Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

A randomized study assessing the systematic search for maxillary sinusitis in nasotracheally mechanically ventilated patients. Influence of nosocomial maxillary sinusitis on the occurrence of ventilator-associated pneumonia.

Holzapfel L, Chastang C, Demingeon G, et al.

Am J Respir Crit Care Med 1999; 159:695-701. [abstract] [full-text for subscribers]

Reviewed by Kshama Daphtary, MD and Kathleen Meert, MD, Children's Hospital of Michigan.

Review posted July 9, 1999

---

I. What is being studied?:

The study objective:

To determine whether a systematic search for and subsequent treatment of nosocomial sinusitis decreases the occurrence of ventilator-associated pneumonia (VAP) in nasotracheally intubated patients.

The study design:

Prospective, randomized, controlled study.

The patients included:

1. Age > 15 years
2. Nasotracheal intubation
3. Anticipated duration of mechanical ventilation > 7 days

The patients excluded:

1. Pregnancy
2. Tracheotomy
3. Bleeding disorder ( platelet count < 30,000/mm3, prothrombin time < 30%, activated partial thromboplastin time two times higher than the normal value )
4. Therapeutic doses of heparin
5. Selective digestive tract decontamination

The interventions compared:

The study group had sinus CT performed at days 4 and 8 after intubation and every 7 days thereafter if the body temperature was > 38°C. The required duration of fever prior to CT scan was not mentioned. If the CT scan showed an air-fluid level and/or an opacification within the maxillary sinus and if the patient had a body temperature of > 38°C, a transnasal puncture was performed. The aspirated material was examined macroscopically and subjected to aerobic and anaerobic cultures. A drain was inserted in all patients who had a transnasal puncture. Sinusitis was diagnosed if a) sinus CT scan findings were consistent with sinusitis (air-fluid level and/or opacification), b) mechanical ventilation had been ongoing for at least 48 hours, c) macroscopically, the sinus aspirate was purulent, d) quantitative culture of the aspirated material > 1000 cfu/ml. Treatment of sinusitis consisted of intravenous antibiotics adjusted to the susceptibility of the cultured organism and sinus lavage performed every 8 hours.

The control group did not undergo sinus CT scan or puncture. Control patients may have received antibiotics for what the authors refer to as "nondocumented infection" or fever without a source.

The outcomes evaluated:

The main outcome was the occurrence rate of VAP. Criteria for VAP were:

1. a new infiltrate on chest radiograph consistent with pneumonia,
2. rectal temperature of > 38.4°C or < 36°C,
3. leukocyte count of > 12,000 cells/mm3 or < 5,000 cells/mm3 and/or grossly purulent tracheobronchial secretions,
4. a specimen obtained using a fiberoptic bronchoscope protected specimen brush from the involved pulmonary lobe with > 1000 cfu/ml.

The other two outcomes evaluated were:

1. Occurrence rate of nosocomial septicemia, which was defined as two separate blood cultures yielding the same organism or as a single blood culture associated with tachycardia > 90/min, fever > 38.4 °C or hypothermia < 36°C.
2. Overall survival.

All the outcomes were calculated from the date of randomization.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes, patients were allocated to either the study group or the control group by randomization in blocks of four. A computer program was used to provide randomization.

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Was followup complete?

Yes, follow-up was complete. Patients who were discharged from the ICU or who died without developing the event were censored at the time of their discharge or death. This means that patients who were discharged or died were included in the analysis (Kaplan-Meier procedure) but only up until the time of discharge or death.

Were patients analyzed in the groups to which they were randomized?

Yes, patients were analyzed in the groups to which they were randomized.

Secondary questions:

3. Were patients, health workers, and study personnel "blind" to treatment?

No. The design of this study did not permit "blinding" of patients, health workers or study personnel to the intervention.

4. Were the groups similar at the start of the trial?

The two groups were similar at the start of the trial with respect to age, sex, Simplified Acute Physiologic Score, proportion of patients with chronic obstructive pulmonary disease, cardiac insufficiency and immunodepression. Both groups had similar percentages of patients with admitting diagnosis of coma. However, the study group tended to have a higher percentage of patients with pneumonia (24% vs. 16%) and infection (14% vs. 7%) and a lower percentage of surgical patients (15% vs. 20%) and patients with multiple trauma (2% vs. 9%) at the time of admission than the control group

5. Aside from the experimental intervention, were the groups treated equally?

Apart from the experimental intervention, both groups received treatment for VAP and septicemia with antibiotics based on the susceptibility of the cultured organism. Patients in both groups also received antibiotics for other reasons including what the authors refer to as "nondocumented infection". The authors do not describe how nondocumented infections were defined or diagnosed.

