Causes of Prolongation of Mechanical Ventilation
By Jun Takezawa, MD
Provision of mechanical ventilation is indicated to improve pulmonary gas exchange and/or to reduce the work of breathing in respiratory failure. Once the cause requiring ventilatory support is overcome, weaning from mechanical ventilation is to be attempted. The greatest benefit of a short duration of mechanical ventilation may be improved mortality because of less frequent development of ventilator-associated complications such as nosocomial pneumonia, barotrauma, and stress ulcer.
The duration of mechanical ventilation can be divided into two processes. The first consists of providing mechanical ventilation to establish adequate pulmonary gas exchange and/or to reduce the work of breathing. The second process is weaning or discontinuation of the ventilatory support. Several factors can be associated with the prolongation of mechanical ventilation, and they can be overviewed according to the Figure.
Figure
Duration of Ventilation and Patient and Disease Variables
Duration of Mechanical Ventilation before Weaning
The duration of mechanical ventilation can be prolonged by several factors such as the primary diagnosis for which the patient was transferred to ICU, the severity of illness, and the developement of complications during the ICU stay. Seneff and associates analyzed the APACHE III database including 17,440 patients in 42 ICUs from 40 U.S. hospitals.1 They analyzed the data to determine the contributing factors to prolongation of mechanical ventilation using multivariate regression analysis. Among the patients who required mechanical ventilation, 1195 patients required mechanical ventilation for more than seven days. Five hundred thirty-eight patients died or were discharged from ICU while still requiring mechanical ventilation. The mean duration of mechanical ventilation was 4.7 days.
Relative contribution of patient and disease variables to the prolongation of mechanical ventilation in the study by Seneff et al is shown in the Figure. Admission diagnosis and acute physiology score (APS) on ICU day 1, calculated by the APACHE III scoring system, were most strongly related to the duration of mechanical ventilation. Among admission diagnoses, pneumonia, ARDS, neuromuscular disease, postoperative head trauma, and postoperative intracranial hemorrhage were significantly related to the duration of mechanical ventilation.
The relationship between ventilator-associated pneumonia (VAP) and mortality was evaluated by Kollef and associates.2 They enrolled 277 patients from three ICUs who required mechanical ventilation for more than 24 hours. Among them, 43 patients (15.5%) developed pneumonia. To determine the factors responsible for ventilator-associated pneumonia, multivariate analysis was used. A multiple organ dysfunction syndrome (MODS) index greater than 3, age older than 60 years, prior administration of antibiotics, and supine position during the first 24 hours were significantly associated with VAP. Mortality among the patients who developed pneumonia was 37.2%, as compared to 8.5% in those without pneumonia (P < 0.01).
A similar study was reported by Fagon and colleagues.3 They recruited 48 patients who required mechanical ventilation and developed pneumonia during their stay in ICU. Case matched patients who did not develop pneumonia were used as a control. The odds ratio for mortality among patients with pneumonia was 2.0 (95% CI 1.61-2.49). Mean hospital stay in patients who developed pneumonia was 34 days and 21 days for patients without pneumonia (P < 0.02). Although these two studies did not describe the effect of VAP on duration of mechanical ventilation, it can be speculated that VAP prolongs the duration of mechanical ventilation. However, it is not clear whether the patients who subsequently developed pneumonia were already more severely ill at the time of admission to the ICU. To clarify this issue, a further study is required that takes matched severity of illness at study entry into account.
Selective decontamination of the digestive tract (SDD) has been claimed to decrease the incidence of VAP and might also decrease both the duration of mechanical ventilation and mortality. More than 20 randomized control trials (RCTs) have been performed on this issue. In a meta-analysis by Sun et al, when overall RCTs were evaluated in terms of mortality, SDD seemed to decrease mortality in patients with high severity.4 The latest RCT on SDD was reported by Quinio and associates,5 in which the patients were randomized to receive either an SDD regimen (76 patients) or a placebo (72 patients). Although the number of patients who developed bronchopneumonia was less in the SDD group, no difference was found between the groups in the incidence of other infections, ICU stay, duration of mechanical ventilation, or mortality. They noted a tendency for more Gram-positive colonization in the SDD group. However, they could lower the cost of antibiotics by 42%. From those studies, we see that it may be possible that SDD can reduce mortality in patients with a high APS. However, it may be restricted to a certain patient population, and prevalence of nosocomial infection by multi-drug resistant organisms must be taken into account before the routine use of this modality in the ICU.
Duration of Weaning
The duration of weaning can be affected by several factors, such as the use of sedatives, nursing work force, the weaning protocol used, and ventilatory modes used, including T-piece trials. Barrientos-Vega et al randomly assigned 108 patients who required mechanical ventilation for more than 24 hours to receive a continuous infusion of either midazolam (54 patients) or propofol (54 patients) as a sedative.6 Both drugs were titrated to obtain the best patient-ventilator synchrony, as evaluated by the Ramsey scoring system. All patients received adjunctive morphine at a dose of 0.5 mg/kg/day. Although the duration of mechanical ventilation with a sedative was similar between the groups, the duration from the onset of weaning to T-piece trial, and the time from T-piece trial to extubation, were significantly shorter in propofol group (48.9 vs 4.0 hours and 49.0 vs 30.8 hours, respectively). Therefore, it would appear that the use of propofol as a sedative can cut in half the duration of mechanical ventilation in comparison with midazolam. However, overall medical cost was similar between the two groups.
