difference (Tables 4 and 5).
Only two patients in the group undergoing open suction
system had used antacids during their ICU stay and
neither developed VAP.
In the univariate analyses, use of prior antibiotics did
not show statistically significant differences; however, OR
tended to identify antibiotics as an exposure factor for the
development of VAP in the open group (OR ¼ 1.201;
CI ¼ 0.944–1.530; P ¼ 0.135) compared with the closed
group (OR ¼ 0.963; CI ¼ 0.802–1.156; P ¼ 0.687).
With the multivariate analyses, a group variable was
included represented by the closed suction system (inter-
vention fi 1) with reference to the open suction system
(control fi 0), as the study aimed to identify risk variables
and their association with VAP. These analyses showed
that the closed suction group has less chance of developing
VAP. In other words, patients undergoing closed suction
had a 0.014 less chance of developing VAP when receiving
prior antibiotics, which in turn increased the chance by
2.29 per day (Table 6).
Discussion
In a literature review of the last 10 years, only two studies
with similar design were found. The first was a multi-
centre study to evaluate colonization, incidence and
mortality between patients undergoing open and closed
endotracheal suction (Deppe et al., 1990). It was found
that tracheal colonization was greater in the closed
suctioned group (67%) compared with the open suction
group (39%) (P ¼ 0.02). However, there was no statis-
tically significant difference in the incidence of nosocomial
pneumonia between the closed (26%) and open groups
P-value
(29%). Survival analysis demonstrated that the probability
0.932 of survival without developing nosocomial pneumonia was
0.214 greater among closed suction patients compared with open
0.308 suction patients (P < 0.03). Nosocomial pneumonia was
diagnosed using clinical criteria by means Gram staining
and culture of sputum.
Silva (1993) carried a similar study to evaluate pul-
monary infection comparing open and closed endotracheal
suction systems using clinical diagnostic criteria. The
results showed that the incidence of pulmonary infection
with an open suction system was 84.9% and 60.8% with
closed suction. However, it was not reported whether
cases of reintubation occurred, and this is considered the
main risk factor for the development of pulmonary
infection.
Similar results were found in our study, as there were
no statistically significant differences in VAP (P ¼ 0.278)
between the open and closed suction groups. However,
the frequency of cases of VAP was greater in the open
(11 cases) compared with the closed suction group (seven
cases).
According to Craven et al. (1998), endotracheal suction
is the main route of entry of bacteria to the inferior
respiratory tract. The literature on risks and complications
associated with this technique and their prevention is vast
and focuses on therapeutic and technological advances.
Therefore, this subject is always discussed on professional
updating courses and in-service education programmes
(Sampaio, 1998). In our centre, nursing staff receive
exhaustive training in performing endotracheal suction by
open and closed methods following a protocol, in order to
make practice homogeneous.
In fact, we have observed in practice, that the anxiety to
relieve a hypersecretive patient makes staff ignore some
steps in the technique, which can increase complications
inherent to the procedure such as oxygen desaturation in

 2003 Blackwell Publishing Ltd, Journal of Clinical Nursing, 12, 484–489

488 S. S. Zeitoun et al.
patients with elevated inspired oxygen fraction, nosoco-
mial pneumonia, elevation in intracranial pressure, atelec-
tasis and haemodynamic instability (Ritz et al., 1986;
Taggart et al., 1988; Deppe et al., 1990; Crosby &
Parsons, 1992; Johnson et al., 1994).
In agreement with Crimlisk et al. (1994) and Blackwood
& Webb (1998), we emphasize the following positive
aspects of closed suction systems: guarantee of aseptic
chain in case of inadequate handwashing; catheter con-
tamination by a non-sterile object; decrease of environ-
mental and staff contamination, because of avoiding circuit
condensate aerosol and/or tracheal secretion expelled
during suctioning reaching the operator; readiness for
suction in patients with bronchial secretions because the
equipment remains attached between endotracheal tube
and mechanical ventilator; maintenance of positive end
expiratory pressure, avoiding microatelectasis; and
removal of the need to hyperoxygenate patients before
the procedure, unless oxygen desaturation be detected.
Zeitoun et al. (2000) carried out a study with 59
patients to evaluate the cost of equipment used in
open and closed suction systems and concluded that the
closed technique was showed cheaper (R$60.34) compared
with the open method (R$69.73; where R stands for ‘Real’,
the currency of Brazil) (P ¼ 0.001).
With regard to the variables (smoking, alcoholism,
previous pneumopathy, diabetes mellitus and renal failure)
analysed as risks factors for the development of VAP, and
that also define the epidemiological profile of each patient,
there were no statistically significance differences when
the prevalent of these variables in the groups were
calculated (Table 1).
When these same variables were compared with the
development of VAP in both groups, there were also no
statistically significance different (Tables 2 and 3).
Despite the harmful effects of smoking, it was not possible
to establish an association with VAP because of the small
number when the sample was stratified. For the study to
have greater power (80–90%), it should be repeated as a
multicentre study.
Medeiros (1993) carried out a study of risks factors for
the development and mortality of nosocomial pneumonia
in adult ICUs and found that alcoholism, chronic
obstructive pulmonary disease and other variables acted
as risk factors.