The number of patients who received antimicrobial agents in each group before the onset of sinusitis and/or VAP is not mentioned. Since the study group tended to have a larger number of patients admitted with infection and pneumonia compared to the control group, a larger number of patients in the study group received antimicrobial agents at the onset of the study. This could interfere with the results of the study and could independently explain the lower incidence of VAP and nosocomial septicemia in the study group. Exposure to antibiotics has been found to confer protection against VAP and respiratory tract infections and what we may be seeing in this study is the "prophylactic antibiotic effect" in the study group who received more antibiotics given to treat pneumonia and infection present at the time of admission. (1, 2)

The study group received antibiotics for a documented nosocomial infection more frequently (146 antibiotic prescriptions vs. 95, p = 0.03) and for a longer duration (8.1 + 10.5 days vs. 5.4 + 8.5 days, p = 0.008) than the control group. Presumably the difference in treatment rates is due to the finding of sinusitis, and does not represent an independent difference of care between the two groups.

III. What were the results?

1. How large was the treatment effect?

VAP
Between group comparison:
37 patients in the study group (1- month Kaplan- Meier estimate, 34%) and 51 patients in the control group (1- month Kaplan- Meier estimate, 47%) developed VAP (p=0.02, log-rank test). The RR is 0.61. We do not have access to the raw data or number of censored cases, therefore we could not confirm the calculations of RR reported by the authors.

Within group comparison:
VAP developed in 37 study group patients. VAP occurred after sinusitis in only 11 of these. VAP occurred before the diagnosis of sinusitis in 7 patients, on the same day in 5 patients and in 14 patients who never developed sinusitis. In the study group, considering sinusitis as a time-dependent event, Cox's model found that sinusitis increased the risk of VAP by 1.69 (p = 0.17).

Nosocomial septicemia
Between group comparison:
36 patients in the study group (1-month Kaplan-Meier estimate, 30%) and 46 patients in the control group (1-month Kaplan-Meier estimate, 35%) developed nosocomial septicemia (p = 0.11, log rank test). The RR is 0.71.

Within group comparison: Considering sinusitis as a time-dependent event, Cox's model found that sinusitis increased the risk of nosocomial septicemia by 2.29.

Mortality
Mortality at the end of 2 months was estimated to be 36% in the study group and 46% in the control group (p = 0.03, log-rank test). RR = 0.71.

2. How precise was the estimate of the treatment effect?

Between group analysis:

RR	95% CI
VAP	0.40 - 0.93
Nosocomial septicemia	0.46 - 1.09
Mortality	0.52 - 0.97

Using Cox's model considering sinusitis as a time-dependent event (within group analysis):

RR	95% CI
VAP	0.80 - 3.56
Nosocomial septicemia	1.10 - 4.74

IV. Will the results help me in caring for my patients?

1. Can the results be applied to my patient care?

No. This study was carried out on adults with a mean age of 60.8 + 17 years. These adult patients had different underlying diseases than would be expected for typical pediatric ICU patients. Specifically, a larger number of patients with cardiac disease and chronic obstructive pulmonary disease were included.

In our ICU most patients are nasotracheally intubated. Results would be even less applicable for units where orotracheal intubation is routine.

2. Were all clinically important outcomes considered?

Yes, all the clinically important outcomes were considered. However, no data on risks associated with transporting the patients for CT scan or sinus puncture was provided.

3. Are the likely treatment benefits worth the potential harms and costs?

The authors conclude that the occurrence of VAP in nasotracheally intubated patients can be prevented by a systematic search and treatment of nosocomial sinusitis. However, the results do not support this conclusion. The confidence interval for the relative risk for VAP in patients developing sinusitis includes 1 and is therefore not significant.

Mortality was found to be lower in the study than the control group. The authors postulate that the lower mortality was due to lower rate of VAP. However, the cause of death for these patients is not provided and we are left to assume that mortality was related to infection in all cases.

The risks to obtain CT scans, sinus punctures and lavage were not evaluated. Without obvious benefit, it does not appear that a systematic search for sinusitis and subsequent treatment would outweigh the risks of these procedures.

References

1. Liberati A, D'Amico R, Pifferi S, et al. Antibiotic prophylaxis for respiratory tract infections in adult patients in intensive care units (Cochrane Review). In: The Cochrane Library, Issue 4, 1998. Oxford: Update Software. [abstract]
2. Cook DJ, Walter SD, Cook RJ, et al. Incidence of and risk factors for ventilator-associated pneumonia in critically ill patients. Ann Intern Med. 1998; 129:433-440. [abstract]

Comments

Back to the EB Journal Club Index