Kollef et al7 compared the weaning duration of protocol-directed weaning (179 patients) and physician-directed weaning (178 patients). The endpoint of the study was duration of mechanical ventilation (until ventilatory support was discontinued). Patient entry criteria for the study were PaO2/FIO2 greater than 200 mmHg, PEEP less than 5 cm H2O, heart rate less than 140 beats/min, respiratory rate less than 35 breaths/min, patient alertness, and the prior discontinuation of all vasoactive and inotropic agents. Protocol-directed weaning was provided by well-trained nursing staffs and respiratory therapists. The median duration of protocol-directed weaning was 35 hours, and that of physician-directed weaning was 44 hours. Kaplan-Meier analysis revealed a significantly shorter weaning duration in protocol-directed weaning, and there was no difference in the number of patients who required reintubation between the groups. However, the hospital mortality rate was similar and hospital cost savings for patients managed with protocol-directed weaning was $42,960.
Thorens and associates evaluated the effect of nursing workforce on weaning duration in patients with COPD.8 When a sufficient number and quality of nursing workforce is provided, weaning duration of COPD patients can be significantly reduced in comparison to that with insufficient nursing workforce (10.1 vs 19.0 days; nursing index: 0.65 vs 1.03, respectively).
Table
Comparison of the Brochard and Esteban weaning studies
Brochard Esteban
Patient Characteristics
Patient 109 130 APACHE II 20 SAPS 12
COPD(%) 28 40 CNS disease(%) 17 8 Acute lung injury(%) 52 Ventilator variable not described
Re-intubation rates
SIMV(%) 11.6 13.8 PSV(%) 3.2 18.9 T-piece (%) 2.9 15.2
Mode and Technique of Weaning
The efficiency of ventilatory mode, including SIMV, PSV, and T-piece trials, on weaning duration has been studied comprehensively by two groups, with different results.9,10 Detailed comment on those two studies was provided by Kacmarek in a previous issue.11 In addition to the points raised by Kacmarek, in those two studies the patients’ primary disease processes, their severity of illness, and the ventilators used may have been differentthat is, they were conduced on different patient populations with ventilators of different performance (see Table). Therefore, direct comparison or mata-analysis on both studies cannot meaningfully be accomplished.
Esteban and the Spanish lung failure collaborative group conducted another study comparing the extubation outcome after weaning using either a two-hour T-piece trial or pressure support (PSV) of 7 cmH2O.12 Patients who were on mechanical ventilation for more than 48 hours and ready for weaning were enrolled in the study. They were randomized to receive either T-piece trial (246 patients) or PSV weaning (238 patients). The percentage of patients who remained extubated for 48 hours was not different between the T-piece and PSV groups (63 vs 70%, respectively), but weaning failure occurred more frequently in the T-piece trial (22 vs 14%; P = 0.03). ICU mortality among reintubated patients was significantly higher than in successfully extubated patients (27 vs 2.6%; P < 0.001). Although it can be speculated that patients who required reintubation were sicker, the severity of illness as assessed by SAPS II score in the patients who were successfully extubated, who required reintubation, and who failed the weaning trial were not significantly different. A reason for higher mortality in the patients who required reintubation was not described in their study; it could have been due to the development of additional complications after extubation, such as respiratory muscle failure, cardiac failure, ICU psychosis, upper airway stenosis, or inadequate clearance of respiratory secretions.
Summary
The duration of mechanical ventilation before a weaning trial can be affected by primary disease, severity of illness, and ventilator associated complications. Weaning duration can be affected by use of sedative, nursing workforce, and weaning protocol but may not be affected by the ventilatory mode. Mortality of patients who require ventilatory support is affected by primary disease, severity of illness, ventilator-associated complications, and reintubation.
References
1. Seneff MG, et al. Chest 1996;110:694-679.
2. Kollef MH, et al. JAMA 1993;270:1965-1970.
3. Fagon JY, et al. Am J Med 1993;94:281-288.
4. Sun X, et al. Crit Care Med 1996;24:753-755.
5. Quinio B, et al. Chest 1996;109:765-772.
6. Barrientos-Vega R, et al. Crit Care Med 1997;25:33-40.
7. Kollef MH, et al. Crit Care Med 1997;25:567-574.
8. Thorens JB, et al. Crit Care Med 1995;23:1807-1815.
9. Brochard L, et al. Am J Respir Crit Care Med 1994;150:896-903.
10. Esteban A, et al. N Engl J Med 1995;332:345-350.
11. Kacmarek R. Crit Care Alert 1995;3:10-11.
12. Esteban A, et al. Am J Respir Crit Care Med 1997;156: 459-465.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content