When the median value was calculated for the length of
use (in days) of steroids, antibiotics and H2 antagonists
prior to the development of VAP and analysed as a risk
factor for infection associated with the use of open
and closed endotracheal suction systems, it showed
statistically significant differences P ¼ 0.043, P ¼ 0.002
and P ¼ 0.046, respectively, only for patients receiving
open suction, although the length of use of these
medications was less in this group (Table 4).
The use of H2 antagonists for the prevention of stress
ulcers in critically ill patients or those having mechanical
ventilation is considered controversial. Torres et al.
(1990), studying the incidence, risk factors and prognosis
of VAP, found that this was twice as high in patients
receiving antacids and/or H2 antagonists vs. those not
receiving these (37% vs. 19%; P ¼ 0.007). However,
there was no reference to the suction method used.
The American Thoracic Society (1995) reports that in
three meta-analytic studies the use of sulcralfate, which
did not elevate gastric pH, was associated with a reduced
incidence of pneumonia, when compared with the use of a
H2 antagonist. However, the use of H2 antagonist alone
did not increase the risk of pneumonia when compared
with placebo.
These conflicting results with H2 antagonist reflect
the fact that this elevates gastric pH. Furthermore, in the
analyses described there was no consideration of the
impact of concurrent enteral feeding.
Fernandes et al. (2000a) state that use of sucralfate has
been recommended by Hospital Infection Control Prac-
tices Advisory Committee (HICPAC). Fernandes et al.
(2000b) state that sucralfate has been showed to be a
potential substitute for H2 antagonist and antacid, and has
been recommended by the HICPAC until conclusive
evidence on its effect is produced.
Chastre & Fagon (1994) describe how, in a prospective
study of 52 cases of nosocomial pneumonia diagnosed in
an ICU, the infection was directly influenced by the use of
antibiotics prior to its development, with a rate of 65% in
patients receiving antibiotic therapy against 19% in those
not receiving this. These data coincide in part with our
study, which showed, in logistic regression, that prior
antibiotic use is harmful, increasing the chance of
acquiring VAP. However, why this chance increases when
the patient has open endotracheal suction is a question
that deserves more research.
Conclusion
In this study the use of a closed tracheal suction system
did not decrease or prevent the development of VAP
when compared with an open system. However, the
study has two limitations. The first relates to the sample
size: the absence of statistically significant differences for
risk factors may be because of the small numbers that
remained when the samples were stratified. The second
limitation refers to the use of the clinical criteria for the
 2003 Blackwell Publishing Ltd, Journal of Clinical Nursing, 12, 484–489
Issues in clinical nursing
diagnosis of VAP, because of their questionable accu-
racy. We plan to extend our work through a multicentre
study and to use both clinical and microbiological
criteria.
References
American Thoracic Society (1995) Hospital-acquired pneumonia in
adults: diagnosis, assessment of severity, initial antimicrobial
therapy, and preventative strategies. A consensus statement.
American Journal of Respiratory Critical Care Medicine 513, 1711–
1725.
Blackwood B. & Webb C.H. (1998) Closed tracheal suctioning
systems and infection control in intensive care unit. Journal of
Hospital Infection 39, 315–321.
Bonten M.J.M. & Bergmans D.C.J.J. (1999) Nosocomial pneumo-
nia. In Hospital Epidemiology and Infection Control, 2nd edn
(Mayhall C.G., ed.). Lippincott Williams & Wilkins, Philadelphia,
pp. 211–238.
Brun-Buisson C., Sollet J.P., Schweich H., Brière S. & Petit C.
(1998) Treatment of ventilator-associated pneumonia with
Piperacillin-Tazobactam/Amikacin versus Ceftazidime/Amika-
cin. A multicenter, randomized controlled trial. Clinical Infectious
Diseases 26, 346–354.
Chaster J. & Fargon J.Y. (1994) Invasive diagnostic test should be
routinely used to manage ventilated patients with suspect
pneumonia. American Journal of Respiratory Critical Care
Medicine 150, 570–574.
Craven D.E., Steger K.A. & Laforce F.M. (1998) Pneumonia. In:
Hospital Infections, 4th edn. Lippincott-Raven Publishers,
Philadelphia, pp. 487–513.
Crimlisk J.T., Paris R., McGonagle E.G., Calcutt J.A. & Farber
H.W. (1994) The closed tracheal suction system: implications for
critical care nursing. Dimensions of Critical Care Nursing 13(6),
292–300.
Crosby L.J. & Parsons L.C. (1992) Cerebrovascular response of
closed head-injured patients to a standardized endotracheal tube
suctioning and manual hyperventilation procedure. Journal of
Neuroscience Nursing 24(1), 40–49.
Deppe S.A., Kelly J.W., Thoi L.L., Chudy J.H., Longfield R.N.,
Ducey J.P., Truwit C.L. & Antopol M.R. (1990) Incidence of
colonization, nosocomial pneumonia, and mortality in critically ill
patients using a Trach care closed-suction system versus an open-
suction system: prospective, randomized study. Critical Care
Medicine 18(12), 1039–1393.
Fernandes A.T., Ribeiro Filho N. & Barroso E.A.R. (2000a)
Conceito, cadeia epidemiológica das infecções hospitalares e
avaliação custo-benefı́cio das medidas de controle. In Infecça˜o
Hospitalar E Suas Interfaces Na A´rea Da Sau´de (Fernandes A.T.,
ed.). Atheneu, São Paulo, pp. 215–265.
Fernandes A.T., Zamorano P.O. & Torezan Filho M.A. (2000b)
Pneumonia hospitalar. In Infecça˜o Hospitalar E Suas Interfaces Na
A´rea Da Sau´de (Fernandes A.T., ed.). Atheneu, São Paulo,
pp. 516–555.
Gaynes R.P. & Horan T.C. (1996) Surveillance of nosocomial
infections. In Hospital Epidemiology and Infection Control
(Mayhall C.G., ed.). Lippincott Williams & Wilkins, Baltimore,
pp. 1017–1103.
 2003 Blackwell Publishing Ltd, Journal of Clinical Nursing, 12, 484–489
A study of ventilator-associated pneumonia 489
George D.L. (1996) Nosocomial pneumonia. In Hospital Epidemi-
ology and Infection Control (Mayhall C.G., ed.). Lippincott
Williams & Wilkins, Baltimore, pp. 175–195.
Grossi S.A.A. & Santos B.M.O. (1994) Prevenção da hipoxemia
durante a aspiração endotraqueal. Revista Latino-Americana de
Enfermagem 2(2), 87–101.
Hayner C.E. & Baughman R.P. (1995) Nosocomial pneumonia: a
review of diagnostic approaches. Infection Medicine 12(7), 322–330.
Johnson K.L., Kearney P.A., Johnson S.B., Niblett J.B., MacMillan
N.L. & MacClain R.E. (1994) Closed versus open endotracheal
suctioning: costs and physiologic consequences. Critical Care
Medicine 22(4), 658–666.
Kollef H.M., Skubas N.J. & Sundt T.M. (1999) A randomized
clinical trial of continuous aspiration of subglottic secretions in
cardiac surgery patients. Chest 116(5), 1339–1345.
Martino M.D.V. (1998) Infecções do trato respiratório inferior. In
Manual de Microbiologia Clı´nica Aplicada Ao Controle de Infecça˜o
Hospitalar. APECIH, São Paulo, pp. 3–10.
Medeiros E.A.S. (1993) Fatores de risco para aquisição e letalidade
de pneumonia hospitalar em adultos internados em unidade de
terapia intensiva. Doctoral Thesis, Escola Paulista de Medicina,
Universidade Federal de São Paulo, São Paulo, p. 151.
Meduri G.U. (1993) Diagnosis of ventilator associated pneumonia.
Infectious Disease Clinics of North America 7(2), 295–329.
Norman G.R. & Streine D.L. (1994) Biostatistics – the Bare
Essentials. Mosby Year Book, Inc., Canada.
Rello J., Ricart M., Ausina V., Net A. & Prats G. (1992) Pneumonia
due to haemophilus influenzae among mechanically ventilated
patients. Chest 102(5), 1562–1565.
Ritz R., Scott L.R., Coyle M.B. & Pierson D.J. (1986) Contamina-
tion of a multiple-use suction catheter in a closed-circuit system
compared to contamination of a disposable, single-use suction
catheter. Respiratory Care 31(11), 1086–1091.
Sampaio L.A.B.N. (1998) Processo ensino-aprendizagem da técnica
de aspiração endotraqueal: avaliação da implantação de um
modelo padrão em um programa de educação continuada.
Masters degree Dissertation, Escola de Enfermagem, Universi-
dade de São Paulo, São Paulo.
Seneff M.G., Zimmerman J.E. & Knaus W.A. (1995) Severity of
illness indices and outcome prediction: development and evalu-
ation. In Textbook of Critical Care, 3rd edn (Shoemaker W.C.,
Ayres S.M., Grenvik A. & Holbrook P.R., eds). W.B. Saunders,
Philadelphia, pp. 1777–1786.
Silva L.D. (1993) A infecção pulmonar no contexto da UTI: uma
contribuição para sua prevenção através do sistema de aspiração
traqueal empregado pela enfermagem. Revista de Enfermagem da
Universidade Estadual do Rio de Janeiro 1(2), 24–30.
Taggart J.A., Dorinsky N.L. & Sheahan J.S. (1988) Airway
pressures during closed system suctioning. Heart and Lung
17(5), 536–542.
Torres A., Aznar R., Gatell J.M., Jiménez P., González J., Ferrer
A., Celis R. & Roisin R.R. (1990) Incidence, risk and prognosis
factors of nosocomial pneumonia in mechanically ventilated
patients. American Review of Respiratory Disease 142, 523–528.
Zeitoun S.S., Gonçalves S.R., Barros A.L.B.L. & Diccini S. (2000)
Sistema aberto de aspiração endotraqueal versus sistema fechado
de aspiração endotraqueal: relação custo/benefı́cio e implicações
para a assistência de enfermagem. Acta Paulista de Enfermagem 13,
230–233.